Sage Journals: Discover world-class research

Abstract

Background:

The first weeks of postnatal age is the most hazard time when three-quarters of neonatal death encountered. Africa is the region where most neonatal mortality is recorded and the problem is unresolved yet. Because evidence matter to have significant health improvement, this analysis aimed to generate continental based data on early neonatal death and its predictors.

Methods and Materials:

PICO research question approach was used to search citations from PubMed, Research 4 life, Cochran Library, Google scholar, Epistimonikos, Scopus and Google up to February 2025. Joanna Briggs Institute (JBI) prevalence/cohort/case control studies critical appraisal tools were used to assess the quality of the included articles. All studies conducted in Africa that reported the magnitude of early neonatal mortality, its associated factors and studies that reports both were included. Data was extracted using Microsoft Excel spreadsheet and imported into STATA version 17 for analysis. Publication bias was evaluated through funnel plots and further examined using Egger’s and Begg’s tests. A random effects meta-analysis model with 95% CI was computed to estimate the pooled effect size.

Result:

After a comprehensive search and screening process, 18 articles encompassing 15 892 samples were identified from 2853 citations. The pooled magnitude of early neonatal mortality among neonates admitted to neonatal intensive care units found to be 14% (CI: 0.11-0.17), the highest estimate observed in Morocco 17% (CI: 14-21), followed by in Ethiopia 15% (CI: 10-19) with substantial heterogeneity (I²) level of 96.68% and P value of .00. Because of significant publication bias, the pooled magnitude was imputed, shifting the early neonatal mortality from 14% to 10% (CI: 0.102-0.111). Prematurity, asphyxia and hypothermia were the three most commonly reported factors in the included articles. Consequently, the meta-analysis result showed that prematurity, asphyxiated and hypothermia increases the risk of early neonatal mortality with pooled adjusted odds ratio of 5.4 (CI: 2.83-8.10), 3.35 (CI: 2.07-4.62) and 4.38 (CI: 3.52-5.24), respectively.

Conclusion:

This review reviled that early neonatal mortality in Africa is still a serious issue requiring further intervention, highlighting challenges within the region health care system. Africa needs to intensify efforts to prevent preterm birth and improve preterm care; to prevent the incidence of asphyxia and establish hypothermia free delivery and newborn care services.

Keywords

early neonatal mortality predictors Africa

Background

Neonatal also termed as newborn period is the first four weeks of a child’s life in which many critical events including cardiovascular adaptation immediately at birth, the establishment of feeding patterns, initiation of parent infant bonding and an increased risk of severe infections.¹ The day of birth, the first 7 days and the first 28 days of life are the most critical time for child’s survival.² Children face the highest risk of dying in their first month of life at an average global rate of 17 deaths/1000 live births in 2019, down by 52% from 37 deaths/1000 in 1990.³

Early neonatal death defined as the death of a newborn between 0 and 7 completed days after birth.^4
-7 The first 3 days accounts about 30% of under-5 child deaths, the majority of deaths are due to asphyxia, prematurity and malformation while sepsis covers one half the deaths.⁸ A continuum of quality care during pregnancy, labor, delivery and throughout the neonatal period will be key to reduce neonatal mortality.⁹ Despite the existence of strategies to reduce neonatal mortality, the world lost nearly 2.4 million children in the first month of life in 2019 (6700 neonatal deaths every day, a third of all neonatal deaths within the first day after birth, and three-quarters in the first weeks of life.³ Africa mainly the Sub-Saharan region had the highest neonatal mortality rate in 2019 at 27 deaths/1000 live births, followed by Central and Southern Asia with 24 deaths/1000 live births. A child born in sub-Saharan Africa or Southern Asia is 10 times more likely to die in the first month life than a child born in high income countries.¹⁰

By 2030, all countries are expected to accomplish the target of reducing neonatal death to 12 or fewer/1000 live births, ensuring that no newborn is left behind.¹¹ However, the current neonatal mortality reduction rate has not been sufficient to indicate that the SDG targets will be realized. For example, in Ethiopia (second populous country form Africa), neonatal mortality declined from 39 deaths/1000 live births in 2005 to 29 deaths/1000 live births in 2016 before increasing to 33 deaths/1000 births in 2019.¹²

Africa is a region affected by both manmade and natural disasters that disrupt the health system affecting effort to reduce neonatal mortality. Evidences suggests that the burden of neonatal mortality remains a significant challenge, hindering progress toward the African Health Strategy Plan 2016 to 2030, aimed to reduce morbidity and eliminate preventable deaths from communicable and non-communicable diseases and other health conditions in Africa by 2030 through the achievement of universal health coverage.¹³

In order to achieve both global and continental health plan, comprehensive evidences plays paramount role for maternal and neonatal health. Evidences generated from systematic review and meta-analysis reflected to be more robust and objective compared to individual studies that call attention for policy makers to decide and map strategic planning at the continental level. Countries with comparable health programs and plans need universal evidence to escort harmonized action. Therefore, the purpose of this systematic review and meta-analysis was to generate continent based evidence on the early neonatal death and its associated factors with different variable in Africa.

Research Questions

What is the pooled magnitude of early neonatal mortality in Africa?

What are the factors associated to early neonatal mortality in Africa?

