Birth prevalence and risk factors of neural tube defects in Ethiopia: a systematic review and meta-analysis

Abstract

Objective

This study aims to estimate the prevalence of neural tube defects (NTDs) and to identify potential risk factors in the Ethiopian context.

Study design

Systematic review and meta-analysis.

Study participants

A total of 611 064 participants were included in the review obtained from 42 studies.

Methods

PubMed (Medline), Embase and Cochrane Library databases in combination with other potential sources of literature were systematically searched, whereby studies conducted between January 2010 and December 2022 were targeted in the review process. All observational studies were included and heterogeneity between studies was verified using Cochrane Q test statistics and I² test statistics. Small study effects were checked using Egger’s statistical test at a 5% significance level.

Result

The pooled prevalence of all NTDs per 10 000 births in Ethiopia was 71.48 (95% CI 57.80 to 86.58). The between-study heterogeneity was high (I²= 97.49%, p<0.0001). Birth prevalence of spina bifida (33.99 per 10 000) was higher than anencephaly (23.70 per 10 000), and encephalocele (4.22 per 10 000). Unbooked antenatal care (AOR 2.26, 95% CI (1.30 to 3.94)), preconception intake of folic acid (AOR 0.41, 95% CI (0.26 to 0.66)), having chronic medical illness (AOR 2.06, 95% CI (1.42 to 2.99)), drinking alcohol (AOR 2.70, 95% CI (1.89 to 3.85)), smoking cigarette (AOR 2.49, 95% CI (1.51 to 4.11)), chewing khat (AOR 3.30, 95% CI (1.88 to 5.80)), exposure to pesticides (AOR 3.87, 95% CI (2.63 to 5.71)), maternal age ≥35 (AOR 1.90, 95% CI (1.13 to 3.25)), maternal low educational status (AOR 1.60, 95% CI (1.13 to 2.24)), residing in urban areas (AOR 0.75, 95% CI (0.58 to 0.97))and family history of NTDs (AOR 2.51, 95% CI (1.36 to 4.62)) were associated with NTD cases.

Conclusion

The prevalence of NTDs in Ethiopia is seven times as high as in other Western countries where prevention measures are put in place. Heredity, maternal and environmental factors are associated with a high prevalence of NTDs. Mandatory fortification of staple food with folic acid should be taken as a priority intervention to curb the burden of NTDs. To smoothen and overlook the pace of implementation of mass fortification, screening, and monitoring surveillance systems should be in place along with awareness-raising measures.

PROSPERO registration number

CRD42023413490.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Screening, data extraction and quality assessment were done by two authors independently.

• Recent studies were included to better estimate the birth prevalence and determine factors associated with neural tube defects.

• Utilisation of the updated Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement coupled with the quality monitoring checklist.

• Lack of publication from Benishangul Gumuz, Gambella, Afar, Central Ethiopia and Southwest Ethiopia regions.

Introduction

Neural tube defects (NTDs) are the second most common form of congenital anomaly (CA) following congenital heart defects. Those defects occurred due to perturbation in the process of neurulation, which is usually expected to be completed within 4 weeks after fertilisation.^{1 2} NTDs are broadly classified as closed and open based on their physical characteristics.³ Physical characteristics along with the location and extent of the defect had a role in determining the defect’s prognosis.⁴

Globally, more than a quarter million births were affected by NTDs, and the prevalence is more prominent in Eastern Mediterranean, South-East Asian and African regions.^{5 6} Spinal bifida and anencephaly are two common forms of NTDs reported from different corners of the world and contribute to 29% of neonatal death in low-income countries.^{5 7} However, the paucity of routinely generated data to monitor the trend and effectiveness of measures implemented to curb congenital animalities is the major gap to be filled in resource-constrained settings.⁸ Acknowledging the gap, the WHO recommends the establishment of a surveillance system and CAs registry mechanism.⁹ Despite the recommendation, the implementation was not materialised in most African countries with few expectations like South Africa, which had a relatively matured system.^{10 11}

