Impact of respiratory distress syndrome and birth asphyxia exposure on the survival of preterm neonates in East Africa continent: systematic review and meta-analysis

Abstract

Introduction

Several kinds of researches are available on preterm mortality in the East Africa continent; however, it is inconsistent and inconclusive, which requires the pooled evidence to recognize the burden in general.

Purpose

To collect and synthesis evidence on preterm mortality and identify factors in the East Africa continent.

Methods

PubMed, Google Scholar, Hinary, Cochrane library, research gate, and institutional repositories were retrieved to identity eligible articles through Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. The articles were selected if the publication period is between 2010-2021 G.C. Data were extracted by a standardized JBI data extraction format for mortality rate and stratified the associated factors. Then exported to STATA 14 for further analysis. I² and Egger's tests were employed to estimate the heterogeneity and publication bias respectively. Subgroup analysis based on country, study design, year of publication, and the sample size was also examined.

Result

This meta-analysis included 32 articles with a total of 21,405 study participants. The pooled mortality rate among preterm in the East Africa continent was found to be 19.2% (95% CI (confidence interval (16.0–22.4)). Regarding the study design, the mortality rate was found to be 18.1%, 19.4%, and 19.7% concerning the prospective cohort, retrospective cohort, and cross-sectional studies. The pooled odds of mortality among preterm with respiratory distress syndrome decreased survival by nearly three folds [AOR (Adjusted odds ratio = 3.2; 95% CI: 22, 4.6)] as compared to their counterparts. Similarly, preterm neonates presented with birth asphyxia were nearly three times higher in death as compared with preterm without birth asphyxia [AOR = 2.6; 95% CI: 1.9, 3.4].

Conclusion

Preterm mortality was found to be unacceptably high in Eastern Africa continent.

Fortunately, the main causes of death were found to be respiratory distress syndrome and birth asphyxia which are preventable and treatable hence early detection and timely management of this problem are highly recommended to improve preterm survival.

Preterm, Respiratory distress syndrome, Asphyxia, Impact, Systematic Review, Meta-analysis, East Africa.

1. Introduction

Preterm refers to a baby born before 37 weeks of pregnancy has been completed [1]. Preterm birth can be further sub-divided based on gestational age: extremely preterm (<28 weeks), very preterm (28 - <32 weeks) and moderate preterm (32 - <37 completed weeks of gestation) [2].

The Neonatal period is the most vulnerable time for a child's survival [3]. In 2016, 2.6 million deaths, or roughly 46% of all under-five deaths, occurred during this period [4]. Neonatal mortality (NM) is a major public health challenge worldwide [5, 6]. In 2019, approximately 17 deaths per 1,000 live births had been reported worldwide [3]. Of these, approximately 70% of the neonatal mortalities in resource-limited setting predominantly in East Africa [5, 6, 7].

Worldwide, neonatal mortality due to preterm accounts for 15 to 36, percent [8]. However, in low to middle-income countries neonatal mortality contributed by preterm ranges from 34–40%, and preterm is the second leading cause of under-five mortality [9, 10].

Globally, 13 million preterm were born annually and the highest percentage resource-limited setting [11], and eastern Africa countries have a lion sharing of this burden [10, 12, 13, 14]. Besides, a global action report 2020 showed that the preterm birth rate was 11%, and millions of children died due to preterm birth before the age of 5 years. Meanwhile, preterm birth is the leading cause of death among children [15, 16]. In 2019, 47% of all under-5 deaths occurred in the newborn period, and close to 75 % dying within the first week of life due to preterm birth [17, 17]. Even though premature babies can be saved with feasible and cost-effective care [9, 18], still it is a leading cause of infant mortality in developing countries [9, 10, 19].

In Sub-Saharan Africa had the highest neonatal mortality rate in 2019 at 27 deaths per 1,000 live births, followed by Central and Southern Asia with 24 deaths per 1,000 live births [18]. Besides, in Sub-Saharan Africa, East Africa countries accounts the highest number of neonatal mortality, 2019 [19,20].

Globally, the burden of preterm birth is disproportionately concentrated in East Africa and Asia, which account for 85% of all preterm births [21]. Preterm birth is a significant challenge in developing countries due to the rapid increase in their incidence and their disproportionate contribution to increased neonatal mortality rates [22].

