Genus

Bitew et al. Genus (2020) 76:37

https://doi.org/10.1186/s41118-020-00106-2

ORIGINAL ARTICLE Open Access

Machine learning approach for predicting

under-five mortality determinants in
Ethiopia: evidence from the 2016 Ethiopian
Demographic and Health Survey
Fikrewold H. Bitew1*, Samuel H. Nyarko1,2, Lloyd Potter1,2 and Corey S. Sparks1

* Correspondence: fikre.wold@gmail.
com Abstract
1
Department of Demography,
College for Health, Community & There is a dearth of literature on the use of machine learning models to predict
Policy, University of Texas at San important under-five mortality risks in Ethiopia. In this study, we showed spatial
Antonio, 501 W. Cesar Chavez Blvd, variations of under-five mortality and used machine learning models to predict its
San Antonio, TX 78207, USA
Full list of author information is important sociodemographic determinants in Ethiopia. The study data were drawn
available at the end of the article from the 2016 Ethiopian Demographic and Health Survey. We used three machine
learning models such as random forests, logistic regression, and K-nearest neighbors
as well as one traditional logistic regression model to predict under-five mortality
determinants. For each machine learning model, measures of model accuracy and
receiver operating characteristic curves were used to evaluate the predictive power
of each model. The descriptive results show that there are considerable regional
variations in under-five mortality rates in Ethiopia. The under-five mortality prediction
ability was found to be between 46.3 and 67.2% for the models considered, with the
random forest model (67.2%) showing the best performance. The best predictive
model shows that household size, time to the source of water, breastfeeding status,
number of births in the preceding 5 years, sex of a child, birth intervals, antenatal
care, birth order, type of water source, and mother’s body mass index play an
important role in under-five mortality levels in Ethiopia. The random forest machine
learning model produces a better predictive power for estimating under-five
mortality risk factors and may help to improve policy decision-making in this regard.
Childhood survival chances can be improved considerably by using these important
factors to inform relevant policies.
Keywords: Machine learning, Under-five mortality, Determinants, Ethiopia

Introduction
Globally, an estimated 5.4 million children under the age of 5 are said to have died in
2017 alone (UNICEF, WHO, World Bank Group, and United Nations, 2018). Mean-
while, the global under-five mortality rate is said to have declined by 58%, from 93
deaths per 1000 live births in 1990 to 39 in 2017 (UNICEF, WHO, World Bank Group,
and United Nations, 2018). Yet still, the under-five mortality rate in low-income
© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit
line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by
statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Bitew et al. Genus (2020) 76:37 Page 2 of 16

countries was 69 deaths per 1000 live births in 2017—almost 14 times the rate in
high-income countries (5 deaths per 1000 live births) (UNICEF, WHO, World
Bank Group, and United Nations, 2018). It has been observed that more than half
of these deaths are due to infectious diseases (such as pneumonia and diarrhea)
that are preventable and treatable through simple, affordable interventions (World
Health Organization, 2017).
Despite the considerable improvements over the past decades, sub-Saharan Africa re-
mains the region with the highest level of under-five mortality in the world, with about
half of the global under-five mortality burden (UNICEF, WHO, World Bank Group,
and United Nations, 2018). Ethiopia appears to have the fifth-highest number of new-
born deaths in the world, following India, Pakistan, Nigeria, and the Democratic Re-
public of Congo (UNICEF, 2017). It is estimated that about 472,000 children die in
Ethiopia each year before their fifth birthday, which places Ethiopia sixth among the
countries in the world in terms of absolute numbers of under-five deaths (Federal Min-
istry of Health, 2005). In Ethiopia, the under-five mortality rate has declined by two-
thirds from the 1990 figure of 204 per 1000 live births to 58 per 1000 live births in
2016, and thus, achieving the target for Millennium Development Goal 4 (MDG 4)
(You, Hug, Ejdemyr, Idele, et al., 2015). Despite this achievement, the under-five mor-
tality rate in Ethiopia remains higher than those of many low and middle-income coun-
tries (LMIC).
Previous studies have provided much evidence on the socioeconomic and demographic
factors that are associated with under-five mortality in Ethiopia (Ayele & Zewotir, 2016;
Ayele, Zewotir, & Mwambi, 2017; Bereka, Habtewold, & Nebi, 2017), using traditional re-
gression models. In this study, we predict the important determinants of under-five mor-
tality in Ethiopia using non-traditional regression models drawing on nationally
representative data. Specifically, we employed machine learning techniques to predict
under-five mortality risks in the study sample. The main aim of this study is to show the
spatial distribution of under-five mortality and the potential of machine learning algo-
rithms in predicting important sociodemographic factors underlying the spatial variations
in under-five mortality. As such, we initially develop a spatial visualization of the under-
five mortality rate by region in Ethiopia. This is to visually highlight the spatial disparities
in under-five mortality in the country while predicting the most important factors under-
lying these disparities. This study informs and strengthens appropriate extant policies or
intervention strategies aimed at reducing under-five mortality in the country. It also un-
derscores the potential role of the machine learning approach in demographic research.