Ethiopia is one of the countries with a very high burden of stillbirth and neonatal mortality, with a rate of 24.6, and 33 occurrences per 1000 live births, respectively.^{12 13} In Ethiopia, the prevalence of NTDs ranges between 60 and 63 per 10 000 births, which is nearly two times higher than the estimated prevalence in East Africa.^{14 15} Furthermore, in Ethiopia, out of every 10 000 births, 104 result in death due to NTDs specifically craniorachischisis (it is a severe form of NTDs characterised by openings in both the brain and spinal cord).¹⁶

Globally, various prevention modalities spanning from the advanced genetic test to the simplest micronutrient fortification with folic acid were put forward to avert untoward health consequences of NTDs and circumvent the economic burden imposed on the health system.^17–19 Along with the establishment of the surveillance programme, prenatal examination (ie, antenatal ultrasonography, amniocentesis and genetic counselling) is one of the identification platforms that is designed for screening NTD during pregnancy and taking timely measures, including elective termination of the pregnancy.⁶ Ultrasound examination, measuring homocysteine concentration and gauging maternal serum alpha-fetoprotein levels are some of the commonly used investigative mechanisms. However, in Ethiopia, those advanced laboratory investigations are not feasible and affordable for routine use.²⁰ Henceforth, the most suitable, practicable and scalable means of curbing the burden of NTDs is the fortification of staple foods with folic acid.^{21 22} Despite its comparative advantage and ability to prevent most NTDs, except few exceptions that are resistant to folic acid fortification,^{23 24} mass fortification of staple foods has not yet been fully implemented in most sub-Saharan countries.^{14 25}

In general, the formation of NTDs could stem from hereditary origins to environmental factors, although the mechanism of NTDs pathogenesis has not been well elucidated.^{26 27} A history of NTDs, family history of NTDs and consanguineous marriages specifically with first-degree relatives are identified as hereditary factors that contribute to NTDs.^28–32 Maternal nutrition status specifically related to the consumption of folic acid at preconception and during the first trimester of pregnancy is also intertwined with NTDs.³³ Exposure to teratogenic factors such as herbal or unknown medication,³⁴ pharmaceutical elements (antiepileptic medications, opioid medication and antiretroviral treatment),^{35 36} industrial waste and pollutants (polycyclic aromatic hydrocarbons, heavy metals and pesticides)^37–42 have a strong connection with the formation of NTDs. The risk of developing NTDs is not only limited to the aforementioned factors, it could also be associated with a medical history and personal habits of the mother, such as smoking habits and alcohol consumption.^43–46

Recent evidence coming out of the Ethiopian perinatal death surveillance and response system indicates that CAs have a significant role in perinatal mortality.^47–49 In addition to this, limited capacity to manage NTD cases at tertiary facilities, resulting in poor prognoses after treatment due to high rates of readmission and reoperation, is another critical factor contributing to the escalated national burden at this point.^{50 51} In Ethiopia, preventative measures have not yet been endorsed and implemented in such a way that would put a solid footing to combat the untoward impacts of NTDs due to a lack of feasibility assessment, local capacity and policy directives that would enforce preventive measures.²⁵ In addition to the aforementioned reasons, the current review is also conducted to fill gaps in the existing work of literature in terms of geographical coverage and sample size. Considering all this, this review is carried out to provide a consummate and updated picture of the magnitude and possible risk factors that could contribute to NTDs in Ethiopia.

Materials and method

Protocol and registration

Before conducting the review, the protocol for the review was registered at the International Prospective Register of Systematic Review (PROSPERO), University of York Center for Reviews and Dissemination with link https://tinyurl.com/nt3fvcxz and ref no ID CRD42023413490. As depicted in online supplemental table 1, the updated Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement coupled with the checklist was used to report this review.⁵²

Study design and search strategy

Three databases (ie, PubMed (Medline), Embase and Cochrane Library) in combination with other potential sources of grey literature were used for the review. Articles published in English language between January 2010 and December 2022, which are exclusively focused on Ethiopia, were eligible for further consideration in the review process. As depicted in online supplemental table 2, the search terms of interest included Medical Search Headings (MeSH) and keywords “neural tube defects” OR “congenital abnormalities” OR “anencephaly” OR ”encephalocele” OR “spina bifida” OR “meningocele” OR “myelomeningocele” AND [“Ethiopia”]. Duplicate studies were primarily extracted and excluded via EndNote software. To maintain the quality and foster the pace of extraction, two independent reviews were assigned to identify studies based on their titles and abstracts. Those identified lists of articles during the initial search were transferred to EndNote to vigilantly screen duplication. The selection of articles has proceeded per the defined set of criteria, the studies were selected if they (1) were conducted in Ethiopia; (2) were observational studies (cross-sectional, cohort and case controls), and randomised controlled trials reporting the incidence or prevalence estimates with and without confidence bounds of neural tube defects (encephalocele, spina bifida or anencephaly), (3) published in the English language and (4)) population-based (all cases in a defined geographical area, or hospital or community-based surveillance). Commentaries, systematic reviews and studies not conducted on human subjects were excluded from the review.

Data extraction and quality assessment

After the extraction of potential studies, two reviewers independently screened the full-text articles for eligibility. Afterward, the quality of studies was objectively measured using Joanna Briggs Institute checklist designed for various study methods.⁵³ Disagreement between the reviews on the quality of studies was handled through the direct involvement of the primary author in the evidence-based discussion. Based on the quality assessment criteria specific to each study design, only studies that achieved a score of 50% or above were deemed eligible for further analysis.⁵⁴ Subsequently, pertinent data elements were extracted from articles that met the inclusion criteria. Year of publication, region, year of study, study design, sample size, study outcome, the total number of NTDs as well as the specific number of birth anencephaly, spina bifida, meningocele and encephalocele were extracted through a predesigned data extraction tool.

Statistical analysis

The entire analyses of the study were conducted using STATA V.17. The primary outcome of the review was the birth prevalence of NTDs reported per 10 000 births (live births only or live births+stillbirths only or live births+stillbirths+spontaneous abortions) and the secondary outcome of the review was to identify predictors of NTDs. The meta prop command of STATA was used to compute the pool effect estimate.⁵⁵ Meta-analysis was performed with the DerSimonian-Laird random-effects model with Hartung-Knapp-Sidik-Jonkman variance correction.⁵⁶ This random effect model was selected due to its potential to capture the variation and uncertainty of the true effect stemming from the study population, study design and sample size.⁵⁷ Individual and pooled estimates were graphically displayed using a forest plot. Moreover, a leave-one-out sensitivity analysis was done to point out the influence of a single study on the overall estimate. Among studies heterogeneity was assessed using I² statistics, expressed as % (low (25%), moderate (50%) and high (75%) and Cochrane’s Q statistic (significance level < 0.05).⁵⁸ Subgroup analysis with multivariate meta-regression was performed to objectively identify the source of heterogeneity. Finally, to ascertain the presence of publication bias both the funnel plot and Egger’s test were conducted, and the p<0.05 value was used as a cut-off point to declare the bias.

Patient and public involvement

Patients and public were not involved in the design, conduct, reporting and dissemination plans of this research.

Results

A total of 285 articles were retrieved from various databases, out of which 91 articles were eligible for full-text review. Of the included studies, 42 studies met the inclusion criteria to be part of the review (figure 1).

Figure 1.

Flow diagram showing the procedures of selecting studies for the meta-analysis on the birth prevalence and risk factors of neural tube defects in Ethiopia, 2022. NTD, neural tube defect.

Characteristics of included studies

Twenty-two cross-sectional, 19 case-control and 1 cohort study were included in the final review. With respect to the geographical coverage of studies, a total of eight studies were cross-regional studies (studies taken in two or more regions of the country), among them three studies took place in Addis Ababa and Amhara region simultaneously,^59–61 four studies were conducted in east part of county including Oromia, Harari, Somalia and Dire Dawa city administration^62–65 and one study was carried out in Sidama and Southern Ethiopia region.⁶⁶ On the other hand, the remaining studies were exclusively conducted in four regions and one city administration (Addis Ababa=5,^67–71 Amhara=13,^72–84 Harari=1,⁸⁵ Oromia=10^86–95 and Tigray=5^96–100). As displayed in the annex section, online supplemental table 3, a total of 611 064 participants were involved in 42 studies. From those 42 studies, 6621 NTD cases were identified with a minimum of 6 cases from Addis Ababa to a maximum of 1873 cases from a cross-regional study.^{59 67}

Meta-analysis

Birth prevalence of NTDs

Twenty-two primary studies were included to estimate the birth prevalence of NTDs in Ethiopia. Accordingly, the pooled prevalence of NTDs was 71.48 cases (95% CI 57.80 to 86.58) per 10 000 births (figure 2).

Figure 2.

Forest plot showing the pooled birth prevalence of neural tube deficits in Ethiopia. ES,Estimated Prevalence.

Handling heterogeneity

To handle the observed heterogeneity among extracted studies various modalities were used including sensitivity analysis and meta-regression. Per the leave-one-out sensitivity analysis approach, none of the studies excessively influenced the estimated birth prevalence of NTDs as online supplemental figure 1. To examine the level of heterogeneity, subgroup analysis was performed using region, study period, sample size and study design. Based on region, the highest birth prevalence of NTDs was observed in the Tigray region (159.15 NTD cases per 10 000 births), whereas the lowest prevalence was reported from Harari (12.36 NTD cases per 10 000 births). With the respective study period, the birth prevalence of NTDs was higher among studies conducted during the era of Sustainable Development Goals, which was 81.89 NTD cases per 10 000 births per pooled report from 16 studies, while the prevalence was 49.67 cases per 10 000 births from 6 studies conducted during millennium development goals. Moreover, pooled NTDs of smaller sample size studies was 133.97 per 10 000 births (table 1).

Table 1.

Subgroup analysis of the birth prevalence of neural tube defects by study characteristics in Ethiopia

Subgroup	Category	No of studies	Birth prevalence of NTDs per 10 000 births (95% Cl)	Heterogeneity I2 (%)	P value
Region	Addis Ababa	4	71.55 (50.17 to 96.51)	69.80	<0.001
	Amhara	6	46.07 (37.65 to 55.32)	80.15_	_<0.001
	Harari	1	12.36 (8.31 to 18.38)	_	_
	Oromia	4	64.04 (37.61 to 97.11)	87.33	_<0.001
	Tigray	3	159.15 (101.84 to 228.83)	_	_
	Cross-regional *	4	87.77 (57.83 to 123.83)	98.79	_<0.001
Study period	During MDGs†	6	49.67 (38.75 to 61.92)	94.59	<0.001
	During SDGs‡	16	81.89 (57.86 to 109.55)	97.80	<0.001
Sample size	Smaller size§	3	133.97 (68.14 to 220.25)	_	_
	Mega size¶	19	67.04 (53.43 to 82.17)	97.78	<0.001
Study design	Cross-sectional	19	66.91 (53.70 to 81.52)	97.04	<0.001
	Case–control	2	53.32 (42.40 to 65.46)	_	_
	Cohort	1	107.08 (98.32 to 116.61)	_	_

*Study conducted in two or more regions of Ethiopia.

†Study took place between 2010 and 2015.

‡Study conducted between 2016 and 2022.

§Study with a sample size below 1000 observations.

¶Study with a sample size of 1000 or more observations.

Meta-regression

To objectively identify the source of variation, multivariable meta-regression was performed and as a result, none of the variables were indicated with the source of heterogeneity as indicated in online supplemental table 4.

Publication bias

Assessment of small study effects using funnel plots showed a largely symmetric distribution of studies around the random effects estimate of the birth prevalence of NTDs (figure 3). Moreover, Egger’s regression test was p=0.1749, which shows the absence of publication bias.

Figure 3.

Funnel plot showing the results of the publication bias among studies, 2022.

Birth prevalence of various types of NTDs

Next, we assess specific NTDs—Spina bifida, anencephaly and encephalocele. Among the NTDs, the birth prevalence of spina bifida was 33.99 (95% CI 25.59 to 43.52) cases per 10 000 births per pooled report from 21 studies (figure 4). Egger’s regression test was p=0.1945, which shows the absence of publication bias, and per leave-out analysis, no studies affected the pooled prevalence. According to subgroup analysis by region, the highest burden was reported from the Tigray region (102.02 cases per 10 000 births), and the lowest was observed in the Harari region (10.30 cases per 10 000 births), see the detailed in online supplemental figure 2.

Figure 4.

Forest plot showing the pooled birth prevalence of spina bifida in Ethiopia. ES, Estimated Prevalence.

Similarly, according to a complied report from 17 studies, the birth prevalence of anencephaly was 23.70 (95% CI 13.96 to 35.84) cases per 10 000 births (figure 5). Egger’s regression test was p=0.2276, which shows the absence of publication bias, and per leave-out analysis, no studies affected the pooled prevalence. Per subgroup analysis by region, the highest burden was observed in the Tigray region (45.60 cases per 10 000 births) while the lowest was reported in the Amhara region (13.66 cases per 10 000 births) (see the detail in online supplemental figure 3).

Figure 5.

Forest plot showing the pooled birth prevalence of anencephaly in Ethiopia. ES, Estimated Prevalence.

Moreover, encephalocele was reported less frequently, per the aggregate report of 10 studies, the birth prevalence was 4.22 (95% CI 1.49 to 8.08) cases per 10 000 births (figure 6). No publication bias was reported according to the Egger’s test for encephaloceles, which was p=0.0824.

Figure 6.

Forest plot showing the pooled birth prevalence of encephaloceles in Ethiopia. ES, Estimated Prevalence.

Risk factors of neural tube defects

A total of 19 potential risk factors, composed of 5 major categories, were included in the review and 13 of them were significantly associated with NTDs. The predictors indicate that, under the category of maternal sociodemographic characteristics, being older than 35 (AOR 1.92, 95% CI (1.13 to 3.25)) and being an uneducated woman (AOR 1.60, 95% CI (1.13 to 2.24)) increased the risk of having NTDs affected perinate. Residing in an urban area (AOR 0.75, 95% CI (0.58 to 0.97)) reduced the probability of having NTDS affected perinate by 25%.

Among risk factors indicated under obstetric and medical-related, nutritional characteristics of mothers, being unbooked for antenatal care (ANC) (AOR 2.26, 95% CI (1.30 to 3.94)), having a family history of NTDs (AOR 2.51, 95% CI (1.36 to 4.62)) and having a chronic illness (AOR 2.06, 95% CI (1.42 to 2.99)) increased risk of giving birth with NTDs. However, preconception intake of folic acid lowers the risk of having NTDs affected perinate by 59% (AOR 0.41, 95% CI (0.26 to 0.66)).

With respect to exposure status to pharmaceuticals and environmental hazards, utilisation of herbal medicine (AOR 3.91, 95% CI (2.61 to 5.84)) and illicit drugs (AOR 3.59, 95% CI (2.18 to 5.19)) as well as a history of exposure to pesticides (AOR 3.87, 95% CI (2.63 to 5.71)) and living around a polluting industry (AOR 1.77, 95% CI (1.62 to 1.93)) increase the risk of having NTDs affected birth. On the other hand, personal habits such as consuming alcohol (AOR 2.70, 95% CI (1.89 to 3.85)) and smoking cigarettes (AOR 2.49, 95% CI (1.51 to 4.11)) heightened the risk of having a child with NTDs. Moreover, the habit of chewing khat (khat is called Catha edulis Forsk by its scientific name, is an evergreen, mild-narcotic, flowering tree that mostly grows in East Africa¹⁰¹) is found to increase the risk of giving birth with NTDs in three folds (AOR 3.30, 95% CI (1.88 to 5.80)). However, six risk factors, namely the sex of perinate, balanced diet consumption, maternal mid-upper arm circumference status, utilisation of oral contraceptives and consanguinity marriage, were not associated with NTDs formation, as their CIs contained an AOR of 1. Lastly, as depicted in table 2, Eggers’s regression test revealed that there was no sign of publication bias, as per the defined criteria.

Table 2.

Pooled OR of the associations between neural tube defects and their associated factors

Group of characteristics	Characteristics	Category	Pooled OR (95% CI)	No of studies	Egger’s test	I2 (%)
	Maternal age
1		≥35	1.92 (1.13 to 3.25)	20	0.9081	86.66
		≤35	1
	Maternal education
		Uneducated	1.60 (1.13 to 2.24)	13	0.7707	75.39
		Educated	1
	Consanguinity marriage
		Yes	2.68 (0.38 to 18.87)	6	0.5535	84.15
		No	1
	Maternal residence
		Urban	0.75 (0.58 to 0.97)	20	0.6409	71.13
		Rural	1
2	ANC follow-up
		Without ANC follow-up	2.26 (1.30 to 3.94)	16	0.3268	89.96
		With ANC follow-up	1
	Folic acid intake
		Yes	0.41 (0.26 to 0.66)	25	0.9614	88.82
		No	1
	Family history of neural tube deficit
		Yes	2.51 (1.36 to 4.62)	11	0.1484	70.52
		No	1
	History of chronic illness
		Yes	2.06 (1.42 to 2.99)	21	0.0883	79.53
		No	1
	Oral contraceptive use
		Yes	1.22 (0.54 to 2.76)	5	0.1894	81.57
		No	1
	Maternal MUAC
		MUAC≤23	1.61 (0.27 to 9.46)	3	0.8926	87.87
		MUAC≥23	1
	Balanced diet consumption
		Yes	0.59 (0.15 to 2.29)	5	0.3883	96.23
		No	1
3	Herbal medication
		Yes	3.91 (2.61 to 5.84)	5	0.9632	0
		No	1
	Illicit drug intake
		Yes	3.59 (2.18 to 5.19)	16	0.5313	90.83
		No	1
	Pesticide exposure
		Yes	3.87 (2.63 to 5.71)	10	0.3211	20.94
		No	1
	Living near the polluting industry
		Yes	1.77 (1.62 to 1.93)	3	0.4071	0.00
		No	1
4	Alcohol intake
		Yes	2.70 (1.89 to 3.85)	18	0.8129	74.73
		No	1
	Smoking status
		Smoker	2.49 (1.51 to 4.11)	14	0.058	60.04
		Nonsmoker	1
	Chewing khat
		Yes	3.30 (1.88 to 5.80)	7	0.8176	79.00
		No	1
5	Sex of neonate
		Male	0.88 (0.70 to 1.09)	17	0.8568	66.13
		Female	1

1:Maternal sociodemographic characteristics.

2:Obstetric and medical related, nutritional characteristics of mothers.

3:Status of exposure to pharmaceuticals and environmental hazards.

4:Personal habits.

5:Newborn-related characteristics.

ANC, antenatal care; MUAC, maternal mid-upper arm circumference.

Discussion

In this meta-analysis, the prevalence of NTDs in Ethiopia was nearly 1 in every 10 births. The birth prevalence of spina bifida was nearly two times higher than anencephaly and eight times higher than encephalocele. Sociodemographic characteristics (maternal age and maternal education status), medical history (chronic medical illness), health-seeking behaviour (ANC visit), nutritional status (intake of folic acid), environmental exposure (pesticide exposure and living close to industrially polluted areas), hereditary factors (family history of NTDs), exposure to pharmaceuticals (illicit and herbal medication) and personal habit (drinking alcohol, smoking cigarettes and chewing khat) were associated with NTDs.

The pooled birth prevalence of NTDs in Ethiopia was three times higher than in other East African countries and two to five times higher than sub-Saharan Africa and the Africa continent. In comparison to most Western countries, it becomes ten times higher.^{14 102–106} This high birth prevalence of NTDs in Ethiopia could be explained due to a lag in taking more drastic measures in primary prevention, such as the mandatory fortification of staple food with folic acid. In line with this, there is ample evidence indicating that effective utilisation of mass fortification not only significantly reduces the burden but also the medical treatment costs of NTDs.^{7 17} Additionally, notable regional variation was observed in the burden of NTDs in Ethiopia. In this regard, the burden is nearly 2–3 times higher in the regions of Tigray, Harari and Somalia as compared with the national estimate.^{64 96} This variation may be explained by the fact that 78% of women within the reproductive age group had low RBC folate concentrations and the concentrations become much lower among pregnant women.^{107 108} Additionally, ecologically, the burden is disproportionately concentrated in the highland sorghum khat mixed farming system.¹⁰⁹ In this context, the eastern parts of the county, including the Somali and Harari regions, are grouped under the aforementioned farming system, which may have contributed to a high burden of NTDs. With regards to the Tigray region, the risk of NTDs is higher as compared with the remaining part of the country, partly explained by the high prevalence of folate deficiency among women of reproductive age.¹¹⁰ However, it is important to consider that the burden might have worsened due to the conflict that broke out in 2020. This conflict has resulted in the disruption of the region’s health system, as evidenced by the damaged health infrastructures and abscondment of healthcare professionals.^{111 112} In summary, the findings imply that regional-level health system status and farming ecosystem should be considered in order to acrimoniously confront the burden of NTDs.

The review unravels that sociodemographic characteristics such as maternal education and maternal age were associated with NTDs. Perinates born from old age women and uneducated women have a higher risk of being affected by NTDs as compared with other respective categories. A similar finding was reported from Colombia¹¹³ and Saudi Arabia¹¹⁴; nevertheless, studies from Eritrea and Pakistan revealed that the burden is more concentrated among younger-aged mothers.^{32 115} This might be explained by older women’s less likelihood of taking supplements that contain folic acid.¹¹⁶ With respect to education, the findings agreed well with studies conducted in Pakistan³² and India.¹¹⁷ The meta-analysis of 45 studies also illustrated⁴⁵ those uneducated women have a higher risk of giving birth to a child with NTDs. In Ethiopia, only a limited number of women were aware of and had access to periconceptional folic acid usage to avert the untoward consequence of NTDs, except for women residing in urban areas.^118–121 In connection with this, per our review, intake of folic acid at preconception and during the early stages of pregnancy has reduced the risk of giving birth to a child with NTDs. However, utilisation of folic acid is more skewed to those who are educated and those who visited a health facility for ANC and family planning services due to better knowledge and access to utilisation.^{119 120} To this effect, women who had ANC visits had better access to use folic acid, which significantly lowered the risk of NTDs as portrayed in our review as well. Considering the benefit of health education for informed decision-making, Ethiopia launched health extension programmes in 2003, to provide health education and basic health service at a community level.^{122 123} In addition, various measures such as mass mobilisation and awareness-raising activities were taken to improve the uptake of ANC service, since it serves as one communication platform to actualise planned targets at the national and local levels.¹³ However, in recent days, the health extension programme falls short of bearing fruits due to a critical gap in the skills and motivation of health extension workers.¹²⁴ Moreover, the suboptimal provision of quality ANC service is a challenge that comes up with wide-scale coverage of the service.^{125 126} With this backdrop, taking measures to optimise the health extension programme and improve the quality of ANC services could be used as an effective channel of communication to increase the uptake of folic acid.

The review also revealed that chronic illness is associated with NTDs. The finding was congruent with studies conducted in India,⁴³ Saudi Arabia¹¹⁴ and Canada¹²⁷ as well as a meta-analysis of 25 studies and multicounty studies of 29 counties.^{103 128} This is related to maternal diabetics, which affect genes involved in neural tube formation by altering routine regulation of miRNAs.¹²⁹ Moreover, pregnant mothers who have chronic hypertension are expected to take medication containing ACE inhibitors and medication also increases the risk of having an NTD child.¹³⁰ In tandem with this, our review also revealed that the intake of both prescribed and unprescribed drugs including herbal medication during the early onset of pregnancy was associated with NTDs. This finding was also supported in studies conducted elsewhere.^{34 35 131} In line with this, in Ethiopia, a significant number of pregnant mothers use herbal medication to alleviate symptoms, during the first trimester of pregnancy, shorten the duration of labour and reduce labour pain.^132–134 Nevertheless, the safety of those medications has not yet been guaranteed by health authorities due to lack of supportive national-level policies and laws.¹³⁵ However, working on improving access to maternal services could reduce the utilisation of herbal medication and possible complications that come up with the medications.¹³⁶ In general, intake and prescription of medication should be considered at all stages of pregnancy to reduce medication induced NTDs. Besides, policies should be designed to integrate traditional medicine into the existing health system.

This review at hand revealed that environmental exposure such as contact with pesticides and living around industrial polluted areas is associated with NTDs. This finding was parallel with studies conducted in Turkey,⁴⁰ China¹³⁷ and the USA.¹³⁸ This might be explained by the fact that most chemical compounds listed as environmental pollutants are lipophilic and can pass through the placental barrier and have the potential to disrupt normal embryonic development including the neurulation process. This deviation from normal embryonic development results in NTDs.^{28 139} In Ethiopia, organohalogenated compounds (ie, organochlorine pesticides, polychlorinated biphenyls and polybrominated diphenyl ethers) are one of the environmental pollutants and its residues were detected in surface water, soil and human breast milk.^{140 141} Moreover, the risk of exposure to those pollutants is higher among residents who live around large-scale farms, river basins and textile industries.^142–144 In general, the observed gap in awareness, waste disposal and monitoring should be addressed to further minimise the impact on human and livestock health as well as the ecosystem.^145–149

This review also revealed that having a family history of NTDs increases the risk of NTDs. A similar finding was reported from studies conducted in other settings.^{26 150} The risk of giving birth to a child with NTDs with first and subsequent pregnancy is higher among those who had NTDs with their first-degree and second-degree relatives. Despite lack of clarity on the pattern of inheritance, close examination of family history during preconception care had a role in the identification of possible hereditarily acquired risk and could pave the way for a timely recommendation of folic acid supplementation.¹⁵¹

Lastly, our review illustrated that personal habits such as smoking cigarettes, drinking alcohol and chewing khat are strongly connected with NTDs. The finding was corroborated by studies carried out elsewhere.^152–154 The plausible explanation related to this could be the fact that both habits (ie, smoking cigarettes and drinking alcohol) have the potential to severely compromise the regulation mechanism of various genes in the neurulation process.^{155 156} Moreover, heavy alcohol use during pregnancy impairs the transportation of folic acid.¹⁵⁷ The combination of all these alterations from normal processes could make the fetus prone to NTDs. In compliance with this, per the animal model study, chewing khat in pregnancy has negative consequences on the fetus, which is expressed in terms of growth retardation and developmental anomalies.¹⁵⁸ In addition, farmers use various pesticides to increase the productivity of khat, and pesticide residues are commonly found in khat beyond the recommended level.^{159 160} This will heighten the risk of giving birth to a child with NTDs among mothers who are chronic khat users that could come from both the khat and the pesticides. Overall, the finding implied those personal habits should be avoided before having pregnancy and should be integrated well as one component of addiction treatment during preconception care.¹⁶¹

The review has the following limitation that needs to be acknowledged. (1) Absence of publication from Benishangul Gumuz, Gambella, Afar, Central Ethiopia region and Southwest Ethiopia region and this could jeopardise the inclusiveness of the review. (2) The high rate of pregnancy loss due to congenital defects could obscure the true burden of NTDs hence might be difficult to get a true picture from the currently available data. (3) Lack of studies conducted at a community level is another drawback of the review that could compromise its representativeness.

Conclusion

The prevalence of NTDs in Ethiopia is profoundly higher compared with neighbouring countries. This high prevalence might be explained by regional variation, sociodemographic characteristics, medical history, health-seeking behaviour, nutritional status, hereditary factors and exposure to environmental and pharmaceutical elements. As a way forward, considering its feasibility and efficacy, mandatory fortification of staple food should be implemented, primarily targeting high-risk regions. This should be accompanied by establishment of a national-level surveillance system to closely monitor the progress. In addition, awareness-raising measures should be taken by using effective modalities of communication such as health extension works and maternal services. Furthermore, the preconception service can be used as one platform to identify high-risk mothers for a timely recommendation of periconceptional folate supplementation.

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information.

Ethics statements

Patient consent for publication

Not applicable.