According to Ethiopia Demographic Health Survey (EDHS) 2019, neonatal mortality was increased to 30 death per 1000 live births from 29 per 1000 in 2016 and preterm is the second leading factor [23], which is 10 folds of the mortality rate in the developed nations [3].

Different research conducted on preterm mortality in east Africa with a huge discrepancy range from 4.4 %- to 41% in Zambia and Sudan [24, 25] respectively. Besides, preterm mortality has been interlinked with different contributing factors including birth asphyxia, feeding difficulties, hypothermia, hypoglycemia, respiratory syndrome disease, jaundice, and necrotizing enter colitis [26, 27, 28]. Thus, this inconsistency, inconclusive, and uncertain requires pooled evidence to recognize the burden of Eastern African countries. In addition to this, the findings from this review will be utilized to guide the development of guidelines and to enhance the efforts of stakeholders towards the improvement of preterm survival. Therefore, the present study aimed to collect and synthesis evidence on preterm mortality and identify factors in the East Africa continent.

2. Methods

2.1. Design and search strategy

2.1.1. Review question

The review questions of this systematic review and meta-analysis were what is the pooled mortality rate among preterm neonates in the East Africa continent? In addition, what are the factors associated with preterm mortality in the East Africa continent?

2.1.2. Reporting

The review protocol has been sent to the PROSPERO database for registration Standard Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) checklist was used to present the results of the review (Additional file).

2.2. Inclusion and exclusion criteria

Both published and unpublished cross-section, case-control, and cohort studies that reported the mortality rate or survival rate and factors associated with it among neonates in East African country context were included f. East African countries include (Sudan, South Sudan, Kenya, Uganda, Djibouti, Eritrea, Ethiopia, Somalia, Tanzania, Rwanda, Burundi, Comoros, Mauritius, Seychelles, Mozambique, Madagascar, Zambia, Malawi, Zimbabwe, Reunion, and Mayotte). The articles were selected if their publication period is between 2010-2021 G.C. However, studies with no abstracts, case series, case reports, and qualitative studies were excluded from the study.

2.3. Search strategies

This review identified studies that provide information on the mortality rate or survival rate in the continent of Eastern Africa. In the searching engine, PubMed, Google Scholar, Hinary, Cochrane library, research gate, and institutional repositories were retrieved. The search linked keywords were combined with Boolean operators “AND and OR^`` in the context of population, exposures/intervention, comparison, and condition/outcome (PICO) format. Besides, the related articles for the references of the relevant articles were also searched and investigated. Those search terms or phrases included were: “preterm”, “term”, “neonate”, “newborn”, “small gestation age”, “low birth weight”, “mortality”, “survival”, and Eastern Africa. Using those key terms, the following search map was applied: (prevalence OR magnitude OR epidemiology) AND (predictor OR risk factor OR determinate OR associated factors) AND (preterm [MeSH Term] OR term OR neonate OR newborn OR small gestational age OR low birth weight) AND (mortality [MeSH Terms] OR survival) AND (Eastern Africa) OR developing country on PubMed database. These search terms were further paired with the names of each East African country including Sudan, South Sudan, Kenya, Uganda, Djibouti, Eritrea, Ethiopia, Somalia, Tanzania, Rwanda, Burundi, Comoros, Mauritius, Seychelles, Mozambique, Madagascar, Zambia, Malawi, Zimbabwe, Reunion, and Mayotte.

2.4. Study selection and screening

All titles and abstracts were screened exhaustively by independent authors (ESC and BMB.) to identify potentially relevant articles. Whenever further information is needed, we made some efforts to communicate via authors by email. The retrieved studies were exported to the citation manager (Zotero) and then duplicate articles were excluded. Disagreements were discussed during a consensus meeting with other reviewers (FAG, HSH, and TM) for the final selection of studies to be included in the systematic review and meta-analysis.

