• MONDAY, SEPTEMBER 23, 2019
  • ISSN 2399-1623
Blessing Jaka Akombi, Pramesh Raj Ghimire, and Andre Masumbuko Renzaho

Abstract

Background

Neonatal mortality is a major public health problem and accounts for about one third of the deaths of children under-5 years. The aim of this study was to estimate the rate and predisposing factors associated with neonatal mortality in the African Great Lakes region (Burundi, Congo Democratic Republic, Kenya, Rwanda, Tanzania and Uganda).

Methods

The study used cross-sectional data from the most recent Demographic and Health Survey of countries in the African Great Lakes region. A total sample of 56600 neonatal deaths of singleton live-born infants was reported across the region. Neonatal mortality was defined as the death of a newborn within the first 28 days of life. Cox regression model was used to identify factors associated with neonatal mortality in the African Great Lakes region.

Results

The Democratic Republic Congo had the highest neonatal mortality rate per 1000 births (19, 95% confidence interval (CI)=17-21) within the region followed by Uganda (18, 95% CI=16-21) and Tanzania (16, 95% CI=14-19). Multilevel logistic regression analysis shows that factors significantly associated with neonatal mortality were: Sex of child (male neonates; hazard ratio (HR)=0.79, 95% CI=0.65-0.96), perceived birth size (small neonates; HR=1.88, 95% CI=1.47-2.39), birth order and birth interval (first child; HR=3.05, 95% CI=2.28-4.07), 4th or more child with > 2 years birth interval (HR=0.48, 95% CI=0.36-0.66), previous death of baby (HR=10.3, 95% CI=7.75-13.68), non-usage of contraceptives (HR=1.81, 95% CI=1.43-2.29), non-receipt of tetanus toxoid injection (HR=1.36, 95% CI=1.04-1.78), no antenatal care visits (HR=1.62, 95% CI=1.06-2.48), and delivery through caesarean section (HR=1.89, 95% CI=1.40-2.55).

Conclusions

To meet the Sustainable Development Goals (SDGs) target for neonatal mortality, interventions should focus on increasing access to good quality healthcare services for women and their newborns as well as family planning services and home-based neonatal care services.


Newborns face the highest risk of dying during the neonatal period – first 28 days of life, at a rate of 19 deaths per 1,000 live births (1, 2). Neonatal mortality accounts for 46% of all under-five mortality with most deaths occurring in low- and middle-income countries (1, 2). Globally, 2.6 million babies died before turning one month old in 2016, with about 50 per cent of these deaths occurring in the first day of life and 75 per cent in the first week of life (1, 2). The overall progress in reducing neonatal mortality rates has been slower than that reported in under-five mortality; while neonatal mortality declined by 49 per cent from 1990 to 2016, the under-five mortality rate declined by 62 per cent within the same period (1, 2). Though the global neonatal mortality rate has decreased from 37 deaths per 1,000 live births in 1990 to 19 in 2016, the rate is still unacceptably high in certain regions and countries around the world (1, 2).

Sub-Saharan Africa (SSA) is one of the regions with the highest prevalence of neonatal mortality (38 per cent) with some degree of variation across its countries and only minor declines over the last two decades (3-5). Within SSA, neonatal mortality rate is unevenly distributed and countries in the Great Lakes region (Burundi, Democratic Republic of Congo (DRC), Kenya, Rwanda, Tanzania and Uganda) contribute to this burden due to decades of political instability, conflicts, poor quality of healthcare governance, inadequate health financing and human health resources, low standard health service delivery, and poor socioeconomic status (6, 7).

The causes of neonatal mortality are well-known which include: preterm birth complications, intrapartum-related complications, birth asphyxia, congenital anomalies and infections such as neonatal sepsis, tetanus, meningitis and pneumonia (1, 2). These causes are often exacerbated by underlying determinants which could be related to the environment of the newborn – whether it is an impoverished family, a marginalized community or a country consumed by conflict. These underlying determinants operate as socioeconomic factors, maternal lifestyle factors, health service factors and environmental factors which could be mitigated through a system-wide approach that strengthens the health systems to provide quality health care. Studies have shown that a vast majority of neonatal deaths are preventable through improved access to good quality maternal and newborn health services delivered by skilled health professionals, clean water supply, proper antenatal and postnatal nutrition for mother and newborn, disinfectants and skin-to-skin contact (1, 8-10). However, in resource-limited countries experiencing prolonged periods of conflict and political instability like those located in the Great African Lakes region, the need for interventions that target the most significant underlying factors is essential to assist in proper planning and strategic allocation of scarce resources. Hence, the aim of this study was to ascertain the most significant factors associated with neonatal mortality across countries in the Africa Great Lakes region in order to assist in national and regional policy prioritization. Findings from this study will contribute to the existing body of evidence buttressing the need for improved quality of health service in tackling the problem of neonatal mortality and also assist in the planning and evaluation of health policies and programmes.

METHODS

This study used datasets from the most recent Demographic and Health Survey (DHS) conducted in all six countries of the African Great Lakes region (Burundi, DRC, Kenya, Rwanda, Tanzania and Uganda). The DHS is a population-based, nationally representative household survey which collects data on maternal and child health, gender, fertility, HIV/AIDS, malaria, family planning and nutrition in developing countries. The DHS programme adopts standardized methods involving uniform questionnaires, manuals, and field procedures to gather information that is comparable across countries. The surveys employ a stratified, multi-stage (cluster), random sampling design and is conducted about every 5 years, to allow comparisons over time. Information was obtained from eligible women aged 15-49 years in each country. The survey uses verbal autopsy questionnaire to collect data for neonatal deaths. Detailed survey methodology and sampling methods used in gathering the data have been reported elsewhere (11-16).

A pooled analysis was conducted to determine the significant factors associated with neonatal mortality across the Great Lakes region. However, Kenya was excluded from the multilevel modelling due to the large number of missing values in the dataset. The analyses were restricted to most recent singleton live births reported in the 5 years preceding the surveys in order to minimise recall bias by the mothers, thus strengthening the internal and external validity of this study. Furthermore, singleton selection eliminated any confounding effect of multiple births that are prone to mortality due to biological factors.

Outcome measure

The main outcome measure for this study was neonatal mortality which refers to the death of a newborn within the first 28 days of life (1). The study outcome was reported as a binary variable with ‘Neonatal death’ coded as 1 and ‘Live birth’ coded as 0. A birth history was used to estimate neonatal mortality in all the individual country DHS.