Methods
Data source
This study is based on data from the 2016 Ethiopian Demographic and Health Survey
(EDHS), the most recent in the demographic and health survey series that is conducted
every five years. The EDHS is a nationally representative household survey that collects
data on a wide range of population, health, and nutrition indicators to improve mater-
nal and child health in Ethiopia (Central Statistical Agency (CSA) [Ethiopia],, and ICF
International, 2016). The survey used a multi-stage stratified sampling technique based
on the 2007 National Population and Housing Census of Ethiopia to select respondents
Bitew et al. Genus (2020) 76:37 Page 3 of 16

from a total of 624 clusters (187 urban and 437 rural) (Central Statistical Agency (CSA)
[Ethiopia],, and ICF International, 2016). The unit of analysis is under-five children
with a total sample size of 10,641 selected from 645 clusters across Ethiopia. This is
based on children’s data obtained from retrospective information from mothers about
their children that died before age 5 within the 5 years preceding the survey (2011 to
2016).

Study variables and measurements

In this study, the outcome variable—under-five mortality—was measured as a binary
outcome. Thus, under-five mortality was measured as being alive (coded as 0) or dead
(coded as 1) for all the models.
The predictors (features) used in this study include individual, household, commu-
nity, and health service factors. The individual-level factors consisted of maternal and
child characteristics. Maternal factors include mother’s age at birth (< 20, > 20), educa-
tion (no education, primary, secondary/higher), contraceptive use (yes/no), and
mother’s body mass index (BMI) (underweight/overweight and normal). Child factors
include whether the child was wanted (child wanted then, wanted later, not at all), sex
of the child, birth order (1–2, 3/later), births in last 5 years, and previous birth interval
(< 2, 2–4, > 4 years), as well as whether the child was breastfed within 1 h of birth. The
household factors used are the source of drinking water (improved/unimproved), time
to the water source, toilet facility (improved/unimproved), and household wealth index
(low, middle, high), and household size. The community factors comprised residence
type (urban/rural) and geographical region (Tigray, Afar, Amhara, Oromia, Somali,
Benishangul-Gumuz, Southern Nations Nationalities and People Region (SNNPR),
Gambella, Harari, Dire Dawa, and Addis Ababa). The health service factors include
antenatal visits (0, 1–4, 5+ visits), place/mode of delivery services (facility with caesar-
ean section (CS) services, facility without CS, home), and postnatal visits within 2
months after delivery (yes/no). The selection of these predictor variables was based on
information from existing literature on the subject (Aheto, 2019; Bereka et al., 2017;
Yaya, Bishwajit, Okonofua, & Uthman, 2018).

Analytic strategy
The R programming language (version 3.6.0) and the caret package (Kuhn, 2020) was
used to perform the data processing and analysis. We first developed a spatial map for
crude under-five mortality rates by regions in Ethiopia to document the regional dis-
parities in under-five mortality in the country. In this regard, we estimated the rates
under-five mortality by region and then merged them with an Ethiopian regional sha-
pefile before mapping it.
We also used three widely used machine learning (ML) algorithms—logistic regres-
sion, a Random Forest (RF), K-nearest neighbors (KNN) models—to predict under-five
mortality determinants in Ethiopia and compared the results of the best algorithm to
the results of the traditional logistic regression model. These three models were se-
lected for various reasons. The logistic regression is typically used to analyze binary
data and is commonly used as an inferential tool in population health research, but can
be also used as a binary classification model. The KNN model is chosen based on its
Bitew et al. Genus (2020) 76:37 Page 4 of 16

ability to detect linear and nonlinear boundaries between groups. The K is a value that
represents the number of nearest neighbors which is the core deciding factor in this
classifier. It relies on finding the best value of k so that the k closest observations are
used to predict the value of a given observation. Thus, when k = 1 then the new data
object is simply assigned to the class of its nearest neighbor. The “nearness” of observa-
tions is widely measured using Euclidean distance between observations even though
there are various numerical measures (Ali, Neagu, & Trundle, 2019; Larose, 2015). The
main concept behind KNN depends on calculating the distances between the tested,
and the trained data samples to identify its nearest neighbors. The RF model is com-
monly used in machine learning situations because it is highly flexible and provides
better predictive performance. The RF model repeatedly samples the variables in the
training data set several times, each time using a random set of predictor variables to
produce decision trees. After many of these trees are formed, the forest is examined to
see which variable consistently produce a better prediction. These groups of relatively
uncorrelated models can produce ensemble predictions that are more accurate than
any of the individual predictions. This is because the trees protect each other from their
errors (as long as they do not all constantly err in the same direction).
ML was born from pattern recognition and the theory that computers can learn with-
out being programmed to perform specific tasks. It allows computers to learn from
complex data sources, to potentially find previously unseen insights without being ex-
plicitly programmed where to look (Elisa, 2018). It can also be used to automate tasks
by building analytical models using algorithms that iteratively learn from data. In
demographic parlance, ML appears to address some of the major challenges in demo-
graphic research by helping to draw insights using available datasets collected for differ-
ent purposes at different points in time, which in most cases may be challenging to
incorporate in the traditional techniques. It may be also used to predict future occur-
rences of the principal components of population change (fertility, mortality, and mi-
gration) and associated factors. As such, ML techniques can be both used to predict
previously identified proximate correlates and new “significant” demographic variables,
and also shed light on how important previously used variables are in terms of
prediction.
In this regard, we randomly sampled and trained 80% of the total sample, which was
eventually used for 10-fold cross-validation to tune the model parameters. The
remaining 20% random sample was used as test data to predict the measures of model
performance. Because the outcome is unbalanced (there is a low fraction of under-five
mortality in the data), the data were down-sampled so the proportions of data in the
training set are equivalent to the cases who were alive after 5 years, and those who had
died before 5 years. Model accuracy metrics such as sensitivity, specificity, positive pre-
dictive value, and negative predictive values were calculated to show how well the
models perform in terms of predicting the dead and alive cases. Sensitivity (“positivity
in health”) refers to the proportion of subjects who have dead cases (reference standard
positive) and give positive test results. Specificity (“negativity in health”) is the propor-
tion of subjects that are alive (reference standard negative) and give negative test re-
sults. Positive predictive value is the proportion of positive results that are true
positives (i.e., truly dead) whereas negative predictive value is the proportion of negative
results that are true negatives (i.e., truly alive). Predictive values vary depending on the
Bitew et al. Genus (2020) 76:37 Page 5 of 16