2.5. Data extraction and quality assessment

Two authors (WNA and DKM) independently and then in collaboration extracted all relevant information by using a standardized JBI data extraction format. For each included article, the following data were extracted: authors’ name, year of publications, study region, sample size, study design, and study setting. Besides, factors associated with mortality were recorded in a standardized JBI data abstraction format. Any disagreements between authors were resolved through discussions with the third and fourth authors when required. The retrieved data was crosschecked with the included papers, then modifications and editions of mistyping data were made when required.

2.6. Statistical analysis

The authors were edited, cleaned, and checked for completeness of the extracted data in an excel sheet, then exported into STATA 14 for further analysis. Pooled overall mortality rate and classified factors associated with mortality were estimated by a random effect meta-analysis model. Heterogeneity between the studies was assessed using the I² statistic and the value of I², they show the overall variation across the studies presented by low, moderate, and high with the percentage of 25, 50, and 75%, respectively [29]. Publication bias was estimated through a funnel plot and the Egger test [30]. Moreover, subgroup analysis was done by the study country, study design, sample size, and year of publication. Sensitivity analysis was piloted to examine the effect of a single study on the overall estimation.

3. Results

3.1. Search result

Our search yielded a total of 3699 records, 2038 from PubMed, 71 from Cochrane Library, 41 from Hinariy, 1547 from Google Scholar, 11 from CINAHL, and 06 from Scopus sources. After removal of duplicates, we screened the titles and abstracts of 922 records articles. Finally, a full-text review was conducted for 32 studies with a sample size of 21,405 participants were included to assess the overall mortality rate among preterm births from 11 East African countries including Ethiopia, Tanzania, Kenya, Burundi, Eritrea, Uganda, Tanzania, Malawi, Uganda, Zambia, Sudan, Mozambique, and Zimbabwe (Figure 1).

Figure 1.

The PRISMA flow chart that shows the searching process in East Africa from January 2012–December 2020.

3.2. Characteristics of the included articles

This meta-analysis included 32 different studies covering a total of 21,405 preterms. The studies were conducted from 2012 to 2020 among more than 11 Eastern Africa countries including 12 from Ethiopia represented by 14 studies [31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44]. Tanzania [14, 45, 46, 47] and Uganda [12, 48, 49, 50] represented by 04 studies, Besides, Burundi [51, 52] and Eretria [53, 54] represented by 02 studies. While Malawi [55], Kenya [56], Zambia [24], Sudan [25], Mozambique [57], and Zimbabwe [58] are represented by 01 studies. Regarding study design, 13 studies employed retrospective cohort [12, 14, 37, 38, 39, 40, 41, 42, 43, 44, 45, 56, 57], whereas 07 prospective cohort studies [31, 32, 46, 47, 48, 49, 50], 01 case-control type [36], and 11 cross-sectional design [24, 25, 33, 34, 35, 51, 52, 53, 54, 55, 58]. All studies included in this review were observational studies which were conducted in the hospital with sample sizes ranging from 100 participants reported from a study in Ethiopia and Sudan [25, 36], and 29811 from Uganda [12] respectively (Table 1).

Table 1.

Distribution of mortality among preterm in East Africa from January 2012–December 2020.