Methods and Materials

This systematic review and meta-analysis was conducted based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).¹⁴ The protocol of this review was registered in International Prospective Register of Systematic Reviews (PROSPERO) with registration number of CRD42023439499.¹⁵

Searching Strategies

The search approach aimed at accessing both published and unpublished citations. First, the searching attempt in PubMed performed, and terminologies was identified for the main searching term development. Then PICO searching framework to develop question was used considering P – Patient or Population, I – intervention/exposure, C – comparator/control and O – outcome neonates as patients or population and death as an outcome. Boolean terms or Boolean operators or logical operators of “AND,” “OR” and “NOT” were used to improve searches and combine keywords in databases or search engines, making searches more precise. According searching term were constructed using the advanced searching engine, for example, from PubMed “(‘Early Neonatal death’ [Title/Abstract]) OR (‘Early Neonatal Mortality’ [Title/Abstract]) OR (‘First week Neonatal Mortality’ [Title/Abstract]) OR (‘"First seven days Neonatal Death’ [Title/Abstract]) AND (‘associated factors’ [Title/Abstract]) OR (‘causes’ [Title/Abstract]) OR (‘risk factors’ [Title/Abstract]) OR (‘predictors’ [Title/Abstract]) AND (‘Africa’ [MeSH Terms]) OR (‘African countries’ [MeSH Terms]) NOT (‘Adolescents’ [MeSH Terms]) NOT (‘Under-Five’ [MeSH Terms]) NOT (‘Animal’ [MeSH Terms]) NOT (‘adult children’ [MeSH Terms]) AND (‘2010/1/1:2025/1/1’ [pdat]) AND (‘english’ [Filter]).” Generally, PubMed, Research 4 life, Cochran Library, Google scholar, Epistimonikos, Scopus and Google were the databases used to search the citations. Searching was tried to include grey literatures from GreyNet International.

Exclusion and Inclusion Criteria

Study Population: research articles conducted among early neonatal mortality were considered.

Study Settings: Studies conducted across different countries of the continent from 2010 to February 2025 were incorporated.

Study Design: observational studies including cross-sectional, case-control and cohort studies with original data reporting at least the magnitude of early neonatal death and its associated factors were included. There is a deviation of inclusion studies from the protocol. Because of the magnitude, deviation to include studies conducted in the community, from live birth that includes demographic health survey studies and articles only conducted from death were excluded. Therefore, the studies included for this review were articles only done from admission neonates.

Language: articles published in English language were included.

Publication Status: both published and unpublished articles were included in this review. Studies which did not report the outcome, published before 2010 were excluded.

Outcome Measures

In this review and meta-analysis, 2 research questions were assessed. The first was to assess the pooled overall magnitude of early neonatal death in Africa and it was calculated by dividing the number of neonates died within their first weeks of life by the total number of neonates admitted during study periods, and multiplied by 100. The second research question was to identify the factors of early neonatal mortality in Africa. The most frequently reported factors of early neonatal mortality were considered in the final analysis to identify predictors. The most 3 frequently identified factors were selected for the final meta-analysis. Odds ratio (OR) will be used to express the pooled effect. Pooled effect can be calculated at least from 2 effect sizes. However, the intension was to address the most dominant factors causing to early neonatal mortality and assuming to ensure the analysis targeted the most consistently reported and impactful predictors, enabling the policy makers to develop and forwards urgent interventions.

Screening and Data Extraction

Preliminary search was performed in PubMed by D.B.M., T.A.E., F.K., M.W., A.T. and A.E. After this, the main searching had been done and first all citations from each databases were exported to EndNote version 21.5 and duplicates were removed. Then, 4 reviewers (D.B.M., F.B.G., M.A.A. and N.M.M.) screened titles and abstracts against the inclusion criteria. Then, the full text of the articles were accessed, and independent assessment was carried out by 4 reviewers (A.Sh.G., T.A.T., YOG. and Y.G.A.) based on the predetermined inclusion and exclusion criteria. Discrepancies between the reviewers were resolved through discussion and common consensus. Data was extracted from the included papers by 4 authors (D.B.M., D.S.S., M.A.A. and F.B.G.) independently from a random sample of 20% of the papers to check consistency; consequently. Then the extracted data was send to each author for final approval before data analysis. The plan was to include all citations that has report of either magnitude or factors or both without considering the context of the study. However, based on the genuine agreements of the reviewers targeting to synthesis valid evidence, only institution-based studies involving admitted neonates were included. The reviewers argued that pooling studies conducted using DHS data, studies on admitted neonates, community-based studies, studies analyzing only causes of death and studies based on total births together would introduce bias.

Assessment of Study Quality

A structured data abstraction form using Microsoft Excel was developed. Study author, publication year and study design, sample size, magnitude of early neonatal death, associated factors and other useful variables were extracted. The Joanna Briggs Institute (JBI) prevalence, case control and cohort study critical appraisal tools were used to assess quality of the study¹⁶ (Supplementary Table 1).

Data Synthesis and Statistical Analysis

Data was extracted using Microsoft Excel spread sheet software and imported into STATA version 17 software for meta-analysis. The pooled effect size with a 95% confidence interval of African early neonatal mortality was determined using an inverse variance random effects model. Heterogeneity across the studies was assessed using the I² statistic, where 25%, 50% and 75% represent low, moderate and high heterogeneities, respectively. A funnel plot and Begg’s and Egger’s tests were used to check publication bias. The final report for the magnitude was imputed because of significant publication bias. Moreover, subgroup analyses based on study area (country) were conducted. Adjusted odds ratios (AOR) for the predictors and percentages for the magnitude were used to report the pooled magnitude and odds of predictors, respectively.