prevalence of the target condition in the population being studied, even if the sensitivity
and specificity remain the same (Price & Christenson, 2007).
Metrics such as the area under curve (AUC) and receiver operating characteristic (ROC)
curve were also used to evaluate model performance in distinguishing between the dead
and alive cases. The ROC curves compare sensitivity versus specificity across a range of
values to determine the ability to predict a dichotomous outcome. The AUC is a measure
of the ability of a classifier to distinguish between classes and is used as a summary of the
ROC curve (Florkowski, 2008). Thus, the higher the AUC, the better the performance of
the model at distinguishing between the positive and negative classes (Florkowski, 2008).
The results of all the models were weighted using person weights provided by the
data. For the traditional logistic regression model, we infer the importance and signifi-
cance of predictors using odds ratios and confidence intervals derived from the model
estimation, while for the ML models, the Mean Decrease in Gini was calculated for
each variable, which is a measure of variable importance for these models. The top 10
categories of variables based on their Mean Decrease in Gini were automatically gener-
ated and then presented in diagrams for each ML model.

Results
Descriptive results of the background characteristics
Table 1 shows the results of under-five mortality by the sample characteristics. Of the 10,
641 under-five children in the sample, there appears to be a significant difference in mortal-
ity prevalence between both sexes with female children experiencing higher (6.7%) than
males (4.2%). There were also considerable differences by birth intervals with under-five
mortality being more prevalent among children with less than 2 years of birth intervals
(9.3%) than children with 2–4 and over 4 years of birth intervals (4.45% and 4.53%, respect-
ively). Under-five mortality was also significantly prevalent among children using unim-
proved water sources (5.8%) than those who used improved water sources (2.9%).
Significant differences were also observed regarding antenatal visits and postnatal care, with
under-five mortality being considerably prevalent among children whose mothers did not
receive antenatal (5.6%) and postnatal care (4.2%). Children who were breastfed within more
than 1 h of birth had a significantly higher prevalence of death (9.8%) than those breastfed
within 1 h of birth (4.5%) while there was also evidence of a significant difference in under-
five mortality regarding the number of people in the household. The rest of the characteris-
tics did not show any significant difference in mortality prevalence among their categories.
Spatial distribution of under-five mortality
Figure 1 shows the spatial distribution of crude under-five mortality rates by regions
in Ethiopia. The under-five mortality rate in the map is presented as the number of
under-five deaths per 1000 live births. The Afar region recorded the highest under-five
mortality rate of 125 per 1000 live births, followed by Benshangul–Gumuz, and Somali,
which recorded 98 and 94 per 1000 live births, respectively. The lowest under-five mor-
tality rate is recorded in Addis Ababa, with a rate of 39 per 1000 live births.

Predicting under-five mortality

Below, we report the results of the three machine learning models (logistic regression,
random Forests, and the K-nearest neighbor models) (Table 2). The under-five
Bitew et al. Genus (2020) 76:37 Page 6 of 16