First Author/year	Country	Study design	Study type	Sample size	Magnitude (%)	Quality status
Yehuala et al. (2015)	Ethiopia	Retrospective Cohort	Hospital-based	485	25	Low risk
Tamene et al. (2020)	Ethiopia	cross-sectional study	Hospital-based	686	36	Low risk
Yismaw et al. (2019)	Ethiopia	Retrospective Cohort	Hospital-based	516	29	Low risk
Abebe et al. (2019)	Ethiopia	cross-sectional study	Hospital-based	415	26	Low risk
Mengist et al. (2020)	Ethiopia	Prospective Cohort	Hospital-based	774	19	Low risk
Chengo et al. (2020)	Tanzania	Prospective Cohort	Hospital-based	311	15	Low risk
Muchem et al. (2018)	Kenya	Retrospective Cohort	Hospital-based	2080	6	Low risk
Mmbaga et al. (2016)	Tanzania	Retrospective Cohort	Hospital-based	1178	18	Low risk
Seid et al. (2019)	Ethiopia	cross-sectional study	Hospital-based	1488	9	Low risk
Zuniga et al. (2013)	Burundi	cross-sectional	Hospital-based	153	20	Low risk
Shah et al. (2012)	Eritrea	cross-sectional	Hospital-based	1502	6	Low risk
Ndelem et al. (2016)	Burundi	cross-sectional	Hospital-based	437	28	Low risk
Andegio et al. (2020)	Eritrea	cross-sectional	Hospital-based	242	17	Low risk
Namazzi et al. (2020)	Uganda	Prospective Cohort	Hospital-based	242	21	Low risk
Moshiro et al. (2019)	Tanzania	Prospective Cohort	Hospital-based	241	21	Low risk
Orsido et al. (2019)	Ethiopia	Retrospective Cohort	Hospital-based	212	40	Low risk
Roro et al. (2019)	Ethiopia	Retrospective Cohort	Hospital-based	206	14	Low risk
van den et al. (2015)	Malawi	cross-sectional study	Hospital-based	449	22	Low risk
Egesa et al. (2020)	Uganda	Prospective Cohort	Hospital-based	311	36	Low risk
Dessu et al. (2018)	Ethiopia	Retrospective Cohort	Hospital-based	107	35	Low risk
Opio et al. (2019)	Uganda	Prospective Cohort	Hospital-based	128	8	Low risk
Sania et al. (2013)	Tanzania	Retrospective Cohort	Hospital-based	1032	3	Low risk
Dessu et al. (2020)	Ethiopia	Prospective Cohort	Hospital-based	216	8	Low risk
Farah et al. (2018)	Ethiopia	Retrospective Cohort	Hospital-based	432	7	Low risk
Paul et al. (2020)	Uganda	Retrospective Cohort	Hospital-based	2981	5	Low risk
Gudayu (2012)	Ethiopia	Retrospective Cohort	Hospital-based	1733	33	Low risk
Endalam et al. (2020)	Ethiopia	Retrospective Cohort	Hospital-based	535	31	Low risk
Miyoshi et al. (2019)	Zambia	cross-sectional study	Hospital-based	1704	4	Low risk
Kolobo et al. (2020)	Ethiopia	case-control	Hospital-based	100	-	Low risk
Salih et al. (2013)	Sudan	cross-sectional study	Hospital-based	100	41	Low risk
Garcia- et al. (2017)	Mozambique	Retrospective Cohort	Hospital-based	147	12	Low risk
Nyakan et al. (2019)	Zimbabwe	cross-sectional study	Hospital-based	262	14	Low risk

3.3. Meta-analysis

The pooled magnitude of mortality: Most of the studies (n = 31) have reported a mortality rate among preterm neonates. The mortality rate was ranged from highest reported from a study in Sudan 41% [25], and the least was from a study in Zambia 4.4% [24]. The pooled mortality rate among preterm neonates in East Africa using random-effects model analysis was found to be 19.2% (95%CI; 16.0–22.4); I² = 98.5%; p < 0.001) (Table 2 &Figure 2).

Table 2.

The pooled mortality rate of preterm from 32 studies in East Africa from January 2012–December 2020.

Study	ES [95% Conf. Interval]			% Weight
Yehuala et al. (2015)	25.150	21.289	29.011	3.27
Tamene et al. (2020)	36.150	32.563	39.737	3.29
Yismaw et al. (2019)	28.880	24.960	32.800	3.26
Abebe et al. (2019)	25.540	21.346	29.734	3.24
Mengistu et al. (2020)	18.600	15.856	21.344	3.34
Chengo et al. (2020)	15.110	11.131	19.089	3.26
Muchem et al. (2018)	6.440	5.382	7.498	3.42
Mmbaga et al. (2016)	18.420	16.205	20.635	3.37
Seid et al. (2019)	9.270	7.800	10.740	3.41
Zuniga et al. (2013)	20.260	13.890	26.630	3.02
Shah et al. (2012)	5.790	4.614	6.966	3.41
Ndelema et al. (2016)	27.920	23.706	32.134	3.24
Andegior et al. (2020)	16.530	11.846	21.214	3.19
Namazzi et al. (2020)	21.070	15.935	26.205	3.15
Moshiro et al. (2019)	20.750	15.634	25.866	3.15
Orsido et al. (2019)	39.620	33.035	46.205	2.99
Roro et al. (2019)	13.590	8.906	18.274	3.19
van den et al. (2015)	22.270	18.428	26.112	3.27
Egesa et al. (2020)	35.690	30.359	41.021	3.13
Dessu et al. (2018)	34.580	25.564	43.596	2.69
Opio et al. (2019)	7.810	3.165	12.455	3.20
Sania et al. (2013)	3.490	2.373	4.607	3.42
Dessu et al. (2020)	8.330	4.645	12.015	3.28
Farah et al. (2018)	6.940	4.549	9.331	3.36
Paul et al. (2020)	4.800	4.036	5.564	3.42
Gudayu (2012)	33.120	30.905	35.335	3.37
Endalam et al. (2020)	31.210	27.290	35.130	3.26
Miyosh et al. (2019)	4.400	3.420	5.380	3.42
Salih et al. (2013)	41.000	31.357	50.643	2.61
Garca- et al. (2017)	11.560	6.386	16.734	3.15
Nyakan et al. (2019)	13.740	9.565	17.915	3.24
D + L po ES	19.203	16.019	22.387	100.00