Results

Study Selection

By following the modified PRISMA guideline, a total of 2853 articles were extracted from PubMed (n = 323), Research 4 life (n = 157), Cochran Library (n = 1686), Google scholar (n = 587), Epistimonikos (n = 57), Scopus (n = 32) and others (n = 11). Some databases that includes CINAHL listed in the protocol were not included in this review because of the issue of no search result had been presented. During the retrieve process, first 42 citations were removed because of duplications and then citations were reviewed for publication year, country and animal-human, neonatal. Following this, 2149 citations were removed because of non-neonatal topics, 58 due to scope of review area, 499 because of the lack of early and 11 citations were removed because they were review. Then, 99 articles were reviewed for abstract and 51 of them had been rejected because of the exclusion criteria. Finally, 48 articles were reviewed for full and 21 of them were included in the final report of the systematic review and meta-analysis (Figure 1). The rest 27 articles (Sabiri et al, 2012, Tura et al, 2020, Fekadeselassie Belege Getaneh¹*, Lakew Asmare², Abel Endawkie², Alemu Gedefie³, Amare Muche², Anissa Mohammed², Aznamariam Ayres², Dagnachew Melak², 2024, Ouahid et al, 2019, Carlo et al, 2010, Chomba et al, 2016, Chaibva et al, 2019, Wuni et al, 2023, Krüger et al, 2012, Kamfwa, 2016, Shayo et al, 2022, Tekeba et al, 2024, Camara et al, 2021, Tesfay et al, 2022, Turnbull et al, 2011, Lohela et al, 2012, Okot et al, 2024, Tamir et al, 2024, Bellizzi et al, 2018, Damtew et al, 2024, Elias et al, 2022, Aregahegn Wudneh, Tesfaye Gugsa Alemu, Abbas Ahmed, Mesfin Abebe, 2023, Dahiru, 2017, McKinnon et al, 2014, Ramaiya et al, 2013, Engmann et al, 2012, Dahiru, 2015) were excluded after full document screening because of different criteria that include the study context and study area (Supplementary Table 2).

Figure 1.

PRISMA 2020 flow diagram for systematic review of early neonatal mortality in Africa, 2025.

Characteristics of the Included Studies

Included studies were conducted from 9 countries, among this 62% were from Ethiopia, 9.5% from Uganda and; DRC, Gahana, Burkina Faso, Morocco, Tanzania, Burundi each covers 4.76% of the total studies. This review has involved a total of 17 172 study subjects with the minimum sample size 200¹⁷ and maximum sample of 3789.¹⁸ The highest magnitude of early neonatal mortality was recorded as 32.46%¹⁹ in Ethiopia and the lowest mortality was reported as 3.74%²⁰ in Burundi. Most included studies were follow-up and only 1 case control study was included in this review (Table 1).

Table 1.

Study Characteristics of the Included Articles.

Author	Country	Study design	Sample size	Publication year	Objective of the study	Magnitude	Magnitude LCI	Magnitude UCI
Aimée²¹	DRC	Cross-sectional	1441	2021	To identify causes early and late neonatal mortality in	13.31%	11.33%	15.33%
Abera et al²²	Ethiopia	Cross-sectional	289	2021	Mortality and factors	14.56%	9.66%	19.46%
Ahmed et al²³	Ethiopia	Cross-sectional	765	2023	Determinants of early neonatal mortality	12.94%	10.20%	15.68%
Avoka et al²⁴	Gahana	Retrospective chart review	811	2018	Very early neonatal mortality	11.00%	8.84%	13.16%.
Wondie et al¹⁹	Ethiopia	Retrospective follow-up study	761	2023	Incidence and predictors of Mortality LBW	32.46%	29.07%	35.73%
Gebremariam et al²⁵	Ethiopia	Prospective follow-up study	393	2021	Prediction of early neonatal mortality	15.27%	12.03%	19.18%
Roro et al²⁶	Ethiopia	Prospective cohort	2090	2019	Predictors, causes and trends of neonatal mortality	8.80%	6.82%	9.26%
Worku et al¹⁸	Ethiopia	Cross-sectional	3789	2012	Independent predictors of early neonatal mortality	23.20%	21.80%	24.60%
Kyasimire et al²⁷	Uganda	Prospective cohort	384	2024	Clinical profiles, incidence and predictors of early neonatal mortality	12.00%	8.73%	15.23%
Bushra et al²⁸	Ethiopia	Cross-sectional study	432	2024	Magnitude and factors	10.60%	8.06%	13.94%
Kahsay²⁹	Ethiopia	Prospective cohort study	253	2019	Incidence and predictors (survival)	12.65%	8.57%	16.77%
Musooko et al³⁰	Uganda	Prospective cohort	635	2014	Incidence and risk factors for early neonatal death	12.40%	9.89%	15.03%
Malka et al³¹	Ethiopia	Prospective cohort	391	2024	Incidence and predictors of early neonatal mortality	17.40%	13.57%	21.07%
Osamo Kelbore et al³²	Ethiopia	Prospective cohort	689	2024	Incidence and predictors of early neonatal mortality	8.40%	6.35%	10.51%
Kaboré et al³³	Burkina Faso	Case control	940	2016	Identify the risk factors associated with very early neonatal death	.	.	.
Eshete and Abiy³⁴	Ethiopia	Prospective cohort	913	2020	When do newborn die	4.00%	2.49%	4.95%
Kaoutar et al³⁵	Morocco	Cohort	529	2021	Determinants of early neonatal mortality in	17.40%	14.17%	20.63%
Gessesse et al³⁶	Ethiopia	Prospective follow-up study	387	2024	Time to death	15.20%	11.61%	18.79%
Kibona et al¹⁷	Tanzania	prospective cohort	200	2021	Seven day outcomes and risk factors of low birth weight neonates	5.00%	1.98%	8.02%
Birhanu et al³⁷	Ethiopia	Prospective follow-up Cohort	358	2022	Survival and predictors of preterm mortality	29.32%	24.67%	33.97%
Ndayishimiye et al²⁰	Burundi	Prospective cohort	722	2024	To determine the neonatal survival time and identify factors	3.74%	2.35%	2.35%

Magnitude of Early Neonatal Mortality in Africa

The pooled magnitude of early neonatal mortality from the 20 articles was 14% (CI: 0.11-0.17) with significant variability of I² = 97.62%, P = .00). The individual studies reports that 4 % (CI: 0.02-0.05) and the highest level 32% (CI: 0.29-0.36). The 2 highest pooled magnitude of early neonatal mortality reported on the forest plot 32% (CI: 0.29-0.36) and 29% (CI: 0.25-0.34) might be because the studies were conducted among low birth weight and preterm neonates, respectively. The sub group meta-analysis indicates that Morocco is the first with the 1 study based magnitude of 17% (CI: 0.14-0.21) and Ethiopia follows with 16% (CI: 0.11-0.20). Because of high-level publication bias, trim-and-fill analysis was performed and the pooled early neonatal mortality after imputed additional studies shifted from 14% to 10% (CI: 0.102-0.111; Figures 2 and 3).

Figure 2.

Forest plot of the pooled early neonatal mortality in Africa (n = 21).

Figure 3.

Forest plot of the sub-group pooled early neonatal mortality in Africa (n = 21).

Cumulative meta-analysis was performed to see the trends of early neonatal mortality through publication periods. The trends showed that there is reduction of early neonatal mortality from 23% (CI: 0.22-0.25) to 12% (CI: 0.09-0.16) from 2012 to 2021. Unfortunately, after 2021, the trends showed resurge or slight increasing of ENM which needs further explanation or exploration. However, recently by the year 2024, the early neonatal mortality showed that 14% (CI: 0.11-0.17; Figure 4).

Figure 4.

Cumulative meta-analysis of early neonatal mortality trends over time in Africa (2012-2024).

Sensitivity analysis was conducted to check the outlier and see robustness of the findings, however there is no outlier detected during sensitivity analysis result. In the sensitivity analysis, the lowest and highest 2 values were 12.9% (CI: 0.101-0.157) and 14.4% (CI: 0.112-0.177) with true effects of theta 13.9% (CI: 0.106-0.171) indicating all are within the confidence interval of the overall effect size 14% (CI: 0.11-0.17; Table 2).

Table 2.

Sensitivity Analysis of the Articles Leave-One-Out Meta-Analysis Summary.

Author and year	Magnitude	95% (CI)	P value
Aimée et al (2021)²¹	0.139	0.105-0.174	.000
Abera et al (2021)²²	0.139	0.104-0.173	.000
Ahmed et al (2023)²³	0.139	0.105-0.174	.000
Avoka et al (2018)²⁴	0.140	0.106-0.175	.000
Wondie et al¹⁹	0.129	0.101-0.157	.000
Gebremariam et al (2021)²⁵	0.138	0.104-0.173	.000
Roro et al (2019)²⁶	0.142	0.108-0.176	.000
Worku et al¹⁸	0.134	0.101-0.166	.000
Kyasimire et al (2024)²⁷	0.140	0.106-0.174	.000
Bushra et al (2024)²⁸	0.141	0.106-0.175	.000
Kahsay et al (2019)²⁹	0.140	0.105-0.174	.000
Musooko et al (2014)³⁰	0.140	0.105-0.174	.000
Malka et al (2024)³¹	0.137	0.103-0.171	.000
Osamo et al (2024)³²	0.142	0.108-0.176	.000
Eshete & Abiy (2020)³⁴	0.144	0.112-0.177	.000
Kaoutar et al (2021)³⁵	0.137	0.103-0.171	.000
Gessesse et al (2024)³⁶	0.138	0.104-0.173	.000
Kibona et al¹⁷	0.143	0.110-0.177	.000
Birhanu et al³⁷	0.131	0.101-0.162	.000
Ndayishimiye et al²⁰	0.144	0.112-0.177	.000
Theta	0.139	0.106-0.171	.000

Random-effects model number of studies = 20. Method: REML.

Galbraith plot was computed for the included studies and the point indicates that slightly fragmented distributions suggesting heterogeneity. The Galbraith plot indicates that 2 studies are found out of 95% CI of the plot (Figure 4).

To detect the source of heterogeneity, Meta regression was conducted based on country, publication year, sample size and by combining these all resulting R² (%) = 0.00 which are <0.05.

Publication bias was tested both statistical using regression-based Egger test and graphically using funnel plot. The Egger test result P = .0265 indicates that there is significant publication bias and the statistical finding is supported by the graphical test of funnel plot that indicates points in the funnel plot are not symmetrically distributed and; could be concluded that there is the possibility of publication bias (Figures 5 and 6).

Figure 5.

Galbraith plot for meta-analysis of for the outcome of magnitude of early neonatal mortality in Africa.

Figure 6.

Funnel plot for the magnitude of early neonatal mortality in Africa.

Following the detection of significant publication bias, nonparametric trim-and-fill analysis was conducted and 2 studies were imputed to the left making the overall pooled magnitude of early neonatal mortality to shift from 14% to 10% (CI: 0.102-0.111). The result suggests that although the net or pooled neonatal mortality linger substantial, bookkeeping the probable publication bias marks reduction in pooled magnitude of early neonatal mortality, emphasizing the position of allowing for missing articles in meta-analysis (Figure 7).

Figure 7.

Funnel Plot after imputation to estimate the effect size of early neonatal mortality in Africa.

Factors Associated With Early Neonatal Mortality in Africa

In this systematic review and meta-analysis, the most 3 top reported significant associated factors to early neonatal mortality were selected for meta-analysis. Asphyxia or low APGAR score, preterm or prematurity and hypothermia or low body temperature were the top 3 identified factors to early neonatal mortality reported by 9, 5 and 3 studies, respectively. For each identified factors the pooled odds of mortality was calculated. Neonates born preterm, neonates having hypothermia and neonates with low APGAR score or asphyxia had increased pooled risk of early neonatal death by 5.47 (CI: 2.83-8.10), 4.38 (CI: 3.52-5.24) and 4.28 (CI: 2.14-6.42), respectively, with significant heterogeneity I² of 100%, 99.95% and 100%, respectively (Figures 8 -10).

Figure 8.

Pooled odds ratio of preterm as factors of early neonatal mortality in Africa.

Figure 9.

Pooled odds ratio of hypothermia as a factors for early neonatal mortality in Africa.

Figure 10.

Pooled odds ratio of asphyxia as a factor of early neonatal mortality in Africa.

Sensitivity analysis was conducted for all 3 top determinants of early neonatal mortality using leave-one-out meta-analysis reviling there is no outlier preterm theta 5.466 (CI: 2.83-8.100), asphyxia 4.28 (CI: 2.14-6.42) and hypothermia 4.38 (CI: 3.52-5.24) with P value of .00. All 3 factors sensitivity analysis result indicates that there are no studies for all preterm, asphyxia and hypothermia with effect size of out of each over all effects size confidences interval. Random-effects meta-regression result indicates that the source of heterogeneity as a factor preterm has R² (%) = 69.96 in publication year. This implies that the potential source of heterogeneity is in publication year. However the possible source of heterogeneity for asphyxia and hypothermia were not identified by meta- regression of the variables in relation with sample size and publication year of R² (%) result of = 0.00.

Publication bias was also conducted for all 3 factors individually and the result reviled that regression-based Egger test of hypothermia P value .0036, preterm .0662 and asphyxia .3212. These egger test results reviewed that there is publication bias in the articles which reports hypothermia is factors associated with early neonatal mortality with P value of .0036 whereas publications that reports preterm and asphyxia as determinants of early neonatal mortality showed that there is publication bias. Consequently, nonparametric trim-and-fill analysis of publication bias was conducted and studies were imputed on the right sides making to shifts the true effects size from OR: 4.38 (CI: 3.52-524) to 4.24 (4.22-4.26; Supplementary Figures 1 and 2).

Discussion

This systematic review and meta-analysis was conducted to determine the pooled magnitude early neonatal mortality and its top predictors in Africa. The meta-analysis of 21 studies showed that the pooled magnitude of early neonatal mortality among admitted neonates before and after article imputation were 14% (CI: 0.11-0.17) and 10% (CI: 0.102-0.111). This pooled effect estimate is higher compared to similar systematic review and meta-analysis study conducted to estimate the burden of early neonatal mortality in Sub-Saharan, that reports 80.3 (95% CI 66-94.6)/1000 live births.³⁸ The possible discrepancy between these 2 findings could be justified by the context of the studies included for the meta- analysis. This study estimates the magnitude of early neonatal mortality only from admitted neonates whereas the study conducted by Moges et al³⁹ included all studies that reports early neonatal mortality (hospital vs community). Admitted neonates have different causes of admission which may increase the potential of death. In contrast, studies that comprise all live births estimate deaths from the total birth, which may effect in a lower reported hazard of death compared to studies directing only on hospitalized neonates. This magnitude is also higher compared to the total neonatal death reported in United States of America.⁴⁰ This indicates that although Africa is working to improve neonatal survival, the burden of early neonatal death is still devastated problem. This might be the fact that the health care system of Africa is facing numerous challenges that include limited NICU technology adoption, inadequate budgetary allocation to health and poor leadership and management in the health care system of the region.⁴¹ Poverty, limited access to pregnancy follow-up and quality of service, different forms of conflicts that compromised services delivery and program unsustainability could also have their own effects.^42,43

Prematurity was one of the three top reported factors associated with early neonatal mortality in Africa with pooled adjusted odds ratio of 5.4 (CI: 2.83-8.10). This means preterm neonates have 5.4 times higher risk dying within the first weeks of life compared to term newborns. This finding is consistent with evidences reported by one research conducted in India.⁴⁴ This indicates that, the burdens of preterm related complications are still the leading cause of neonatal mortality unresolved yet, despite different initiatives are being implemented so far. The possible justification could be explained by the nature of prematurity that leads to high admission rate, feeding intolerance, less immunity to cope up with sepsis, immature organ including lung for surfactants and others. These alone or in combined forms makes the neonates to require sophisticated treatment and care which are difficult to access in Africa, thereby increased the risk of death.⁴⁵ Africa as region has shortcoming to provide possible preterm health care technologies that includes mechanical ventilator, extracorporeal membrane oxygenation, parenteral nutrition, surfactant administration in addition to the gaps to ensure the full coverage of cost effective intervention that includes kangaroo mother care and breast feeding.⁴⁶ Preterm neonates especially in developing region like Africa where promising working process towards neonatal mortality reduction is witnessed, the problem is still a major mortality cause’s identified in this systematic review and meta-analysis. This may also be explained by the poor infection prevention practice in African NICUs, where hospital-acquired infections reaches up to 13.2% considerably intensifying the risk of problems in preterm neonates with immature immunity and leads to increase the odds of death within the first 7 days of life.⁴⁷ Although, preterm is still a major and significant factors of early neonatal mortality, cumulative meta-analysis trend showed the reduction of odds ratio from 10.46 in the year 2012 to 5.47 in 2024. This indicates the positive effects of different interventions taken to improve preterm survival outcomes.⁴⁸