Table 1 Descriptive statistics of child mortality outcome by study characteristics, EDHS 2016 (N =
10,641)
Characteristics Child alive before Child dead before Chi-square test of
age 5 age 5 equality
Percent/mean Percent/mean
Child dead/alive 94.9 5.1
Child sex p = .0001
Male 95.8 4.2
Female 93.3 6.7
Birth order p = 0.71
1st or 2nd 94.7 5.3
3rd or higher 94.4 5.6
Birth interval p = .0001
< 2 years 90.7 9.3
2–4 years 95.5 4.45
> 4 years 95.5 4.53
Mothers age at first birth p = 0.47
< 20 years 94.3 5.7
> 20 years 94.9 5.1
Age of the mother p = 0.94
15–19 94.9 5.0
20–34 94.4 5.6
35–49 94.6 5.4
Residence p = 0.28
Rural 94.4 5.6
Urban 95.7 4.3
Education p = 0.34
No education 94.1 5.9
Primary 95.1 4.8
Secondary and Higher 95.5 4.5
Wealth index p = 0.63
Low 94.7 5.3
Middle 94.7 5.3
High 94.1 5.9
Water source p = 0.51
Unimproved 94.3 5.73
Improved 94.8 5.23
Time to water source 167.4 164.6 p = 0.89*
Toilet facility p = 0.005
Unimproved 94.2 5.8
Improved 97.1 2.9
Place and mode of delivery services p = 0.07
Fac with CS delivery 96.1 3.9
Fac without CS delivery 94.1 5.9
Home 94.2 5.8
Contraceptive use p = 0.23
Yes 95.1 4.9
Bitew et al. Genus (2020) 76:37 Page 7 of 16

Table 1 Descriptive statistics of child mortality outcome by study characteristics, EDHS 2016 (N =
10,641) (Continued)
Characteristics Child alive before Child dead before Chi-square test of
age 5 age 5 equality
Percent/mean Percent/mean
No 94.2 5.8
Child wanted p = 0.74
Then 94.6 5.4
Later 94.5 5.5
Not at all 93.6 6.4
Antenatal visits p = 0.002
No visit 94.4 5.6
1–4 visits 96.7 3.3
5+ visits 97.6 2.4
Postnatal care visits p = 0.009
No 95.8 4.2
Yes 98.3 1.7
Region p = 0.43
Oromia 94.2 5.8
Addis Ababa 95.9 4.1
Afar 91.5 8.5
Amhara 94.9 5.1
Ben-Gumuz 94.2 5.8
Dire Dawa 93.7 6.3
Gambella 93.3 6.7
Harari 94.5 5.5
SNNP 93.3 6.7
Somali 96.7 3.3
Tigray 93.8 6.2
Mother’s BMI p = 0.41
Underweight 93.7 6.3
Normal 94.6 5.4
Overweight 95.4 4.6
Breastfed p = .0001
Within an hour of birth 95.5 4.5
Greater than an hour of birth 90.2 9.8
Household size 6.1 5.4 p = .0001*
NB all estimates include sample design and person weights, per DHS instructions. *t test was used instead of chi-square.
Significant variables are in bold

mortality prediction accuracy was found to be low for all models, between 46.3 and
67.2% accuracy on the test dataset, with the RF model having the highest overall accur-
acy. The RF model had high sensitivity, meaning that it was more accurate in identify-
ing dead cases, but had low specificity, meaning that it was poor in identifying the alive
cases. However, the model correctly identified 70% of the real dead cases (28/(28+12))
and 67% of real alive cases (698/(698+343)), which means that the RF model is rela-
tively better at predicting both real positive (dead) and negative (alive) cases. The
Bitew et al. Genus (2020) 76:37 Page 8 of 16

Tigray

Afar

Amhara

Benishangul

Dire Dawa
Harari
Addis Adaba

Gambela
Oromia

Somali

SNNPR

U5 Mortality Rate
30 – 40
40 – 50
50 – 60
60 – 70
70 – 80
80 – 90
90 – 100
100 – 110
110 – 130

Leaflet | © OpenStreetMap contributors, © OpenStreetMap contributors © CARTO

Fig. 1 Spatial distribution of crude under-five mortality rates by regions in Ethiopia.Source: Created by the
authors from EDHS estimates

logistic and KNN models both show lower overall accuracy (59.9 and 46.3%, respect-
ively), and lower sensitivity, specificity, and positive as well as negative predictive
values.
A visualization of the receiver operating characteristics (ROC) curve is shown in Fig. 2.
Among the three machine learning models employed in this study, the curve of the RF
model shows the highest AUC value, indicating it is the best at classifying dead and alive
cases, among the models.
Figures 3, 4, and 5 show the variable importance measures, measured by the scaled
mean decrease in the Gini coefficient for each variable, as calculated during the k-fold
cross-validation process. This is an effective measure of how important a variable is for
predicting under-five mortality across all the cross-validation estimates. The three

Table 2 Model accuracy metrics for all models as evaluated on the test data
Confusion matrix Random Forest Logistic regression KNN model
Predicted Predicted Predicted
Alive Dead Alive Dead Alive Dead
Observed Alive 698 343 632 418 475 566
Dead 12 28 16 24 14 26
% % %
Accuracy 67.2 59.9 46.3
Sensitivity 98.3 97.5 97.1
Specificity 7.5 5.34 4.4
Positive predictive value 70.0 59.9 45.6
Negative predictive value 67.0 60.0 65.0
AUC 72.0 66.1 55.5
Bitew et al. Genus (2020) 76:37 Page 9 of 16

Fig. 2 ROC Curves for the three models

figures show very similar results, with household size (nhh) and breastfeeding behavior
(bfeed) being among the top 3 variables in all three models. Other important factors
that appeared in the top five variables are the time to water source (time_water), num-
ber of births (births5_ys), birth interval (b_interval), and child sex (male).
Unlike the ML model results presented above, the traditional logistic regression
model is the only one that allows direct interpretation of the model coefficients
(Table 3). Table 3 shows the estimated odds ratios and confidence intervals for the
model parameters. Factors associated with under-five mortality were sex, birth
order, birth interval, water source, place of delivery, antenatal visit, postnatal care,
breastfeeding, and household size. Increased risks of under-five mortality were
found among males, higher birth order children, and children born in a facility
without C-section services. On the contrary, reduced risks were found among chil-
dren with longer birth intervals, improved water sources, children who received
antenatal and postnatal care as well as those from larger households.

Fig. 3 Variable importance measures for the random Forest model

Bitew et al. Genus (2020) 76:37 Page 10 of 16

Fig. 4 Variable importance measures for logistic regression model

Discussion
This study briefly described spatial variations in under-five mortality and predicted
under-five mortality risks in Ethiopia using machine learning techniques. The spatial
map provides evidence of considerable regional disparities in under-five mortality rates
in Ethiopia similar to what has been observed in Ghana (Aheto, 2019). Tigray and some
regions in the central part of the country show the lowest under-five mortality rates
whereas regions in the eastern and western parts of the country have the highest
under-five mortality rates. Providing evidence on the spatial variations of under-five
mortality in the country may provide the need to better understand the underlying risk
factors. Regarding the predictive analysis, the prediction accuracies and AUC statistics
are found to be highest for the RF model. The RF model shows a higher predictive
power compared to the other ML models included in this study. In this regard, the RF
model shows that household size, time to the water source, breastfeeding behavior,
births in the preceding 5 years, sex of a child, birth intervals, birth order, antenatal

Fig. 5 Variable importance measures for K-nearest neighbor model

Bitew et al. Genus (2020) 76:37 Page 11 of 16

Table 3 Logistic regression analysis of under-five mortality in Ethiopia

Variables Odds ratio Lower 95 % CI Upper 95% CI p value
(Intercept) 0.033 0.006 0.193 0.0001
Mothers age first birth (Ref: < 20)
> 20 0.600 0.353 1.018 0.059
Sex (Ref: female)
Male 2.018 1.398 2.913 0.0001
Birth order (Ref: 1st/2nd)
3rd or higher 2.129 1.131 4.008 0.020
Birth interval (Ref: < 2)
2–4 years 0.527 0.309 0.898 0.019
> 4 years 0.385 0.190 0.779 0.008
Time to water source 1.000 0.999 1.000 0.244
Water source (Ref: unimproved)
Improved 0.585 0.348 0.985 0.044
Toilet facility (Ref: improved)
Unimproved 1.713 0.744 3.943 0.206
Births in last 5 years 1.163 0.744 1.816 0.508
Residence (Ref: rural)
Urban 0.527 0.181 1.541 0.243
Mother’s education (Ref: no education)
Primary 0.928 0.513 1.680 0.805
Secondary/higher 1.856 0.480 7.178 0.370
Wealth index (Ref: low)
Middle 1.342 0.698 2.581 0.378
High 1.694 0.937 3.064 0.082
Contraceptive use (Ref: using)
Not using 1.174 0.735 1.876 0.502
Region
Addis Ababa 1.124 0.485 2.605 0.786
Afar 0.573 0.228 1.435 0.235
Amhara 0.885 0.354 2.211 0.794
Ben-Gumuz 1.494 0.587 3.803 0.400
Dire Dawa 1.021 0.408 2.554 0.965
Gambella 0.623 0.243 1.597 0.325
Harari 1.175 0.495 2.790 0.715
SNNP 1.221 0.376 3.960 0.740
Somali 1.504 0.287 7.881 0.629
Tigray 1.733 0.519 5.787 0.372
Mother’s BMI (Ref: normal)
Overweight 0.527 0.170 1.640 0.269
Underweight 1.402 0.868 2.264 0.168
Place of delivery (Ref: fac with CS delivery)
Facility without CS delivery 2.850 1.182 6.869 0.020
Home 1.185 0.617 2.275 0.610
Bitew et al. Genus (2020) 76:37 Page 12 of 16

Table 3 Logistic regression analysis of under-five mortality in Ethiopia (Continued)

Variables Odds ratio Lower 95 % CI Upper 95% CI p value
Antenatal visits (Ref: no visit)
1–4 visits 0.616 0.381 0.995 0.048
5+ visits 0.437 0.208 0.917 0.029
Postnatal care (Ref: no)
Yes 0.264 0.080 0.872 0.029
Child wanted (Ref: wanted then)
Wanted later 0.768 0.369 1.599 0.482
Not at all 1.407 0.749 2.642 0.289
Breastfeeding (Ref: > an hour of birth)
Within 1 h of birth 0.242 0.147 0.398 0.0001
vHousehold size 0.498 0.345 0.719 0.0001
NB significant variables are in bold

visits, type of water source, and mother’s BMI are the top 10 important predictors of
under-five mortality in Ethiopia. The important role played by some of these factors in
under-five mortality levels is widely documented in the literature (Abir, Agho, Page,
Milton, & Dibley, 2015; Dendup, Zhao, & Dema, 2018; Howell, Holla, & Waidmann,
2016; Yaya et al., 2018).
In comparison, the findings of the best performing ML model appear to be virtually
consistent with the traditional logistic regression analysis which also shows that a
child’s sex, birth interval, birth order, water source, place of delivery, antenatal visits,
postnatal care, household size, and breastfeeding behavior play a significant role in
under-five mortality levels in Ethiopia. Only the number of births in the preceding
5 years and the mother’s BMI appear to play an important role in the ML models
but play an insignificant role in the traditional logistic regression analysis. This is
an indication that ML models may produce some “new variables” or previously un-
seen insights by the traditional regression models which may play a crucial role in
policy decision making. From the traditional logistic regression findings, male chil-
dren have shown a significantly higher risk of dying before age 5 compared with
female children. This is consistent with the finding of a cross-sectional study con-
ducted in Bangladesh (Abir et al., 2015). It has been shown that male children
have an increased risk of dying in the first month of life because of high vulner-
ability to infectious disease. This may be because female neonates are more likely
to develop early fetal lung maturity in the first week of life, which may result in a
lower incidence of respiratory diseases in female compared with male neonates
(Khoury, Marks, McCarthy, & Zaro, 1985). Also, higher birth order of children ap-
pears to be associated with a significantly higher risk of under-five mortality.
Analogously, the unfavorable effect of higher birth order on childhood survival
chances has been well documented in Africa (Howell et al., 2016) as well as some
parts of Asia (Dendup et al., 2018; Hong & Hor, 2013) and may provide a better
understanding of the spatial variations in the country.
Furthermore, the risk of under-five mortality has increased significantly among
children with less than 2 years of birth interval than children with more than 2
years of birth interval. Affirmatively, there is much evidence that longer birth
Bitew et al. Genus (2020) 76:37 Page 13 of 16

intervals improve the survival chance of succeeding children (Kozuki & Walker,
2013; Yaya et al., 2018). A short preceding birth interval can be said to influence
under-five mortality through three main mechanisms: first, closely spaced births
may cause depletion of the mother. The second mechanism is through competition
for scarce household resources among children, while the third is the transmission
of infectious diseases between the closely spaced children (Majumder, May, & Pant,
1997). While the first mechanism is biological, the last two are said to be behav-
ioral effects of a short preceding birth interval (Koenig, Phillips, Campbell, &
Dsouza, 1990).
Additionally, this study finds that the use of unimproved drinking water is associated
with an increased risk of under-five mortality. Lack of access to clean water has been
considered as one of the important factors that contribute to more than 80% of child
deaths in the world (UNICEF, 2018). There is also considerable evidence from studies
in developing countries that show that household sanitation and a clean water supply
promote child health and survival (Ezeh, Agho, Dibley, Hall, & Page, 2014; Mugo,
Agho, Zwi, Damundu, & Dibley, 2018). In Ethiopia, the proportion of the population
using improved drinking water sources is only 57%, and those who use improved sani-
tation are less than 5% (World Health Organization, 2017). This may have serious im-
plications for variations in under-five mortality in the country. This study further
provides evidence that children whose mothers do not use any contraceptives have a
significantly higher risk of under-five mortality than their counterparts whose mothers
use modern contraceptives.
This study also finds that delivery in health facilities without CS services and at home
is associated with a higher under-five mortality risk. This may be mainly related to
dealing with delivery complications that may raise under-five mortality risks. Health fa-
cilities with CS services are very scarce in Ethiopia, and where they are available, trans-
portation challenges encourage women to deliver at home even when facility-based
delivery is available at a minimal cost (Shiferaw, Spigt, Godefrooij, Melkamu, & Tekie,
2013). Moreover, the study finds a positive effect of antenatal and postnatal care
checkups on under-five survival chances. This is consistent with the significant associ-
ation observed between antenatal and postnatal care and lower under-five mortality
risk in the literature (Bitew & Nyarko, 2019; Machio, 2018). The implication is that
children whose mothers do not receive antenatal and postnatal care services may ex-
perience several proximate under-five mortality risk factors, such as congenital and in-
fectious diseases, than their counterparts. This study has also shown a considerable
positive effect of early timing of breastfeeding on childhood survival chances. Breast-
feeding has long been shown as an important protective factor against under-five mor-
tality, particularly among developing countries (Azuine, Murray, Alsafi, & Singh, 2015;
Nyarko, Tanle, & Kumi-Kyereme, 2014) and may play a key part in childhood survival
interventions in Ethiopia. Quite surprisingly, larger household size appears to be associ-
ated with reduced under-five mortality risk in this study, contrary to what is docu-
mented in the literature (Dendup et al., 2018). However, this may well be underscored
by some household-level contextual factors in the country such as availability of consid-
erable social support from parents and siblings.
This study is not without limitations. The survey comprised only surviving women,
and since neonatal and maternal mortalities may occur concurrently, this may have led
Bitew et al. Genus (2020) 76:37 Page 14 of 16

to an underestimation of the under-five mortality rates. Ultimately, unlike the trad-

itional regression models, the ML results appear to be mostly uninterpretable because
they have no regression coefficients and for that matter no direction of effect. In effect,
ML models generally predict or classify specific variables based on the level of import-
ance of their role in determining the under-five mortality levels in the current study. In
this case, extant empirical literature from studies using the traditional methodologies
may be used to determine the direction of these important variables. There are also
possible biases in the memorization or non-disclosure of deaths by mothers which may
underestimate the number of deaths. Nevertheless, machine learning techniques are
considered to be very useful in predicting population health and other phenomena and
lead to better policy decisions (Ashrafian & Darzi, 2018; Holzinger, 2017).

Conclusions
The findings show that considerable regional disparities in under-five mortality rates
persist in Ethiopia, with the highest rates being found in the Afar, Benishangul—
Gumuz, and Somali regions. Also, the RF model provides a moderately better predictive
power than the logistic regression and KNN ML models in predicting under-five mor-
tality determinants in Ethiopia. Even though the RF model and the traditional logistic
regression model have shown similar factors, the RF model appears to reveal some im-
portant factors that may not be identified by the traditional logistic regression model.
This model may, therefore, proffer better policy directions regarding under-five child-
hood survival. Thus, household size, time to the water source, breastfeeding behavior,
number of births in the past 5 years, sex of a child, birth intervals, antenatal visits, birth
order, type of water source, and mother’s BMI may play an important role in under-
five survival chances in Ethiopia. This study highlights the use of machine learning al-
gorithms to predict and better understand very important under-five mortality risk fac-
tors to improve crucial policy directions. As a corollary, ML methods may also apply to
other areas of demographic research including fertility and migration studies. Our find-
ings reinforce the need to focus on the most important predicted factors including
breastfeeding, birth interval control, and antenatal care among others in developing
policies aimed at enhancing childhood survival chances. Also, based on the findings,
expanding access to improved drinking water will help to substantially reduce future
under-five mortality levels in Ethiopia.

Abbreviations
AUC: Area under curve; BMI: Body mass index; CS: Caesarean section; EDHS: Ethiopian Demographic and Health
Survey; KNN: K-Nearest Neighbors; LMIC: Low- and middle-income countries; MDG: Millennium Development Goal;
ML: Machine learning; RF: Random forest; ROC: Receiver operating characteristic; SNNPR: Southern Nations Nationalities
and People Region

Authors’ contributions
FHB conceived and designed the study. FHB and CSS performed the analysis with technical support from SHN. FHB
wrote the initial draft of the manuscript with technical support from SHN, LP, and CSS. All authors critically reviewed
the manuscript for important intellectual content and then approved the final version of the manuscript for
publication.

Funding
No funding was received for this study

Availability of data and materials

The datasets analyzed in this study are freely available at the DHS Program repository
Bitew et al. Genus (2020) 76:37 Page 15 of 16

Competing interests
The authors declare that they have no competing interests.

Author details
1
Department of Demography, College for Health, Community & Policy, University of Texas at San Antonio, 501 W.
Cesar Chavez Blvd, San Antonio, TX 78207, USA. 2Institute for Demographic and Socioeconomic Research, The
University of Texas at San Antonio, 501 W. Cesar Chavez Blvd, San Antonio, TX 78207, USA.

Received: 30 April 2020 Accepted: 16 October 2020

References
Abir, T., Agho, K. E., Page, A. N., Milton, A. H., & Dibley, M. J. (2015). Risk factors for under-five mortality: evidence from
Bangladesh Demographic and Health Survey, 2004–2011. BMJ Open, 5(8), e006722.
Aheto, J. M. K. (2019). Predictive model and determinants of under-five child mortality: evidence from the 2014 Ghana
Demographic and Health Survey. BMC Public Health, 19, 64.
Ali, N., Neagu, D., & Trundle, P. (2019). Evaluation of k-nearest neighbour classifier performance for heterogeneous data sets.
SN Applied Sciences, 1(12), 1559.
Ashrafian, H., & Darzi, A. (2018). Transforming health policy through machine learning. PLoS Medicine, 15(11), e1002692.
Ayele, D. G., & Zewotir, T. T. (2016). Childhood mortality spatial distribution in Ethiopia. Journal of Applied Statistics, 43(15),
2813–2828.
Ayele, D. G., Zewotir, T. T., & Mwambi, H. (2017). Survival analysis of under-five mortality using Cox and frailty models in
Ethiopia. Journal of Health, Population, & Nutrition, 36(1), 25.
Azuine, R. E., Murray, J., Alsafi, N., & Singh, G. K. (2015). Exclusive breastfeeding and under-five mortality, 2006-2014: A cross-
national analysis of 57 low- and-middle income countries. International Journal of MCH AIDS, 4(1), 13–21.
Bereka, S. G., Habtewold, F. G., & Nebi, T. D. (2017). Under-five mortality of children and its determinants in Ethiopian Somali
Regional State, Eastern Ethiopia. Health Science Journal, 11, 3.
Bitew, F., & Nyarko, S. H. (2019). Modern contraceptive use and intention to use: implication for under-five mortality in
Ethiopia. Heliyon, 5, e02295.
Central Statistical Agency (CSA) [Ethiopia], & ICF International (2016). Ethiopia Demographic and Health Survey 2016. Addis
Ababa, Ethiopia, Calverton, MD, USA: Central Statistical Agency, ICF International.
Dendup, T., Zhao, Y., & Dema, D. (2018). Factors associated with under-five mortality in Bhutan: an analysis of the Bhutan
National Health Survey 2012. BMC Public Health, 18, 1375.
Elisa, N. (2018). Could Machine Learning be used to address Africa's Challenges? International Journal of Computer
Applications, 180(18), 0975–8887.
Ezeh, O. K., Agho, K. E., Dibley, M. J., Hall, J., & Page, A. N. (2014). The impact of water and sanitation on childhood mortality in
Nigeria: evidence from demographic and health surveys, 2003–2013. International Journal of Environmental Research and
Public Health, 11(9), 9256–9272.
Federal Ministry of Health (2005). National Strategy for Child Survival in Ethiopia. Addis Ababa: Federal Ministry of Health.
Florkowski, C. M. (2008). Sensitivity, specificity, receiver-operating characteristic (ROC) curves, and likelihood ratios:
communicating the performance of diagnostic tests. The Clinical Biochemist Reviews, 29(Suppl 1), S83.
Holzinger, A. (2017). Introduction to machine learning and knowledge extraction (MAKE). Machine Learning and Knowledge
Extraction, 1(1), 1–20.
Hong, R., & Hor, D. (2013). Factors associated with the decline of under-five mortality in Cambodia, 2000-2010: Further analysis of
the Cambodia Demographic and Health Surveys. Calverton: ICF International.s.
Howell, E. M., Holla, N., & Waidmann, T. (2016). Being the younger child in a large African family: a study of birth order as a
risk factor for poor health using the demographic and health surveys for 18 countries. BMC Nutrition, 2, 61.
Khoury, M. J., Marks, J. S., McCarthy, B. J., & Zaro, S. M. (1985). Factors affecting the sex differential in neonatal mortality: the
role of respiratory distress syndrome. American Journal of Obstetrics and Gynecology, 151(6), 777–782.
Koenig, M. A., Phillips, J. F., Campbell, O. M., & Dsouza, S. (1990). Birth intervals and childhood mortality in rural Bangladesh.
Demography, 27(2), 251–265.
Kozuki, N., & Walker, N. (2013). Exploring the association between short/long preceding birth intervals and child mortality:
using reference birth interval children of the same mother as comparison. BMC Public Health, 13, S6.
Kuhn, M. (2020). Caret: Classification and Regression Training. R package version, 6, 0–85 https://CRAN.R-project.org/package=
caret.
Larose, D. T. (2015). Data mining and predictive analytics. New York: Wiley.
Machio, P. M. (2018). Determinants of neonatal and under-five mortality in Kenya: Do antenatal and skilled delivery care
services matter? Journal of African Development, 20(1), 59–67.
Majumder, A. K., May, M., & Pant, P. D. (1997). Infant and child mortality determinants in Bangladesh: Are they changing?
Journal of Biosocial Science, 29(4), 385–399.
Mugo, N. S., Agho, K. E., Zwi, A. B., Damundu, E. Y., & Dibley, M. J. (2018). Determinants of neonatal, infant, and under-five
mortality in a war-affected country: analysis of the 2010 Household Health Survey in South Sudan. BMJ Global Health,
3(1), e000510.
Nyarko, S. H., Tanle, A., & Kumi-Kyereme, A. (2014). Determinants of childhood mortality in Ghana. International Journal of
Social Science Research, 3, 61–77.
Price, C. P., & Christenson, R. H. (2007). Evidence-based laboratory medicine: principles, practice, and outcomes, (2nd ed., ).
Washington DC: American Association for Clinical Chemistry Press.
Shiferaw, S., Spigt, M., Godefrooij, M., Melkamu, Y., & Tekie, M. (2013). Why do women prefer home births in Ethiopia? BMC
Pregnancy and Childbirth, 13, 5.
UNICEF. (2017). The State of the World’s Children. https://www.unicef.org/sowc/. Accessed March 15, 2019.
UNICEF (2018). Every Child Alive. The urgent need to end newborn deaths. Genèva, Switzerland: UNICEF.
Bitew et al. Genus (2020) 76:37 Page 16 of 16

UNICEF, WHO, World Bank Group & United Nations (2018). Levels and trends in child mortality report 2018. New York: UNICEF.
World Health Organization (2017). World health statistics 2017: Monitoring health for the SDGs, and Sustainable Development
Goals. Geneva: WHO.
Yaya, S., Bishwajit, G., Okonofua, F., & Uthman, O. A. (2018). Under five mortality patterns and associated maternal risk factors
in sub-Saharan Africa: A multi-country analysis. PLoS ONE, 13(10), e0205977.
You, D., Hug, L., Ejdemyr, S., Idele, P., et al. (2015). Global, regional, and national levels and trends in under-five mortality
between 1990 and 2015, with scenario-based projections to 2030: a systematic analysis by the UN Inter-agency Group
for Child Mortality Estimation. Lancet, 386(10010), 2275–2286.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.