Heterogeneity chi-squared = 1947.64 (d.f. = 30) p = 0.000.

I-squared (variation in ES attributable to heterogeneity) = 98.5%.

Test of ES = 0: z = 11.82 p = 0.000.

Figure 2.

Forest plot showing the pooled estimate of mortality among preterm in East Africa, from January 2012–December 2020.

3.4. Publication bias

Egger's regression test value showed that there is a statistically significant publication bias (p < 0.000). Besides, a funnel plot showed an asymmetrical distribution which indicated the presence of publication bias (Table 3 & Figure 3).

Table 3.

Egger's test of the study involved 32 studies on preterm mortality in East Africa from January 2012–December 2020.

Std_Eff	Coef.	Std. Err.	T	P > t	[95% Conf. Interval]
Slope	.5574478	.0524259	10.63	0.000	.4502247	.6646708
Bias	.3443868	.0500503	6.88	0.000	.2420225	.4467511

Figure 3.

Funnel plot to test the publication bias of the 32 studies, log proportion (x-axis) with a standard error of log proportion (y-axis).

Sensitivity analysis: The results of sensitivity analyses using the random effect model suggested that there is no single study that influenced the overall estimation significantly. Besides, the sensitivity analysis is displayed graphically (Table 4).

Table 4.

Summary of the sensitivity on mortality rate among preterm in Eastern Africa from January 2012–December 2020.

Study omitted	Estimate	[95% Conf. Interval]
Yehuala et al. (2015)	18.99383	15.792513	22.195148
Tamene et al. (2020)	18.584991	15.508882	21.6611
Yismaw et al. (2019)	18.861759	15.687027	22.036489
Abebe et al. (2019)	18.983465	15.779122	22.187807
Mengistu et al. (2020)	19.223606	15.987537	22.459675
Chengo et al. (2020)	19.343513	16.096169	22.59086
Muchem et al. (2018)	19.698416	16.247002	23.149832
Mmbaga et al. (2016)	19.229933	15.993988	22.465878
Seid et al. (2019)	19.580208	16.217562	22.942852
Zuniga et al. (2013)	19.169939	15.939093	22.400784
Shah et al. (2012)	19.711596	16.308544	23.114645
Ndelema et al. (2016)	18.900389	15.710692	22.090086
Andegior et al. (2020)	19.292683	16.05171	22.533653
Namazzi et al. (2020)	19.141171	15.912008	22.370335
Moshiro et al. (2019)	19.151745	15.921703	22.381784
Orsido et al. (2019)	18.561008	15.38846	21.733557
Roro et al. (2019)	19.390844	16.14489	22.636799
van den et al. (2015)	19.095337	15.876361	22.314314
Egesa et al. (2020)	18.654905	15.487803	21.822006
Dessu et al. (2018)	18.77424	15.567248	21.981234
Opio et al. (2019)	19.582634	16.333736	22.83153
Sania et al. (2013)	19.783962	16.425308	23.142618
Dessu et al. (2020)	19.576698	16.319117	22.834278
Farah et al. (2018)	19.640337	16.356947	22.923727
Paul et al. (2020)	19.767809	16.256323	23.279293
Gudayu (2012)	18.614647	15.72389	21.505405
Endamaw et al. (2020)	18.777998	15.623595	21.9324
Miyosh et al. (2019)	19.764885	16.34239	23.18738
salih et al. (2013)	18.614325	15.418252	21.810398
Garca- et al. (2017)	19.453926	16.208763	22.699091
Nyakan et al. (2019)	19.389067	16.140238	22.637897
Kolobo et al. (2020)	19.203331	16.019365	22.387295
Combined	19.203331	16.019366	22.387296