Study independent variables

The study independent variables used in this analysis were classified based on Mosley and Chen conceptual framework of factors influencing child survival in developing countries (17). These variables were adapted to the data available in the merged dataset and were categorized as: country level factors, socioeconomic factors, maternal factors, child factors, environmental factors and health service factors as shown in figure 1. The country level factors were the country of survey which includes Burundi, DRC, Kenya, Rwanda, Tanzania and Uganda. Socioeconomic factors were maternal literacy, place of residence, maternal education, maternal occupation, paternal education, household wealth index and maternal working status. The household wealth index serves as an indicator of the economic status of the household which is consistent with household assets, income and expenditure measures. The household assets and facilities used in calculating this index include television (TV), refrigerator, bicycle, radio, motorcycle, car, electricity, type of toilet facility and type of building materials used in the place of dwelling. The index was represented as a score of household assets via the principle components analysis (PCA) method (18). In the DHS data set, the household wealth index was categorised into five quintiles: poorest, poorer, middle, richer and richest. However, in this analysis, the household wealth index was re-categorised into three groups: the bottom 40% of households were referred to as poor households, the next 40% as middle households and the top 20% as rich households. Maternal factors were maternal body mass index (BMI), desire for last child, use of contraceptives, previous death of baby, maternal smoking status and mothers' age at first birth. Child factors were sex of child, perceived birth size, combined birth order and birth interval. Environmental factors were type of cooking fuel, type of water sources and type of sanitation facility. Type of cooking fuel was categorized into solid fuel and non-solid fuel. Solid fuels included (i) traditional biomass (wood, charcoal, agricultural residues, and dung), (ii) processed biomass (pellets, briquettes); and (iii) other solid fuels (such as coal and lignite). Non-solid fuels included (i) liquid fuels (kerosene, ethanol, or other biofuels), (ii) gaseous fuels (natural gas, liquefied petroleum gas, and biogas), and (iii) electricity (19). Type of water sources and type of sanitation facility were classified based on the World Health Organization (WHO) and United Nations Children’s Fund (UNICEF) Joint Monitoring Program (JMP) guidelines (20). Based on these guidelines, water sources were categorized into (i) improved sources which include piped water in a dwelling, plot or yard, public taps or standpipes, tube wells or boreholes, protected dug wells, protected springs and rainwater collection. (ii) Unimproved sources were unprotected dug well, unprotected spring, cart with small tank/drum, tanker truck, bottled water and surface water (river, dam, lake, pond, canal, irrigation channels and stream). Sanitation facility was categorized into (i) improved facility which include households with flush toilet, ventilated or improved pit latrine, pit latrine with slab and composting toilet). (ii) Unimproved facility included traditional pit toilet, pit latrine without slab or open pit and bucket toilet and open defecation (bush or field for defecation). Health service factors were antenatal Fansidar, antenatal iron-folic acid (IFA) supplementation, tetanus toxoid (TT) injection, antenatal care (ANC) visits, delivery assistance, place of delivery, mode of delivery, postnatal check-up and timing of postnatal check-up. However, postnatal check-up and timing of postnatal check-up were not included in the multilevel analysis because they reported a large number of missing values.

Figure 1.  Conceptual framework for analysing factors associated with neonatal mortality in the African Great Lakes region. Adapted from Mosley and Chen [17].
joghr-03-e2019043-f1


Data analysis

All statistical analyses were conducted using STATA statistical software, version 14.1 (Stata Corporation, College Station, TX, USA) with ‘Svy’ commands to adjust for cluster sampling survey design, weights and standard errors (SEs). The neonatal mortality rate and 95% confidence interval (CI) was estimated using the number of resident newborns dying at less than 28 days of age divided by the number of resident live births multiplied by 1,000.

The Cox proportional hazard regression model was used to independently assess the effect of the study variables on the outcome measure (neonatal mortality) after adjusting for potential confounding variables. A pooled analysis was conducted to increase the number of reported deaths in the region. The multivariable analysis used a six staged conceptual model as shown in figure 1. A stepwise backward elimination process was performed at each stage and factors significantly associated with the study outcome were retained. To avoid any statistical bias, we double checked our results by: (1) entering only potential risk factors with a P-value < 0.20 obtained in the univariable analysis for backward elimination process, (2) testing the backward elimination method by including all of the variables (all potential risk factors), and (3) testing and reporting any collinearity in the final model. The hazard ratios (HRs) and their 95% CIs were used to estimate the adjusted risk of the study variables and those with P < 0.05 were retained in the final model.

RESULTS

Characteristics of study variables

Table 1 shows the number of children and their corresponding percentage in each country, as well as the distribution of the study variables in the African Great Lakes region. A total sample of 56600 children were reported across the region. Burundi, DRC, Kenya, Rwanda, Tanzania and Uganda reported 8776 (16%), 10815 (19%), 14189 (25%), 5963 (11%), 6920 (12%) and 9936 (18%) children and their percentage respectively. A weighted total of 866 most recent neonatal deaths occurred within the 5 years’ period preceding the survey in the region. Of which DRC reported 208 deaths, Kenya reported 191 deaths, while Uganda, Tanzania, Burundi and Rwanda reported 181, 113, 95 and 77 deaths respectively.

Table 1.  Distribution of study variables in the African Great Lakes region (N=56600)
Study variables n %
Country level factors
Country of survey:
Burundi 8776 16
DRC* 10815 19
Kenya 14189 24
Rwanda 5963 11
Tanzania 6920 12
Uganda 9936 18
Socioeconomic factors:
Maternal literacy (N=56540):
Can read 39107 69
Cannot read 17433 31
Place of residence:
Urban 15119 27
Rural 41480 73
Maternal education:
Higher education 130 2
Secondary education 677 11
Primary education 4288 72
No education 869 15
Maternal occupation (49123):
Not working 7812 14
Skilled/Professional 12492 22
Agriculture 28819 51
Missing 7477 13
Paternal education (N=42992):
Secondary or higher 15765 28
Primary 20844 37
No education 6383 11
Missing 13608 24
Rich 13083 23
Middle 22863 40
Poor 20654 37
Maternal working status (N=49128):
Currently working 38610 68
Currently not working 10518 19
Missing 7472 13
Maternal factors
Maternal BMI* (N=29659):
≤18 1737 3
18-25 21201 37
25+ 6720 12
Missing 26941 48
Mother’s desire for pregnancy (N=49146):
Wanted then 30792 54
Wanted later 13758 24
Wanted no more 4597 8
Missing 7454 14
Use of contraceptives (N=49168):
Currently using 18406 33
Currently not using 30762 54
Missing 7432 13
Previous death of baby:
No 41101 73
Yes 15499 27
Maternal smoking status:
Non-smoker 56095 99
Smoker 505 1
Mother’s age at first birth (N=56559):
<20 30167 53
20-24 20812 37
25+ 5579 10
Child factors
Sex of child:
Male 28817 51
Female 27783 49
Perceived birth size (N=48621):
Average 24902 44
Smaller than average 7053 12
Larger than average 16666 29
Missing 7979 15
Combined birth order and birth interval:
First child 12560 22
2nd or 3rd child, > 2 years interval 15857 28
2nd or 3rd child, ≤ 2 years interval 4287 8
4th or more child, > 2 years interval 18849 33
4th or more child, ≤ 2 years interval 5046 9
Environmental factors
Type of cooking fuel (N=55292):
Non-solid fuels 1737 5
Solid fuels 53555 95
Type of water sources (N=55291):
Improved water sources 36770 66
Unimproved water sources 18521 34
Type of sanitation facility (N=55283):
Improved facility 24975 45
Unimproved facility 30308 55
Health service factors
Antenatal Fansidar (N=50293):
Yes 22730 40
No 27563 49
Missing 6307 11
Antenatal IFA* supplementation (N=49097):
Yes 34210 60
No 14887 26
Missing 7503 14
Tetanus Toxoid (TT) vaccine (N=49130):
2+ TT 22274 39
1 TT 15444 27
No TT 11411 20
Missing 7470 14
Antenatal care (ANC) visits (N=56397):
4+ANC 29750 53
1-3 ANC 24549 43
N0 ANC 2098 4
Delivery assistance (N=56563):
Health professional* 39590 70
Non-health professional 16972 30
Place of delivery (N=56559):
Health facility 43964 78
Home 12595 22
Mode of delivery (N=56498):
Non-caesarean section 52155 92
Caesarean section 4342 8

DRC – Democratic Republic of the Congo, IFA – iron folic acid, BMI – body mass index

Doctors, nurses, midwives.



Figure 2 shows neonatal mortality rate per 1000 births for countries in the African Great Lakes region. Three countries reported significantly higher rates than the rest of the region, notably DRC (19; 95% CI=17-21), Uganda (18; 95% CI=16-21) and Tanzania (16; 95% CI=14-19). Compared to DRC, neonatal mortality rate was significantly lower in Burundi (11, 95% CI=9-13), Rwanda (13, 95% CI=11-15) and Kenya (13, 95% CI=11-16).

Figure 2.  Neonatal mortality rate per 1000 live births in African Great Lakes region.
joghr-03-e2019043-f2


Table 2 reports the neonatal mortality rates and their 95% CIs by study variable within each country in the African Great Lakes region. Neonatal mortality rate was significantly higher across all countries for mothers who had a caesarean section, did not have antenatal care visits, and reported previous death of a baby as well as for babies who were perceived as smaller than average size. The rate was also significantly higher among women not using contraceptives in all the countries except Burundi and in mothers who are smokers compared to non-smokers in DRC, Kenya and Tanzania. Likewise, the neonatal mortality rate was significantly higher among male children than their female counterpart in Burundi, Rwanda, Tanzania and Uganda. In DRC, Tanzania and Uganda, neonatal mortality rate was reported to be significantly higher among women that use solid fuels for cooking and among women who drank water from unimproved water sources. Furthermore, neonatal mortality rate was significantly higher in households using unimproved sanitation facility in Kenya and Uganda, and also among mothers who did not have TT injection than among mothers who had 2+ TT vaccines in DRC, Kenya and Uganda. Neonatal mortality rate was significantly higher among mothers who did not have antenatal Fansidar and antenatal IFA supplementation in Tanzania and Uganda and was higher across all countries for mothers who had caesarean section.

Table 2.  Neonatal mortality rates and 95% CI in the African Great Lakes Region
Burundi (N=8776) DRC* (N=10815) Kenya (N=14189) (Rwanda (N=5963) Tanzania (N=6920) Uganda
(N=9936)
Study variables Rate (95% CI) Rate (95% CI) Rate (95% CI) Rate (95% CI) Rate (95% CI) Rate (95% CI)
Socioeconomic factors
Maternal literacy:
Can read 11 (10, 13) 20 (18, 22) 13 (12, 15) 12 (10, 14) 18 (16, 20) 17 (15, 19)
Cannot read 10 (9, 12) 18 (16, 20) 14 (13, 16) 16 (14, 18) 12 (10, 13) 20 (18, 22)
Place of residence:
Urban 16 (14, 18) 16 (14, 18) 14 (12, 16) 16 (14, 18) 23 (20, 25) 14 (12, 15)
Rural 10 (9, 12) 21 (18, 23) 13 (11, 15) 12 (11, 14) 14 (12, 16) 20 (18, 22)
Maternal education:
Higher education 72 (68, 77) 26 (23, 28) 14 (12, 16) 0 (0, 0) 0 (0, 0) 13 (11, 15)
Secondary education 13 (11, 14) 19 (17, 21) 8 (6, 9) 12 (10, 13) 16 (14, 18) 15 (13, 17)
Primary education 11 (10, 13) 20 (18, 22) 15 (14, 17) 14 (12, 16) 19 (17, 21) 19 (17, 21)
No education 9 (8, 11) 16 (14, 18) 18 (16, 20) 12 (10, 13) 8 (7, 10) 22 (20, 24)
Maternal occupation:
Not working 10 (9, 12) 18 (16, 20) 10 (9, 12) 13 (11, 15) 17 (15, 19) 14 (13, 16)
Skilled/Professional 10 (8, 11) 20 (18, 22) 11 (9, 13) 11 (9, 12) 20 (18, 22) 18 (16, 20)
Agriculture 11 (9, 13) 19 (17, 22) 15 (13, 17) 13 (12, 15) 1 (1, 2) 20 (18, 22)
Paternal education:
Secondary or higher 13 (11, 15) 18 (16, 20) 11 (9, 12) 15 (13, 17) 21 (19, 23) 14 (12, 16)
Primary 9 (8, 11) 19 (17, 21) 16 (14, 18) 12 (10, 14) 13 (12, 15) 22 (20, 25)
No education 10 (9, 12) 20 (17, 22) 6 (5, 7) 11 (9, 13) 14 (12, 15) 16 (14, 18)
Wealth index:
Rich 15 (13, 17) 13 (11, 15) 10 (9, 12) 12 (10, 13) 18 (16, 20) 14 (13, 16)
Middle 11 (9, 12) 22 (20, 24) 15 (13, 17) 17 (15, 19) 22 (20, 24) 18 (16, 20)
Poor 10 (8, 11) 20 (18, 22) 15 (14, 17) 9 (7, 10) 13 (11, 15) 21 (19, 23)
Maternal working status:
Currently working 11 (10, 13) 20 (18, 22) 14 (12, 16) 13 (11, 15) 16 (14, 18) 18 (16, 20)
Currently not working 9 (8, 10) 17 (15, 19) 9 (8, 11) 12 (10, 14) 17 (15, 19) 18 (16, 20)
Maternal factors
Maternal BMI*:
≤ 18 5 (4, 6) 18 (16, 20) 12 (10, 13) 14 (12, 16) 13 (11, 15) 7 (6, 8)
18-25 14 (12, 16) 19 (17, 21) 13 (11, 15) 11 (9, 12) 15 (13, 17) 18 (16, 20)
25+ 7 (6, 8) 21 (18, 23) 12 (10, 13) 13 (11, 15) 21 (18, 23) 21 (19, 23)
Mother’s desire for pregnancy:
Wanted then 10 (8, 11) 20 (18, 22) 12 (10, 13) 14 (12, 16) 18 (16, 20) 18 (16, 20)
Wanted later 11 (9, 13) 17 (15, 19) 12 (10, 13) 11 (10, 13) 14 (13, 16) 14 (13, 16)
Wanted no more 16 (14, 18) 21 (19, 24) 18 (16, 21) 12 (10, 13) 11 (10, 13) 29 (26, 32)
Use of contraceptives:
Currently using 9 (7, 10) 12 (10, 14) 8 (6, 9) 6 (5, 8) 9 (8, 11) 12 (10, 14)
Currently not using 12 (10, 13) 21 (19, 23) 19 (17, 21) 19 (17, 21) 21 (19, 24) 22 (20, 24)
Previous death of baby:
No 4 (3, 5) 11 (9, 12) 6 (5, 7) 4 (3, 5) 8 (7, 10) 8 (6, 9)
Yes 31 (29, 34) 39 (36, 42) 42 (39, 45) 44 (41, 48) 35 (32, 37) 36 (33, 39)
Maternal smoking status:
Non-smoker 11 (9, 13) 19 (17, 21) 13 (12, 15) 13 (11, 15) 16 (14, 18) 18 (16, 20)
Smoker 5 (4, 6) 28 (25, 30) 115 (110, 121) 17 (15, 19) 40 (37, 43) 0 (0, 0)
Mothers age at first birth:
<20 12 (10, 14) 18 (16, 20) 12 (11, 14) 14 (13, 16) 17 (15, 18) 19 (17, 21)
20-24 9 (8, 11) 22 (20, 24) 16 (14, 18) 12 (11, 14) 15 (13, 17) 19 (17, 21)
25+ 13 (11, 15) 21 (19, 23) 13 (12, 15) 12 (10, 14) 24 (21, 26) 9 (8, 11)
Child factors
Sex of child:
Male 13 (11, 14) 19 (17, 21) 13 (11, 15) 15 (13, 17) 19 (17, 21) 21 (19, 23)
Female 9 (8, 10) 20 (18, 22) 14 (12, 16) 10 (9, 12) 13 (12, 15) 15 (13, 17)
Perceived birth size:
Average 11 (9, 12) 19 (17, 22) 7 (5, 8) 7 (6, 8) 12 (10, 13) 15 (13, 17)
Smaller than average 13 (12, 15) 31 (28, 33) 24 (21, 26) 33 (30, 36) 50 (47, 53) 20 (18, 22)
Larger than average 10 (8, 11) 16 (14, 18) 16 (14, 17) 11 (9, 12) 16 (14, 18) 19 (17, 21)
Combined birth order and birth interval:
First child 13 (12, 15) 34 (31, 36) 13 (11, 14) 18 (16, 20) 23 (20, 25) 23 (21, 26)
2nd or 3rd child, >2 years interval 10 (8, 11) 16 (14, 18) 10 (9, 12) 8 (6, 9) 14 (12, 16) 11 (9, 12)
2nd or 3rd child, ≤ 2 years interval 8 (7, 10) 10 (8, 11) 9 (7, 10) 13 (11, 14) 8 (7, 10) 16 (14, 18)
4th or more child, > 2 years interval 11 (10, 13) 11 (10, 13) 17 (15, 19) 14 (12, 15) 16 (14, 18) 18 (16, 20)
4th or more child, ≤ 2 years interval 10 (8, 11) 32 (29, 34) 22 (20, 25) 18 (16, 20) 14 (12, 16) 28 (26, 31)
Environmental factors
Type of cooking fuel:
Non-solid fuels 0 (0, 0) 12 (11, 14) 10 (9, 12) 0 (0, 0) 47 (44, 50) 25 (23, 27)
Solid fuels 11 (9, 12) 19 (17, 21) 14 (12, 16) 13 (11, 15) 15 (14, 17) 18 (16, 20)
Type of water sources:
Improved water sources 12 (10, 13) 14 (12, 16) 13 (11, 14) 13 (11, 15) 18 (16, 20) 17 (15, 19)
Unimproved water sources 7 (6, 8) 24 (22, 26) 15 (13, 17) 13 (11, 15) 13 (12, 15) 20 (18, 22)
Type of sanitation facility:
Improved facility 12 (10, 13) 18 (16, 20) 11 (10, 13) 13 (12, 15) 16 (14, 18) 14 (12, 16)
Unimproved facility 10 (9, 12) 20 (18, 22) 16 (14, 18) 11 (9, 13) 16 (14, 18) 20 (18, 22)
Health service factors
Antenatal Fansidar:
Yes 11 (10, 13) 18 (16, 20) 10 (9, 12) 0 (0, 0) 17 (15, 19) 15 (14, 17)
No 11 (9, 12) 20 (17, 22) 15 (13, 16) 0 (0, 0) 14 (13, 16) 28 (26, 31)
Antenatal IFA* supplementation
Yes 10 (8, 12) 18 (16, 20) 11 (10, 13) 13 (11, 15) 17 (15, 19) 16 (14, 18)
No 12 (10, 13) 21 (19, 23) 14 (12, 16) 12 (11, 14) 12 (10, 14) 33 (30, 36)
Tetanus Toxoid (TT) injection:
2+ TT 13 (11, 14) 16 (15, 18) 10 (9, 12) 14 (12, 16) 14 (12, 16) 15 (13, 17)
1 TT 9 (7, 10) 17 (15, 19) 12 (11, 14) 12 (10, 13) 21 (19, 23) 21 (19, 23)
No TT 11 (9, 13) 26 (24, 29) 21 (19, 23) 14 (12, 16) 17 (15, 19) 27 (25, 30)
Antenatal care (ANC) visits:
4+ANC 11 (9, 13) 19 (16, 21) 11 (9, 12) 9 (8, 11) 15 (13, 17) 14 (12, 16)
1-3 ANC 11 (9, 12) 16 (14, 18) 15 (13, 17) 16 (14, 18) 17 (15, 19) 24 (21, 26)
N0 ANC 17 (15, 19) 34 (32, 37) 43 (40, 46) 0 (0, 0) 29 (27, 32) 37 (34, 40)
Delivery assistance:
Health professional* 11 (9, 12) 19 (17, 21) 13 (11, 15) 13 (11, 15) 23 (20, 25) 17 (15, 19)
Others 12 (10, 13) 20 (17, 22) 14 (12, 16) 11 (9, 12) 15 (13, 17) 21 (19, 23)
Place of delivery:
Health facility 11 (9, 12) 19 (17, 22) 13 (12, 15) 13 (11, 15) 18 (16, 21) 18 (16, 20)
Home 12 (10, 14) 18 (16, 20) 13 (11, 15) 11 (9, 12) 12 (10, 14) 20 (18, 22)
Mode of delivery:
Non-caesarean section 11 (9, 12) 18 (16, 20) 13 (11, 14) 11 (10, 13) 15 (13, 17) 17 (15, 19)
Caesarean section 19 (17, 21) 30 (27, 33) 21 (19, 23) 25 (23, 28) 30 (28, 33) 35 (32, 38)

DRC – Democratic Republic of the Congo, IFA – iron folic acid, BMI – body mass index

*Doctors, Nurses, midwives.



Factors associated with neonatal mortality

Table 3 shows the unadjusted and adjusted Hazard Ratios (HR) and 95% CI for factors associated with neonatal mortality in the African Great Lakes region. The risk of NMR was significantly higher in DRC compared to Burundi. However, it is important to note that the risk of NMR in Tanzania relative to Burundi approached statistical significance (P=0.066) in being higher.

Table 3.  Unadjusted and adjusted Hazard Ratios (HR) and 95% CI for factors associated with neonatal mortality in the African Great Lakes region
Study Variables Unadjusted Adjusted
Country level factors
Country of survey: HR (95% CI) P-value HR (95% CI) P-value
Burundi 1.00 1.00
DRC 1.77 (1.32, 2.37) <0.001 1.41 (1.04, 1.92) 0.027
Rwanda 1.19 (0.85, 1.66) 0.316 1.05 (0.74, 1.47) 0.792
Tanzania 1.51 (1.09, 2.08) 0.012 1.37 (0.98, 1.92) 0.066
Uganda 1.68 (1.27, 2.23) <0.001 1.26 (0.94, 1.71) 0.127
Socioeconomic factors
Maternal literacy:
Can read 1.00
Cannot read 0.98 (0.81, 1.18) 0.826
Place of residence:
Urban 1.00
Rural 0.93 (0.76, 1.14) 0.493
Maternal education:
Higher education 1.00
Secondary education 1.03 (0.50, 2.10) 0.946
Primary education 1.05 (0.52, 2.10) 0.893
No education 0.74 (0.36, 1.52) 0.410
Maternal occupation:
Not working 1.00
Skilled/Professional 1.12 (0.83, 1.52) 0.455
Agriculture 0.99 (0.75, 1.30) 0.921
Wealth index:
Rich 1.00
Middle 1.23 (0.95, 1.59) 0.117
Poor 1.02 (0.79, 1.32) 0.873
Maternal working status:
Currently working 1.00
Currently not working 1.00 (0.80, 1.25) 0.976
Maternal factors
Mother’s desire for pregnancy:
Wanted then 1.00
Wanted later 0.86 (0.70, 1.07) 0.176
Wanted no more 1.19 (0.88, 1.59) 0.259
Use of contraceptives:
Currently using 1.00
Currently not using 1.99 (1.59, 2.48) <0.001 1.81 (1.43, 2.29) <0.001
Previous death of baby:
No 1.00
Yes 5.21 (4.25, 6.38) <0.001 10.3 (7.75, 13.68) <0.001
Maternal smoking status:
Non-smoker 1.00
Smoker 0.79 (0.29, 2.14) 0.640
Mother’s age at first birth:
<20 1.00
20-24 0.92 (0.75, 1.12) 0.398
25+ 0.88 (0.64, 1.21) 0.424
Child factors
Sex of child
Male 1.00
Female 0.81 (0.67, 0.97) 0.024 0.79 (0.65, 0.96) 0.018
Perceived birth size:
Average 1.00
Smaller than average 1.95 (1.55, 2.47) <0.001 1.88 (1.47, 2.39) <0.001
Larger than average 1.08 (0.87, 1.33) 0.509 1.05 (0.84, 1.33) 0.655
Combined birth order and birth interval:
-2nd or 3rd child, > 2 years interval 1.00
-First child 1.96 (1.51, 2.56) <0.001 3.05 (2.28, 4.07) <0.001
-2nd or 3rd child, ≤ 2 years interval 0.97 (0.65, 1.44) 0.864 0.81 (0.53, 1.23) 0.323
-4th or more child, > 2 years interval 1.19 (0.91, 1.55) 0.196 0.48 (0.36, 0.66) <0.001
-4th or more child, ≤ 2 years interval 2.00 (1.45, 2.77) <0.001 0.79 (0.55, 1.14) 0.213
Environmental factors
Type of cooking fuel:
Non-solid fuels 1.00
Solid fuels 0.61 (0.29, 1.31) 0.206
Type of water sources
Improved water sources 1.00
Unimproved water sources 1.23 (1.02, 1.48) 0.034
Type of sanitation facility:
Improved facility 1.00
Unimproved facility 1.16 (0.96, 1.40) 0.130
Health service factors
Antenatal IFA supplementation:
Yes 1.00
No 1.10(0.90, 1.34) 0.366
Tetanus Toxoid (TT) injection
2+ TT 1.00
1 TT 1.05 (0.84, 1.31) 0.685 1.27 (1.00, 1.61) 0.046
No TT 1.25 (1.00, 1.56) 0.052 1.36 (1.04, 1.78) 0.023
Antenatal care (ANC) visits:
4+ANC 1.00
1-3 ANC 1.17 (0.97, 1.42) 0.103 1.13 (0.92, 1.38) 0.262
N0 ANC 2.34 (1.62, 3.40) <0.001 1.62 (1.06, 2.48) 0.027
Delivery assistance:
Health professional* 1.00
Others 1.07 (0.88, 1.29) 0.513
Place of delivery:
Health facility 1.00
Home 1.00 (0.80, 1.25) 0.977
Mode of delivery:
Non-caesarean section 1.00
Caesarean section 1.92 (1.48, 2.48) <0.001 1.89 (1.40, 2.55) <0.001

DRC – Democratic Republic of the Congo, IFA – iron folic acid

*Doctors, nurses, midwives.



In the multivariate analysis, children born in DRC (HR=1.41, 95% CI=1.04-1.92) were more susceptible to neonatal deaths than those born in other countries within the region. Mothers not using contraceptives (HR=1.81, 95% CI=1.43-2.29) were more prone to having neonatal deaths compared with mothers who were using contraceptives. Mothers who reported previous death of a baby (HR=10.3, 95% CI=7.75-13.68) were more likely to have neonatal deaths. Male children (HR=0.79, 95% CI=0.65-0.96) were more predisposed to neonatal deaths than their female counterpart. Newborns that were perceived as smaller than average were more susceptible to neonatal death (HR=1.88, 95% CI=1.47-2.39). Women who reported having the first child had higher risk of neonatal mortality (HR=3.05, 95% CI=2.28-4.07); whereas, women with 4th or more child with >2 years birth interval had a significantly lower risk of neonatal mortality (HR=0.48, 95% CI=0.36-0.66). Mothers who did not receive any TT injections (HR=1.36, 95% CI=1.04-1.78) and mothers who received one dose of TT injection (HR=1.27, 95% CI=1.00-1.61) were more susceptible to neonatal death than mothers who received 2+ doses of TT injection. Mothers who had no antenatal care visits during pregnancy were more likely to have neonatal deaths HR=1.62, 95% CI=1.06-2.48) compared to mothers who had 4+ antenatal care visits. Women who delivered by caesarean section (HR=1.89, 95% CI=1.40-2.55) were more predisposed to neonatal death than women who had a non-caesarean section delivery.

DISCUSSION

This study examined the rates of, and predisposing factors associated with neonatal mortality across the African Great Lakes region. DRC reported the highest rate of neonatal mortality in the sub-region. DRC is one of 25 priority countries for United States Agency for International Development’s (USAID) work in improving child health as a result of the high rates of newborn deaths. (2). DRC has suffered from war and prolonged conflicts which led to massive disruption to the political, social and economic structure of the country thereby negatively affecting the logistics of food distribution, sanitation, and the delivery of and access to health services (21). USAID has stated that a scale-up of known and proven high-impact interventions could save the lives of more than 1.5 million children by 2020 (2). However, this can only be achieved by strengthening the health system and its utilization in the country.

We found that neonatal mortality rate was significantly higher among male newborn compared to female newborns. Female neonates have a biological advantage in survival over their male counterparts. This could be due to slower lung maturation among male fetuses which leads to a higher incidence of respiratory distress syndrome in male newborns compared with female newborns (22). Globally, it is estimated that approximately 23% of newborn deaths are attributed to respiratory problems (23). Studies have shown that male newborns are more vulnerable to congenital anomalies, infectious diseases such as intestinal infections and lower respiratory infections, as well as perinatal conditions such as birth trauma, fetal macrosomia, intrauterine hypoxia, birth asphyxia, gestational diabetes, prematurity, respiratory distress syndrome and neonatal tetanus (24-27). However, the underlying mechanisms contributing to the observed male disadvantage are still unknown.

In our study, babies who were perceived as being small or as having low birthweight (LBW) by their mothers recorded higher neonatal mortality than larger babies. Studies have shown that LBW contributes to 60% to 80% of all neonatal deaths and it is inversely proportional to neonatal mortality (1, 28). LBW primarily results from premature birth (being born before 37 weeks gestation) or from intrauterine growth restriction (IUGR) (baby not growing well in utero) (28, 29). Prematurity and its consequences contribute to an estimated 50% of neonatal deaths (30). Preterm newborns have greater susceptibility to respiratory distress, jaundice, infections, temperature instability, kernicterus, hypoglycemia, seizures, necrotizing enterocolitis, and periventricular leukomalacia (31). Prevention of preterm births is one of the best ways to prevent babies born with low birthweight. However, in communities were preterm births and low-birth-weight are evident, home-based neonatal care could provide essential postnatal care to mothers as a complementary strategy to facility-based postnatal care. Home-based neonatal care enables continuous coverage of care and improves newborn survival especially for mothers who are unable to access it due to financial, social and physical barriers (32). This care could include home visits in the first month after birth with a focus on promoting essential newborn care practices such as early initiation of breastfeeding and exclusive breastfeeding, screening of high-risk neonates, promoting of hygienic umbilical cord and skin care, management of sepsis, referral of sick infants to health facilities as well as providing of information to mothers on danger sign recognition and skin-to-skin thermal care. In addition, WHO and UNICEF recommends home visits in the first week of life to improve newborn survival (32).

We found that the first child has an increased likelihood of neonatal mortality. This may be as a result of the first child being born to a young mother due to teenage marriage (33). Based on UNICEF’s 2017 report, the proportion of teenage marriages in Uganda, DRC and Tanzania was 40%, 37% and 31% respectively, while in Kenya, Burundi and Rwanda it was 23%, 20% and 7% respectively (33). The mother’s age at birth is important for the quality of decisions she makes. Studies have shown that young maternal age is associated with many health-related complications which arise not only from biological and medical causes but also from underlying sociocultural influences such as lack of education, limited autonomy to make decisions due to family pressure, poor health-seeking behaviours as well as non-usage of available health services and infrequent routine antenatal and postnatal check-ups (34-36). On the other hand, this study also reported that children of 4th or higher birth order with long birth intervals (>2 years) were more prone to neonatal mortality. Children of 4th or higher birth order with long birth intervals (>2 years) are likely to be born to older women and research has shown that advanced maternal age is significantly associated with pregnancy complications such as preterm birth which is a direct cause of neonatal mortality (30, 31, 37).

Findings from this study show that a mother’s report of a previous death of baby was significantly associated with a neonatal death. Studies have shown that the probabilities of neonatal death among second children were highest among mothers whose first child also died during the neonatal period (38, 39). Birth spacing and maternal age at time of birth are important determinants for the survival of newborns (38). Hence, healthy timing and spacing of pregnancy are effective low-cost prevention approach for reducing neonatal mortality especially in the African Great Lakes region (40, 41).

The use of contraceptives could have important effects on child survival. In this study, women not using contraceptives were reported to be more predisposed to neonatal mortality. Studies have shown that non-usage of contraceptives exposes women to unintended pregnancies which lead to poor health-seeking behaviours and negligence (34). The positive impact of contraceptive use is evident when it assists reproductive age women prevent closely spaced births. Strategies to reduce neonatal mortality rates by empowering women to control the timing of their births using modern contraceptive methods have great health benefits for mother and neonate (42, 43). By improving access to and use of family planning methods such as contraceptives use, women can reduce closely spaced births and limit childbearing to their 20s and 30s, thereby reducing their chances of neonatal mortality (42). Therefore, there is a need for governments and non-governmental organizations that are concerned about the level of neonatal mortality to promote and fund family planning services and contraceptive supplies especially in communities of low socioeconomic status.

This study reported a higher likelihood of neonatal mortality among mothers who delivered by caesarean section compared with vaginal deliveries. This could be attributed to the effect of life-threatening pregnancy complications which warrant for emergency caesarean section or iatrogenic prematurity (44). However, contrary to our findings is a cross-sectional, ecological study which reported caesarean delivery rates were inversely correlated with neonatal mortality (45), and another ecological study which also reported that caesarean section available for high-risk pregnancies could contribute to improve neonatal outcomes (46).

Reproductive health care is the care given to a woman across her reproductive life related to reproduction from pre-conception to pregnancy, delivery, and post-delivery. This care is important for the survival and well-being of the mother and her child and encompasses the health care dimensions of family planning and prenatal, natal, and postnatal care with the aim of reducing maternal and child mortality (47).

In our study, failure to utilize available reproductive health care services such as ANC visits, TT injection during pregnancy and contraceptives resulted in a higher susceptibility to neonatal mortality. The main purpose of ANC is to ensure optimal health outcomes for the mother and her baby and its content service package as outlined in the WHO guidelines, which include screening for complications, health education for healthy lifestyle, 2 TT injections and 90 IFA tablets (48). The causes of neonatal deaths could be minimized or prevented through ANC interventions and WHO has recommended four or more ANC visits – the first two during the first two trimesters and the last two visits during the last trimester (48). Mothers who fail to attend the recommended visits are at a higher risk of poor birth outcomes. Furthermore, a study also identified that one of the most effective ANC intervention was TT injection given to women during pregnancy to prevent infant deaths due to neonatal tetanus, a leading cause of neonatal death in developing countries where a high proportion of deliveries take place at home or in places where hygienic conditions may be poor (49). To improve antenatal care service delivery, outreach programmes by skilled health personnel to administer ANC services such as check-ups and TT injection could effectively and significantly reduce neonatal mortality (50).

Strengths and limitations

This study has some strength. First, the dataset used in this study were obtained from nationally representative household surveys and the variables are similarly defined in all surveys, hence comparable across all countries. Second, this study is population-based with a high response rate (>90%) and data were pooled together to create a large sample size of neonatal deaths reported within 5 years preceding the surveys. Third, this study applied appropriate statistical adjustments to data obtained from 5 nationally representative surveys and was able to identify the significant factors associated with neonatal mortality across the region to inform policy prioritization.

Despite these merits, this study also has some limitations. First, due to the cross-sectional nature of the study design, this paper is unable to establish a causal relationship between neonatal mortality and the independent variables. Second, the newborns’ dates of birth and death were from retrospective data based on self-report from mothers and this could be a potential source of recall and misclassification bias. Third, the rationale used by the mothers in estimating the size of their babies is unclear; hence, caution should be taken in interpreting the relationship between the mother’s perceived birth size and neonatal mortality. Finally, the antecedent health and nutritional status of newborns and their mothers, especially for those newborns that had died, and causes of death were not reported in the DHS surveys.

This study is useful in identifying the most significant underlying factors associated with neonatal mortality across the African Great Lakes region to assist in formulating effective preventive and curative interventions that addresses these factors among the most vulnerable sub-populations. This is particularly important as the region is recovering from wars, and our findings should inform development initiatives addressing maternal and child health to reduce socioeconomic inequalities in neonatal mortality. Given the large proportion of under-five deaths occurring during the neonatal period, it is imperative that evidence-based cost-effective interventions aimed at improving child survival during the neonatal period are implemented alongside the development of a coherent neonatal health policy by each country across the African Great Lakes region.

CONCLUSIONS

This study shows that a complex set of factors increase the risk of neonatal mortality: sex of child (male), perceived birth size (smaller than average), birth order and interval (first child and 4th or more child with >2 years birth interval), previous death of baby, non-usage of contraceptives, non-receipt of TT injection, no antenatal care visits and delivery through caesarean section. Hence, interventions to prevent neonatal mortality should focus on developing a coherent neonatal health policy that increases access to affordable health care as well as improves the quality of health care services rendered to women and their newborn. Additionally, cost-effective interventions aimed at increasing maternal awareness on the importance of healthy birth spacing, proper nutrition and health services utilization are needed across the African Great Lakes region. This intervention strategies will align the region with the Sustainable Development Goals (SDGs) target of reducing neonatal mortality to as low as 12 deaths per 1,000 live births by 2030.

Acknowledgements

The authors are grateful to Measure DHS, ICF International, Rockville, Maryland, USA for granting access to the datasets used in this analysis.

Ethics approval: The first author communicated with MEASURE DHS/ICF International, Rockville, Maryland, USA and was granted permission to download and use the DHS datasets for Burundi, Congo DR, Kenya, Rwanda, Tanzania and Uganda. Hence, no ethics approval was needed.

Notes

[1] Financial disclosure Funding: No grant was received for this study from any funding agency in the public, commercial or not-for-profit sectors.

[2] Conflicts of interest Competing interests: The authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare no conflict of interest.

REFERENCES

1UNICEF. Every Child Alive: The urgent need to end newborn deaths. 2018. Available: https://www.unicef.org/publications/index_102640.html. Accessed: 25 June 2019.

2Hug L, Sharrow D, You D. Levels & trends in child mortality: report 2017. Estimates developed by the UN Inter-agency Group for Child Mortality Estimation.

3N Golding, R Burstein, J Longbottom, AJ Browne, N Fullman, and A Osgood-Zimmerman. Mapping under-5 and neonatal mortality in Africa, 2000–15: a baseline analysis for the Sustainable Development Goals. Lancet. 2017;390:p.2171–82. DOI: 10.1016/S0140-6736(17)31758-0. [28958464]

4H Wang, CA Liddell, MM Coates, MD Mooney, CE Levitz, and AE Schumacher. Global, regional, and national levels of neonatal, infant, and under-5 mortality during 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384(974):p.957–79. DOI: 10.1016/S0140-6736(14)60497-9. [24797572]

5C Engmann. . Improving neonatal mortality in sub-Saharan Africa: any cause for optimism? J Perinatol. 2011;31:p.745–8. DOI: 10.1038/jp.2011.53. [22124515]

6BM Afolabi. . Sub-Sahara African Neonates-Ghost to Statistics. J Neonatal Biol. 2017;6:p.2167–0897. DOI: 10.4172/2167-0897.1000246

7GA Kayode, D Grobbee, M Amoakoh-Coleman, E Ansah, O Uthman, and K Klipstein-Grobusch. . Variation in neonatal mortality and its relation to country characteristics in sub-saharan africa. BMJ Glob Health. 2017;2 Suppl 2:e000209. DOI: 10.1136/bmjgh-2016-000260.130. [29104766]

8World Health Organization. WHO recommendations on postnatal care of the mother and newborn. WHO. 2014. Available from: https://apps.who.int/iris/bitstream/handle/10665/97603/9789241506649_eng.pdf?sequence=1. Accessed: 25 June 2019.

9A Conde‐Agudelo and JL Díaz‐Rossello. . Kangaroo mother care to reduce morbidity and mortality in low birthweight infants. Cochrane Database Syst Rev. 2011;3:CD002771. DOI: 10.1002/14651858.CD002771.pub2. [21412879]

10WA Carlo and CP Travers. . Maternal and neonatal mortality: time to act. J Pediatr (Rio J). 2016;92:p.543–5. DOI: 10.1016/j.jped.2016.08.001. [27521567]

11The DHS Program. Available from: http://dhsprogram.com/data/data-collection.cfm. Accessed: 25 June 2019

12Ministry to the Presidency in charge of Good Governance and Planning [Burundi] (MPBGP), Ministry of Health Public Health and the Fight against AIDS [Burundi] (MSPLS), Institute of Statistics and Economic Studies of Burundi (ISTEEBU), and ICF. Third Demographic and Health Survey. Bujumbura, Burundi: ISTEEBU, MSPLS, and ICF; 2017.

13Kenya National Bureau of Statistics, Ministry of Health, National AIDS Control Council, Kenya Medical Research Institute, National Council for Population and Development, and The DHS Program, ICF International Rockville, Maryland, USA. Kenya Demographic and Health Survey 2014, Nairobi, Kenya; 2015.

14National Institute of Statistics of Rwanda (NISR) [Rwanda], Ministry of Health (MOH) [Rwanda], and ICF International. Rwanda Demographic and Health Survey 2014-15. Rockville, Maryland, USA: NISR, MOH, and ICF International; 2015.

15Ministry of Health, Community Development, Gender, Elderly and Children (MoHCDGEC) [Tanzania Mainland], Ministry of Health (MoH) [Zanzibar], National Bureau of Statistics (NBS), Office of the Chief Government Statistician (OCGS), and ICF. Tanzania Demographic and Health Survey and Malaria Indicator Survey (TDHS-MIS) 2015-16. Dar es Salaam, Tanzania, and Rockville, Maryland, USA: MoHCDGEC, MoH, NBS, OCGS, and ICF; 2016.

16Uganda Bureau of Statistics (UBOS) and ICF. Uganda Demographic and Health Survey 2016. Kampala, Uganda and Rockville, Maryland, USA: UBOS and ICF; 2018.

17WH Mosley and LC Chen. . An analytical framework for the study of child survival in developing countries. Bull World Health Organ. 2003;81:p.140–5.[12756980]

18D Filmer and LH Pritchett. . Estimating wealth effects without expenditure data—or tears: An application to educational enrolments in states of India. Demography. 2001;38:p.115–32.[11227840]

19Ghosh Banerjee S, Portale E, Adair-Rohani H, Bonjour S. Tracking Access to Nonsolid Fuel for Cooking. Live Wire, 2014/8. Washington DC: World Bank; 2014.

20World Health Organization, WHO/UNICEF Joint Water Supply, Sanitation Monitoring Programme. Progress on sanitation and drinking water: 2015 update and MDG assessment. Geneva: World Health Organization; 2015.

21EE Lindskog. . The effect of war on infant mortality in the Democratic Republic of Congo. BMC Public Health. 2016;16:p.1059 . DOI: 10.1186/s12889-016-3685-6. [27716146]

22MJ Khoury, JS Marks, BJ McCarthy, and SM Zaro. . Factors affecting the sex differential in neonatal mortality: the role of respiratory distress syndrome. Am J Obstet Gynecol. 1985;151:p.777–82. DOI: 10.1016/0002-9378(85)90518-6. [3976790]

23World Health Organization. The global burden of disease: 2004 update. Geneva: WHO; 2011.

24World Health Organization. Sex differentials in Infant mortality. SEARO. 2019. Available: http://www.searo.who.int/entity/health_situation_trends/data/chi/sex-diff-imr/en/. Accessed: 25 June 2019

25D Zhao, L Zou, X Lei, and Y Zhang. . Gender differences in infant mortality and neonatal morbidity in mixed-gender twins. Sci Rep. 2017;7:p.8736 . DOI: 10.1038/s41598-017-08951-6. [28821800]

26L Hou, X Wang, G Li, L Zou, Y Chen, and W Zhang. . Cross sectional study in China: fetal gender has adverse perinatal outcomes in mainland China. BMC Pregnancy Childbirth. 2014;14:p.372 . DOI: 10.1186/s12884-014-0372-4. [25344636]

27L Aibar, A Puertas, M Valverde, MP Carrillo, and F Montoya. . Fetal sex and perinatal outcomes. J Perinat Med. 2012;40:p.271–6. DOI: 10.1515/jpm-2011-0137. [22505506]

28C Lau, N Ambalavanan, H Chakraborty, MS Wingate, and WA Carlo. . Extremely low birth weight and infant mortality rates in the United States. Pediatrics. 2013;131:p.855–60. DOI: 10.1542/peds.2012-2471. [23545381]

29GR Alexander, M Kogan, D Bader, W Carlo, M Allen, and J Mor. . US birth weight/gestational age-specific neonatal mortality: 1995–1997 rates for whites, Hispanics, and blacks. Pediatrics. 2003;111:p.e61–6. DOI: 10.1542/peds.111.1.e61. [12509596]

30JE Lawn, S Cousens, and J Zupan. . 4 million neonatal deaths: When? Where? Why? Lancet. 2005;365:p.891–900. DOI: 10.1016/S0140-6736(05)71048-5. [15752534]

31S Saigal and LW Doyle. . An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet. 2008;371:p.261–9. DOI: 10.1016/S0140-6736(08)60136-1. [18207020]

32WHO/UNICEF Joint Statement. Home Visits for the Newborn Child: A Strategy to Improve Survival. Geneva: World Health Organization. 2009.

33UNICEF. State of the World’s Children 2016: A fair chance for every child. UNICEF. 2016. Available from: https://www.unicef.org/publications/files/UNICEF_SOWC_2016.pdf. Accessed: 25 June 2019.

34P Mukhopadhyay, RN Chaudhuri, and B Paul. . Hospital-based perinatal outcomes and complications in teenage pregnancy in India. J Health Popul Nutr. 2010;28:p.494 . DOI: 10.3329/jhpn.v28i5.6158. [20941901]

35AK Sharma, K Verma, S Khatri, and AT Kannan. . Pregnancy in adolescents: a study of risks and outcome in Eastern Nepal. Indian Pediatr. 2001;38:p.1405–9.[11752740]

36MS Chahande, AR Jadhao, SK Wadhva, and S Ughade. . Study of some epidemiological factors in teenage pregnancy hospital based case comparison study. Indian J Community Med. 2002;27:p.4 .

37J Yerushalmy. . Neonatal mortality by order of birth and age of parents. Am J Hyg. 1938;28:p.244–70.[12313287]

38MM Rahman, M Kabir, and R Amin. . Relationship between survival status of first child and subsequent child death. J Biosoc Sci. 1996;28:p.185–91. DOI: 10.1017/S0021932000022239. [8935875]

39EJ Clegg and JF Cross. . Aspects of neonatal death in St Kilda, 1830–1930. J Biosoc Sci. 1994;26:p.97–106. DOI: 10.1017/S002193200002109X. [8200884]

40M Norton. . New evidence on birth spacing: promising findings for improving newborn, infant, child, and maternal health. Int J Gynaecol Obstet. 2005;89 Suppl 1 DOI: 10.1016/j.ijgo.2004.12.012. [15820364]

41EK Williams, MB Hossain, RK Sharma, V Kumar, CM Pandey, and AH Baqui. . Birth interval and risk of stillbirth or neonatal death: findings from rural north India. J Trop Pediatr. 2008;54:p.321–7. DOI: 10.1093/tropej/fmn027. [18443009]

42. . Family Planning Can Reduce High Infant Mortality Levels. Issues Brief (Alan Guttmacher Inst). 2002 Apr.; (2):p.1 .[12134892]

43L Chola, S McGee, A Tugendhaft, E Buchmann, and K Hofman. . Scaling up family planning to reduce maternal and child mortality: the potential costs and benefits of modern contraceptive use in South Africa. PLoS One. 2015;10:e0130077. DOI: 10.1371/journal.pone.0130077. [26076482]

44RH Xie, L Gaudet, D Krewski, ID Graham, MC Walker, and SW Wen. . Higher cesarean delivery rates are associated with higher infant mortality rates in industrialized countries. Birth. 2015;42:p.62–9. DOI: 10.1111/birt.12153. [25597509]

45G Molina, TG Weiser, SR Lipsitz, MM Esquivel, T Uribe-Leitz, and T Azad. Relationship between cesarean delivery rate and maternal and neonatal mortality. JAMA. 2015;314:p.2263–70. DOI: 10.1001/jama.2015.15553. [26624825]

46F Althabe, C Sosa, JM Belizán, L Gibbons, F Jacquerioz, and E Bergel. . Cesarean section rates and maternal and neonatal mortality in low‐, medium‐, and high‐income countries: an ecological study. Birth. 2006;33:p.270–7. DOI: 10.1111/j.1523-536X.2006.00118.x. [17150064]

47World Health Organization. Maternal Health. WHO. 2019. Available from: https://www.who.int/maternal-health/en/. Accessed: 25 June 2019.

48Lawn J, Kerber K. Opportunities for Africa’s newborns: practical data policy and programmatic support for newborn care in Africa. Cape Town, South Africa: Partnership for Maternal, Newborn and Child Health [PMNCH]; 2006.

49RJ McCurdy, KH Kjerulff, and J Zhu. . Kjerulff KH, Zhu J. Prenatal care associated with reduction of neonatal mortality in Sub‐Saharan Africa: evidence from Demographic and Health Surveys. Acta Obstet Gynecol Scand. 2011;90:p.779–90. DOI: 10.1111/j.1600-0412.2011.01133.x. [21426311]

50M Arunda, A Emmelin, and BO Asamoah. . Effectiveness of antenatal care services in reducing neonatal mortality in Kenya: analysis of national survey data. Glob Health Action. 2017;10:1328796. DOI: 10.1080/16549716.2017.1328796. [28621201]