In this systematic review and meta-analysis, neonates diagnosed with asphyxia or having low APGAR score had pooled odds ratio of 4.28 (CI: 2.14-6.42) to early neonatal mortality. This finding is consistent with evidence from previous studies.^6,49 This could be justified by the scientific nature of asphyxia can causes the profound systemic and neurologic sequel resulting from decreased blood flow and oxygen to neonates causing for death.⁵⁰ The problem of asphyxia in developing region including Africa is more widespread because of the inadequate access to quality health services, limited infrastructure and sophisticated pregnancy complications.^50,51 Additionally, the risk of early neonatal death among neonates diagnosed with birth asphyxia could be because asphyxia leads to different complications including hypoglycemia, sepsis and intra-ventricular hemorrhage and necrotizing enterocolitis which may increase the risk of death within the first weeks of life.⁵⁰

Hypothermia was another predictor responsible for early neonatal death reported by 3 research articles. The pooled estimate found that neonates with hypothermia have 4.24 times higher odds of early neonatal death compared to neonates without hypothermia. This finding has indirect similarity with the evidence generated to see the global burden of neonatal hypothermia as major challenge for newborn survival.⁵² The possible justification for the finding why neonates with hypothermia had increased risk of early neonatal mortality might be because; it can lead to delay in surfactant production (causes cardio-respiratory instability). It also causes for intra-ventricular hemorrhage, hypoglycemia, respiratory distress syndrome, apnea and poor feeding increase the odds of early neonatal morbidity and mortality.⁵³ Hypothermia can also alter the immunity of neonates by changing cytokine production and immune cell responses, foremost to suppressed immune system making the neonates unable to fight against causative agents of neonatal infections, follow-on more complications and these increase first week risk of death.⁵²

Limitations of the Study

In this systematic review and meta-analysis, significant level of heterogeneity was detected in the estimation of both magnitude and factors, which possess challenging to draw overall definitive conclusions. Publication bias was also another issue observed in indicating that studies with significant findings expected to be available or published that could increase or decrease the overall findings. Additionally, this finding incorporates only articles reported by English language, potentially miss pertinent outcomes causing to minimize exhaustiveness the study.

Conclusion

This systematic review and meta-analysis disclosed substantial magnitude and top three predictors of early neonatal mortality in Africa. Prematurity, asphyxia and hypothermia were the identified factors responsible for early neonatal mortality in the region. Ethiopia is the country with second record of early neonatal mortality and the only country where preterm birth has been reported as significant predictors associated with early neonatal mortality.

It is important to acknowledge the presence of both heterogeneity and publication bias to draw over all conclusion. In the face of considering these limitations, evidences produced from this systematic review highlight the exigent requirement of efforts to enhance neonatal outcomes in Africa. Policy makers and region-based organizations including African union shall prioritize and strengthen to combat the identified factors that increase the risk of early neonatal mortality. Health care providers in Africa shall strengthen the prevention, early intervention and detection of preterm complications, asphyxia and hypothermia. Future systematic review and meta-analysis shall be conducted explicitly to evaluate the effect of each distinct predictor known on early neonatal mortality in Africa.

Generally, the finding of this systematic review and meta-analysis showed that African health care journey to improve neonatal survival needs more efforts. In the region, prematurity, asphyxia and hypothermia are still the responsible factors for early neonatal death that may be mirror image of level of African maternal and neonatal health care services.

Footnotes

Abbreviations and Acronyms

APGAR Appearance, Pulse, Grimace, Activity and Respiration

JBI Joanna Briggs Institute

PICO Population, Intervention, Comparison and Outcome

PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO International Prospective Register of Systematic Reviews

ORCID iD

Dires Birhanu Mihretie

Authors’ Contributions

D.B.M., T.A.E., F.K., M.W., A.T. and A.E:- Developing the idea, developing and commenting the protocol, searching and screening, appraisal, data extraction, developing the whole writing and reviewing process and conducting the whole meta-analysis in different level of engagement.

D.B.M., F.B.G., M.A.A. and N.M.M.:- Screened titles and abstracts against the inclusion criteria. Then, the full text of the articles were accessed, and independent assessment was carried out by 4 reviewers (A.Sh.G., T.A.T., YOG.. and Y.G.A.) based on the predetermined inclusion and exclusion criteria. Discrepancies between the reviewers were resolved through discussion and common consensus. Data was extracted from the included papers by 4 authors (D.B.M., D.S.S., M.A.A. and F.B.G.) independently from a random sample of 20% of the papers to check consistency.

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.

Data Availability Statement

The data associated with this systematic review and Meta -analysis is on the hands of the principal investigators. Interested individuals or groups on the data can contact to have the data.

Supplemental Material

Supplemental material for this article is available online.

References

Doherty

Salik

Physiology, Neonatal. StatPearls (Internet). 2025. https://www.ncbi.nlm.nih.gov/books/NBK539840/

WHO, UNICEF. Reaching the Every Newborn National 2020 Milestones Country Progress, Plans and Moving Forward. WHO, UNICEF; 2017:2015-2018.

UNICEF. Neonatal Mortality. UNICEF for Every Child; 2020.

Gimeno

Parra-llorca

Salles

, et al Early neonatal death: a challenge worldwide. Semin Fetal Neonatal Med. 2017;22(3):2-6.

Abbing-Karahagopian

Bhat

Pathirana

, et al Neonatal death: case definition & guidelines for data collection, analysis, and presentation of immunization safety data. Vaccine. 2016;34:6027-6037.

Mehkarkar

Sonawane

VB.

A study of early neonatal mortality in a tertiary hospital of Maharashtra, India. Int J Contemp Pediatr. 2018;5(5):1869. doi:10.18203/2349-3291.ijcp20183522

WHO South-East Asia Region. Working Definitions. WHO South East Asia Region Neonatal-Perinatal Database; 2024.

Sankar

Natarajan

Das

Agarwal

Chandrasekaran

Paul

VK.

When do newborns die? A systematic review of timing of overall and cause-specific neonatal deaths in developing countries. J Perinatol. 2016;36(S1):S1-S11.

Cooper

Strategies to reduce perinatal mortality. Lancet Glob Health. 2015;4(1):e6-e7.

10.

World Health Organization. Newborns: Improving Survival and Well-Being. WHO; 2020.

11.

World Health Organization UN. Ending Preventable Newborn Deaths and Stillbirths by 2030. WHO; 2020:3.

12.

Institute EPH, Health FM of, ICF TDP. Ethiopia Mini Demographic and Health Survey 2019. 2021:48.

13.

Affairs AUCD of S. Africa Health Strategy 2016-2030. Sustain. 2019;11(1):1-14. doi:10.1016/j.regsciurbeco.2008.06.005%0A. http://scioteca.caf.com/bitstream/handle/123456789/1091/RED2017-Eng-8ene.pdf?sequence=12&isAllowed=y%0A. https://www.researchgate.net/publication/305320484_SISTEM_PEMBETUNGAN_TERPUSAT_STRATEGI_MELESTARI

14.

Moher

Liberati

Tetzlaff

Altman

DG.

Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol. 2009;62(10):1006-1012.

15.

Birhanu

Ayehu

Shimekaw

, et al Magnitude and determinants of maternal mortality in Eritrea: a systematic review and meta-analysis. BMC Pregnancy Childbirth. 2022;20(1):612.

16.

Munn

Moola

Riitano

Lisy

The development of a critical appraisal tool for use in systematic reviews addressing questions of prevalence. Int J Health Policy Manag. 2014;3(3):123-128.

17.

Kibona

Sungwa

Dika

, et al Immediate seven day outcomes and risk factors of low birth weight neonates at referral hospitals in Mwanza City, Tanzania in October 2020. Afr J Health Sci. 2021;34(4):451-463.

18.

Worku

Kassie

Mekasha

Tilahun

Worku

Predictors of early neonatal mortality at a neonatal intensive care unit of a specialized referral teaching hospital in Ethiopia. Ethiop J Health Dev. 2012;26(3):200-207.

19.

Wondie

Zeleke

Wubneh

CA.

Incidence and predictors of mortality among low birth weight neonates in the first week of life admitted to the neonatal intensive care unit in Northwestern Ethiopia comprehensive specialized hospitals, 2022. Multi-center institution-based retrospective follow-up study. BMC Pediatr. 2023;23(1):1-13.

20.

Ndayishimiye

Iradukunda

Masimbi

Ahishakiye

Odjidja

EN.

Survival of newborns and determinants of their mortality in Burundi: a prospective cohort study at Kamenge Teaching Hospital. Res Sq. 2024;12:42-52.

21.

Aimée1

. Early and late neonatal mortality in hospital settings. Causes and factors related to death (case of Sendwe hospital / DRC). Int J Med Sci Clin Res Rev. 2021;2:22-33.

22.

Abera

Bayisa

Bekele

Dessalegn

Mulisa

Gamtessa

. Neonatal mortality and its associated factors among neonates admitted to wollega university referral hospital neonatal intensive care unit, east wollega, ethiopia. Glob Pediatr Heal. 2021;8:2333794X211030157.

23.

Ahmed

Farah

Ali

Ibrahim

. Determinants of early neonatal mortality (hospital based retrospective cohort study in Somali region of Ethiopia). Sci Rep. 2023;13(1):2003-2022.

24.

Avoka

Adanu

Wombeogo

Seidu

Dun-Dery

. Maternal and neonatal characteristics that influence very early neonatal mortality in the Eastern Regional Hospital of Ghana, Koforidua: a retrospective review. BMC Res Notes. 2018;11(1):1-5.

25.

Gebremariam

Tiruneh

Engidaw

et al. Development and validation of a clinical prognostic risk score to predict early neonatal mortality, Ethiopia: a receiver operating characteristic curve analysis. Clin Epidemiol. 2021;13:637-47.

26.

Roro

Tumtu

Gebre

. Predictors, causes, and trends of neonatal mortality at Nekemte Referral Hospital, east Wollega Zone, western Ethiopia (2010–2014). Retrospective cohort study. PLoS One. 2019;14(10):e0221513.

27.

Kyasimire

Tibaijuka

Ochora

et al. Clinical profiles, incidence and predictors of early neonatal mortality at Mbarara Regional Referral Hospital, south-western Uganda. BMC Pediat. 2024;24(1):542.

28.

Bushra

Kassa

Ayana

et al. Epidemiology of early neonatal mortality in an eastern ethiopian nicu: insights from hiwot fana specialized university hospital. Glob Pediatr Heal. 2024;11:2333794X241273134.

29.

Kahsay

. Survival and predictors of early neonatal death in neonatal intensive care unit of Mekelle general and Ayder comprehensive specialized hospitals, Northern Ethiopia, 2018: prospective cohort study. Res Sq. 2019;1-27.

30.

Musooko

Kakaire

Nakimuli

et al. Incidence and risk factors for early neonatal mortality in newborns with severe perinatal morbidity in Uganda. Int J Gynaecol Obstet. 2014;127(2):201-205.

31.

Malka

Solomon

Kassa

Erega

Tufa

. Time to death and predictors of mortality among early neonates admitted to neonatal intensive care unit of Addis Ababa public Hospitals, Ethiopia: institutional-based prospective cohort study. PLoS One. 2024;19(6.0):e0302665.

32.

Osamo

Ermias

Tadesse

Bubamo

Abame

. Early neonatal mortality and predictors in newborns admitted to the neonatal intensive care unit at public hospitals in hadiya zone, central ethiopia: a retrospective cohort study. Glob Pediatr Health. 2024;11:2333794X241275264.

33.

Kaboré

Bertrand

Koulidiati

louis

Millogo

Kouanda

. Factors associated with very early neonatal mortality in Burkina Faso: a matched case–control study. Int J Gynecol Obstet 2016;135(suppl 1):S93-S97.

34.

Eshete

Abiy

. When do newborns die? Timing and cause-specific neonatal death in neonatal intensive care unit at referral hospital in gedeo zone: a prospective cohort study. Int J Pediatr. Published online February 15, 2020. doi: 10.1155/2020/8707652

35.

Kaoutar

El Boussaadni

Oulmaati

. Early neonatal mortality’s determinants in the neonatology department of Mohamed V Hospital in Tangier, Morocco. Integr J Med Sci. 2021;8:1-8.

36.

Gessesse

Belete

Tadesse

. Time, cause of early neonatal death, and its predictors among neonates admitted to neonatal intensive care units at Bahir Dar City public hospitals, northwest Ethiopia: a prospective follow-up study. Front Pediatr 2024;12:1335858.

37.

Birhanu

Gebremichael

Tesfaye

et al. Survival status and predictors of mortality among preterm neonates admitted to neonatal intensive care unit of Addis Ababa public hospitals, Ethiopia, 2021. A prospective cohort study. BMC Pediatr. 2022;22(1):1-12.

38.

World Health Organization. Newborn Mortal. World Health Organization; 2024:1-23.

39.

Moges

Dessie

Anley

et al. Burden of early neonatal mortality in Sub-Saharan Africa. A systematic review and meta-analysis. PLoS One. 2024;19(7):e0306297.

40.

Akinyemi

Fasokun

Weldeslase

Makanjuola

Omokhodion

OV.

Determinants of neonatal mortality in the United States. Cureus. 2023;15(8):14-23.

41.

Oleribe

Momoh

Uzochukwu

BSC

, et al Identifying key challenges facing healthcare systems in Africa and potential solutions. Int J Gen Med. 2019;12:395-403.

42.

World Health Organisation. Every Newborn Action Plan. World Health Organisation; 2014:58.

43.

United Nations Inter-Agency Group for Child Mortality Estimation (UN IGME). Levels and trends in child mortality estimation. PLoS Med. 2025;109:620-621.

44.

Kannaujiya

Kumar

Upadhyay

, et al Effect of preterm birth on early neonatal, late neonatal, and postneonatal mortality in India. PLOS Glob Public Health. 2022;2(6):1-17.

45.

Zivaljevic

Jovandaric

Babic

Raus

Complications of preterm birth – the importance of care for the outcome: a narrative review. Medicine. 2024;60(6):1-10.

46.

Frantzis

Levasseur

Huebner

, et al Infection prevention and control and related practices in African neonatal units: the Pan-African neonatal care assessment study (PANCAS). Int J Hyg Environ Health. 2024;259:114357.

47.

Gashaw

Ali

Berhane

, et al Neonatal sepsis due to multidrug-resistant bacteria at a tertiary teaching hospital in Ethiopia. Pediatr Infect Dis J. 2024;43(7):687-693.

48.

Moxon

Lawn

Dickson

, et al Inpatient care of small and sick newborns: a multi-country analysis of health system bottlenecks and potential solutions. BMC Pregnancy Childbirth. 2015;15(suppl 2):S7.

49.

Usman

Imam

Farouk

Dayyabu

AL.

Newborn mortality in sub-Saharan Africa: why is perinatal asphyxia still a major cause?

Ann Glob Health. 2019;85(1):1-6.

50.

Gillam-Krakauer

Shah

Gown

Jr.

Birth asphyxia. In: StatPearls (Internet). StatPearls Publishing; 2025.

51.

Hug

Alexander

You

Alkema

National, regional, and global levels and trends in neonatal mortality between 1990 and 2017, with scenario-based projections to 2030: a systematic analysis. Lancet Glob Health. 2019;7(6):e710-e720.

52.

Lunze

Bloom

Jamison

Hamer

DH.

The global burden of neonatal hypothermia: systematic review of a major challenge for newborn survival. BMC Med. 2013;11(1):24.

53.

Pellegrino

Kanyangarara

Agbinko-Djobalar

, et al Occurrence of neonatal hypothermia and associated risk factors among low birth weight (LBW) infants in Accra, Ghana. J Glob Health Rep. 2022;6(1):e55766.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.63 MB

Magnitude and Predictors of Early Neonatal Mortality in Africa:- A Systematic Review and Meta-Analysis