3.5. Subgroup analysis of preterm mortality rate in Eastern Africa

The subgroup analysis was employed to estimate the pooled mortality rate by stratifying the studies into different categories. In this regard, the studies were stratified by country, study design, year of publication, and sample size. The mortality rate regarding by country including Ethiopia, Tanzania, Burundi, Eritrea, and Uganda was found to be 23.7%,14.3%,24.5%,10.9%, and 17.2% respectively. Besides, the mortality rate by study design was found to be 18.1%, 19.4%, and 19.7% for a prospective cohort, retrospective cohort, and cross-sectional study respectively. Based on the year of publication, the mortality rate was found to be 19.2% in the study conducted in between 2012-2018. Similarly, the mortality rate was 19.3% in the study period from 2019-2020. Moreover. the mortality rate was found to be19.2 and 17.9% with the sample size <350 and ≥350, respectively (Table 5).

Table 5.

Summary of subgroup analysis on mortality rate among preterm in Eastern Africa by country, design, year of publication, and sample size from January 2012–December 2020.

Variable	Characteristics	Pooled prevalence % (95%CI)	I2, (p-value)
Country	Ethiopia	23.7 (17.2–30.3)	98.2% (<0.001)
	Tanzania	14.3 (4.5–24.1)	98.4% (<0.001)
	Burundi	24.5 (17.0–31.9)	74.1% (<0.049)
	Eritrea	10.9 (0.4–21.4)	94.7% (<0.001)
	Uganda	17.2 (3.8–30.6)	98.1% (<0.001)
By study design	Prospective cohort	18.1 (12.0–24.2)	93.3% (<0.001)
	Retrospective Cohort	19.4 (14.1–24.8)	99.0% (<0.001)
	Cross-sectional	19.7 (14.3–25.1)	98.3% (<0.001)
By the year of publication	2012–2018	19.2 (13.8–24.6)	98.7% (<0.001)
	2019–2020	19.3 (15.0–23.7)	98.3% (<0.001)
By sample size	<350	19.2 (16.0–22.4)	92.8% (<0.001)
	≥350	17.9 (13.9–22.0)	99.0% (<0.001)

3.6. The impact of respiratory distress syndrome on the survival of preterm neonates

Seven studies from 32 studies reported that respiratory distress syndrome has an impact on the survival of preterm [33, 35, 37, 38, 46, 47, 49]. In this regard, the pooled effect of respiratory distress syndrome on the survival of preterm neonates showed that respiratory distress syndromes were 3.2 times increase risk of death in preterm neonates as compared to preterm neonates without respiratory distress syndrome [AOR = 3.2; 95% CI: 22, 4.6] (Figure 4).

Figure 4.

The pooled effect of respiratory distress syndrome on the pooled estimate of mortality among preterm in East Africa, from January 2012–December 2020.

3.7. The impact of birth asphyxia on the survival of preterm neonates

Nine studies from 32 studies reported that birth asphyxia had a significant association with preterm mortality [33, 35, 36, 37, 38, 39, 41, 47, 49]. In this concern, the pooled effect of birth asphyxia on the survival of preterm neonates showed that preterm presented with birth asphyxia nearly 2.6 times [AOR = 2.6; 95% CI: 1.9, 3.4] had a higher risk of death as compared to preterm without birth asphyxia (Figure 5).

Figure 5.

The pooled effects of birth asphyxia on the pooled estimate of mortality among preterm in East Africa, from January 2012–December 2020.

4. Discussion

In this systematic review and meta-analysis, the pooled mortality rate among preterm in East Africa was found to be 19.2%. This is consistent with the studies conducted in Asia, India, and Nigeria [48, 49, 51]. However, The mortality rate reported in this study is higher than studies reported from Germany [59], Iran [60], Australian [61], China [62], Libya [63], Mexico [64], Spain [65], and France 4.9% [66]. Moreover, the mortality rate of preterm associated with respiratory distress syndrome and asphyxia in this study is much higher than studies conducted in global, western Europe, Eastern Europe & Central Asia, and North America [67, 68]. In contrast, the mortality in this study was lower than the study conducted in Ghana, Brazil, and Pakistan [9, 69, 70].

The discrepancy might be due to sampling size, study settings, study period, and/or characteristics of study participants. Moreover, in developing countries, there is low skilled care at birth, antenatal care visits, low advancement of medical care, and, delayed health-seeking behavior than industrialized countries. Moreover, few women and newborns stay in the health facility for the recommended 24 h after birth in a resource limited-setting, which is the most critical time when complications can present [7, 71].

Subgroup analysis of mortality rate by country includes Ethiopia, Tanzania, Burundi, Eritrea, and Uganda was found to be 23.7%, 14.3%, 24.5%, 10.9%, and 17.2%. This variation can be explained by socioeconomic and cultural variation between the countries, the health status of the mother or father of the neonate, sample size, and study period.

Subgroup analysis regarding the study design showed that the mortality rate was found to be 18.1%, 19.4%, and 19.7% for the prospective cohort, retrospective cohort, and cross-sectional respectively. The reason could be a cross-sectional study including prevalence and incidence, while cohort studies only considered the incidence cases. Besides, subgroup analysis based on the year of publication advocated that the mortality rate was found to be 19.2% in studies conducted from 2012-2018. Similarly, it was 19.3% from 2019-2020. Moreover, the mortality rate regarding sample size subgroup analysis was found to be 19.2% and 17.9% with sample size <350 and ≥350 respectively.

The odds of mortality were higher nearly by three folds [AOR = 3.2; 95% CI: 22, 4.6] among preterm neonates with respiratory distress syndrome as compared to preterm neonates without respiratory distress syndrome. This is also supported by the findings in other developing countries [10, 35, 37, 46].

Respiratory distress syndrome is more fatal and prevalent in preterm neonates predominantly in resource-limited settings [72]. Besides, the treatment modalities, including antenatal corticosteroids, surfactants, and advanced respiratory care of the neonate are limited in Easter African countries [71]. The odds of mortality were increased 2.6 times among preterm with birth asphyxia when compared to their counterparts [AOR = 2.6; 95% CI: 1.9, 3.4]. Similar findings were also reported from previous studies [10, 73, 74]. Indeed, preterm neonates with birth asphyxia usually present hypoxic-ischemic encephalopathy, seizures, and cerebral palsy due to hypoxia. In this regard, respiratory distress syndrome and birth asphyxia.

4.1. Limitations of the study

In this study, only quantitative observational studies published in English were included in the analysis, and case series, case reports, and qualitative findings were excluded. Besides, we determined the subgroup analysis in different strata, heterogeneity was observed in some stratified groups. Thus, these are the limitations of the studies that the reader advised to be considered.

5. Conclusion

Preterm mortality was found to be unacceptably high in Eastern Africa continent.

Fortunately, the main causes of death were found to be respiratory distress syndrome and birth asphyxia which are preventable and treatable, hence early detection and timely management of this problem are highly recommended to improve preterm survival.

Declarations

Author contribution statement

Ermias Sisay Chanie: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Abebew Yeshambel Alemu, Demewoze Kefale Mekonen, Biruk Demissie Melese, Binyam Minuye, Habtamu Shimels Hailemeskel and Worku Necho Asferie: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data.

Wubet Alebachew Bayih, Tigabu Munye, Tekalign Amera Birlie, Abraham Tsedalu Amare, Nigusie Selomon Tibebue, Chalie Marew Tiruneh, Getasew Legas, Fisha Alebel Gebre Eyesus, and Demeke Mesfin Belay: Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability statement

Data will be made available on request.

Declaration of interests statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.

Appendix A. Supplementary data

The following is the supplementary data related to this article: