Stunting
Stunting is a syndrome that is linked to poor nutrition and recurrent infections. These infections
include parasitic infections, particularly malaria and intestinal helminths, and diarrhea,
especially in conditions of poor sanitation and hygiene (Millward, 2017; Vonaesch et al., 2018).
Identified as a cyclical process, growth faltering (i.e., stunting) often starts in utero, connects
to maternal nutrition, to an intergenerational cycle of growth failure and conveys to generations
through the mother (de Onis & Branca, 2016; Millward, 2017). This can be aggravated by the
environment in which a child lives. Although the physiopathology of stunting is not well
understood, authors 24 have theorized that pediatric environmental enteropathy (PEE), a
chronic inflammation of the small intestine could play a major role (Vonaesch et al., 2018).
They argued that PEE is closely linked to stunting and is characterized by the predominance of
the inflammatory process dominated by villous atrophy, crypt hyperplasia, increased
permeability, inflammatory cell infiltrate, and modest malabsorption caused by fecal bacteria
ingested in large quantities by young children living in conditions of poor sanitation, and
hygiene is crucial in the understanding of childhood stunting (Millward, 2017; Vonaesch et al.,
2018). In their study Brown et al. (2015) concluded that PEE/stunting in the first two years of
life was characterized by enteric infections caused by Shigella and enterotoxins. Long-term
consequences related to childhood stunting syndrome are varied and of massive severity. These
include increased morbidity and mortality; reduced physical, neurodevelopmental, and
economic capacity, and an elevated risk of metabolic disease into adulthood (de Onis & Branca,
2016). Researchers have documented poor school uptake among stunted children and a high
risk of mortality and susceptibility to infections (Tariku et al., 2017). Stunting has important
economic consequences for both sexes at the individual, household, and community level.
Authors have shown an association between shorter adult stature and labor-market outcomes
such as lower earnings and poorer productivity (de Onis & Branca, 2016). Increases in
morbidity, mortality, and health expenditure and subsequent reductions in human capital
investment, physical capital investment, and labor supply are all various pathways through
which undernutrition could affect aggregate economic growth (McGovern et al., 2017). Ill
health leads to reductions 25 in productivity and could negatively affect individuals’ and
society’s economic growth, especially when a significant proportion of a country’s population
is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The key study
variables identified as underlying causes for stunting are discussed in terms of the study
population’s economic resources, social determinants of health, and environmental conditions.
Parental Employment Parental employment status or the occupational standing of the
household head has been associated with child's nutrition status (stunting). Rashad and Sharaf
(2019) conducted a study to estimate the causal impact of women’s work on child nutritional
status as measured by the height-for-age score. The authors found a negative correlation
between under-5 nutritional status and their mother’s employment and reduced odds of stunting
for children whose parents (households’ heads) were farmers or self-employed relative to
households headed by housewives. Rashad and Sharaf (2019) argued that it is because couples’
income combined could procure better services than those gained by a single parent (mothers).
Parental Education Parental education has been linked to childhood mortality through different
pathways and is often associated with the socioeconomic status of the household. Frost et al.
(2005) mentioned five pathways through which the linear relationship between parental
education and mortality among under-five children occur: (a) socioeconomic 26 status, (b)
health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e) reproductive
behaviors. These pathways operate at both individual and community levels. Mosley and Chen
(1984), in their seminal work, observed a strong linear relationship between mother’s
educational achievement and reduction in infant and under-5 mortality in South Saharan
African countries. Likewise, researchers found that high literacy levels among mothers in a
given country of sub-Saharan Africa promoted the utilization of health care facilities and the
lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information
Researchers have documented poor school uptake among stunted children and a high risk of
mortality and susceptibility to infections (Tariku et al., 2017). Stunting has important economic
consequences for both sexes at the individual, household, and community level. Authors have
shown an association between shorter adult stature and labor-market outcomes such as lower
earnings and poorer productivity (de Onis & Branca, 2016). Increases in morbidity, mortality,
and health expenditure and subsequent reductions in human capital investment, physical capital
investment, and labor supply are all various pathways through which undernutrition could
affect aggregate economic growth (McGovern et al., 2017). Ill health leads to reductions 25 in
productivity and could negatively affect individuals’ and society’s economic growth,
especially when a significant proportion of a country’s population is affected by stunting
(McGovern et al., 2017). Underlying Causes of Stunting The key study variables identified as
underlying causes for stunting are discussed in terms of the study population’s economic
resources, social determinants of health, and environmental conditions. Parental Employment
Parental employment status or the occupational standing of the household head has been
associated with child's nutrition status (stunting). Rashad and Sharaf (2019) conducted a study
to estimate the causal impact of women’s work on child nutritional status as measured by the
height-for-age score. The authors found a negative correlation between under-5 nutritional
status and their mother’s employment and reduced odds of stunting for children whose parents
(households’ heads) were farmers or self-employed relative to households headed by
housewives. Rashad and Sharaf (2019) argued that it is because couples’ income combined
could procure better services than those gained by a single parent (mothers). Parental Education
Parental education has been linked to childhood mortality through different pathways and is
often associated with the socioeconomic status of the household. Frost et al. (2005) mentioned
five pathways through which the linear relationship between parental education and mortality
among under-five children occur: (a) socioeconomic 26 status, (b) health knowledge, (c)
attitudes towards health care, (d) female autonomy, and (e) reproductive behaviors. These
pathways operate at both individual and community levels. Mosley and Chen (1984), in their
seminal work, observed a strong linear relationship between mother’s educational achievement
and reduction in infant and under-5 mortality in South Saharan African countries. Likewise,
researchers found that high literacy levels among mothers in a given country of sub-Saharan
Africa promoted the utilization of health care facilities and the lowering of risk for under-5
child mortality (Anyamele et al., 2017; McTavish et al., 2010). Nkurunziza et al. (2017) found
that children whose mothers had no education were more likely to be stunted (cOR = 2.3; 95%
CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI: 1.3-2.9; p < 0.001) relative to
those whose mothers who completed secondary school and above. Likewise, Acharya et al.
(2018), in their analysis of determinants of childhood stunting in Democratic Republic of
Congo, found a negative association between level of education and childhood stunting.
Similarly, Birhanu et al. (2017) found an increased mortality rate for infants and under-5
children whose parents were illiterate as well as a strong association between stunting and
parental education level. The number of stunted children was found to be four times higher
among illiterate mothers as compared to those from educated mothers (Birhanu et al., 2017).
Similarly, the likelihood of childhood stunting from illiterate fathers was eight times higher as
compared to those children who had educated fathers (Birhanu et al., 2017). Even though the
pathways are clearly understood, authors agree that parental education is a key determinant of
childhood nutritional status as it plays a central role by empowering both women and men to
offer 27 adequate childcare in terms of health, child feeding, and child education (Geberselassie
et al., 2018). Researchers noted a protective effect against stunting in children whose fathers
completed secondary school as compared to children whose fathers completed primary school
(Geberselassie et al., 2018). However, this protective effect might be missing in single mother
households. Clark and Hamplová (2013) explored single motherhood and child mortality in
sub-Saharan Africa and found that children born to never-married single mothers were
significantly more likely to die before age five in six countries as compared with children
whose parents were married (odds ratios range from 1.36 in Nigeria to 2.61 in Zimbabwe). In
nine countries, they found children of formerly married mothers were at a significantly higher
risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya) relative to children
having married parents (Clark & Hamplová, 2013). These findings are supported by Härkönen
(2018) who found a universal growing negative educational gradient of single motherhood
across countries. The negative educational gap promotes social inequality among single
mothers and their children who live in a cycle of poverty, and experience health related risks,
and poor educational achievement. Therefore, policy makers, public health organizations and
country governments should empower single mothers with better education and environmental
incentives to help ensure positive child health outcomes both at the community and individual
levels (Frost et al., 2005). Birth Interval 28 The relationship of birth interval and malnutrition
has been linked to maternal education. The general consensus among researchers is that chronic
malnutrition (stunting) is promoted by large sibling numbers and by later birth order.
Researchers in sub-Saharan Africa found boys to be more affected by chronic malnutrition
(stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found that among
mothers who were single, first-born boys were more exposed to childhood mortality and not
the first born girls’ counterpart. They found that sister and not brother played a protective role
towards lowering childhood mortality rate for second- born children. Similarly, Bukusuba et
al. (2017) in their case control study among children aged six and 59 months in Uganda found
that boys were significantly more likely to be stunted compared to girls (OR= 2.2, 95% CI:
1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of childhood stunting
for children born within a preceding birth interval less than 24 months. Equally, Liwin and
Houle (2019) found the risk of dying more than five times higher among children aged 12–23
months in the interval of zero to 11 months between the index child and the succeeding while
for a succeeding birth interval of 0–23 months the risk of dying among children aged 24–59
months increased by two-fold (Liwin & Houle, 2019). Economic Resources and Living
Conditions Linked to education level of parents and opportunity for employment is the
potential for improved household income which also plays a pivotal role in determining child
health outcome. Household income is tied to parental education and child health 29 and impacts
food intake through improved food purchasing power, quality of medication via its access and
affordability, and household sanitation (Umar Farooq et al., 2019). In their study in Cameroon
and DRC, Ntoimo and Odimegwu (2014) found higher probabilities of childhood stunting in
single mother households being significantly affected by economic resources and parental
education. Moreover, Mikalitsa (2015) while examining the intra household allocation,
household headship and nutrition of under-fives in western Kenya argued that better nutrition
status of children was not always linked to high income in a household, but rather to who
controls the income. The author found a linear association between improved childhood
nutritional status in households and active participation of women in decision making
(Mikalitsa, 2015). Households with uneducated parents were inclined to have low-income
levels. These parents spent less on nutritional adequate foods and their children were more
susceptible to growth failure due to lack of access to sufficient food of adequate quality as well
as poor living conditions, lack of access to basic health care services and greater exposure to
diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in their discourse on effect
of remittances on diet and nutrition found that remittances could lead to household food
security and underweight reduction even though remittances were found to play a little effect
on chronic undernourishment. In addition to the economic resources available to a household,
where individuals in a household reside the effect of the environs are important considerations
to stunting outcomes. In particular, is the recognized rural-urban inequality on childhood
stunting and that malnutrition in under-five children has been variably distributed according to
30 geographic regions. Kismul et al. (2018) when analyzing the determinants of stunting in the
Democratic Republic of Congo found a significant variability in the prevalence of stunting
between rural and urban regions with rural areas having a larger percentage of children living
with stunting than children living urban areas. In another study on environmental predictors of
stunting among under-5 children in Somalia, authors found variation in stunting distribution
spatially and temporally due to rains seasons and vegetation (Kinyoki et al., 2016). Similarly,
Stifel et al. (2018) in keeping with previous findings showed that rural populations were
characterized as having worse nutrition outcomes than those of urban populations. However,
they demonstrated that the nutritional inequality that exists between rural-urban regions stems
from differences in wealth, education, health, and non-road infrastructure services across rural
and urban areas (Stifel et al., 2018). Thus, reduction of childhood stunting requires
improvement of parental education, improvement of and access to healthy diets, and
improvement of household’s wealth as these are identified in the literature as the major
predictors of stunting (Headey et al., 2017). A component of the environment is water, but also
relevant to hygiene and sanitation and a key nutrient for health is access to safe water in rural
and urban communities. Childhood stunting has been negatively associated with poor access
to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al. (2014) estimated
that fewer than one in five people globally wash their hands with soap after defecation, Bain et
al. (2014) demonstrated the presence of disparities in access to safe water and sanitation
between rural and urban communities with access to both water and sanitation 31 services in
rural generally much lower than in urban areas, especially in low to middle income countries
(LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented complex direct
biological routes and many broader, less direct routes including the socioeconomic aspects
such as accessibility and affordability of water supplies and sanitation facilities through which
water supply, sanitation, and hygiene (WASH) could lead to stunting. Cumming et al. (2014)
also had shown that over onethird of the world’s population was deprived from domestic access
to safe water and sanitation. Aheto et al. (2015) in their study on determinants of malnutrition
in Ghana found that absence of toilet facilities in households was associated with increased risk
of malnutrition. Similarly, Haile et al. (2016) analyzed secondary data of the 2011 Ethiopian
Demographic and Health Survey (EDHS) and found lack of a better-quality latrine to be
associated with high probabilities of stunting from the community-level factors. Torlesse et al.
(2016) found a prevalence of 28.4% and 6.7% for stunting and severe stunting, respectively
among children in their study on determinants of stunting in Indonesia. After controlling for
potential covariates, they found an association between household sanitary facility and
household water treatment. They reported an increased adjusted odds of childhood stunting up
to more than three times in households that drank unprocessed water if the household utilized
a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-7.28, p < 0.001) as compared to households
that drank treated water and utilized unimproved latrine where the adjusted odds on child
stunting was not significantly higher (AOR = 1.27, 95 % CI: 0.99-1.63, p = 0.06; Torlesse et
al., 2016). 32 Health-seeking behavior hinges on the direct route between socio-economic
position (SEP) and health outcomes (Benova et al., 2014). Various factors that influence health
seeking behavior have been reported including socioeconomic and demographic
characteristics, perceived need, accessibility, and service availability (Sarker et al., 2016).
Uggla and Mace (2016) opined that parental investment in health-seeking behaviors was
independently influenced by the association between maternal factors (age, health, and marital
status) and child factors (birth order, health, sex, and age). The authors found higher odds of
investment among children with lower birth order, older mothers and mothers with better health
status. Similarly, scholars opined that higher maternal education and wealth, and better access
to health services were associated with higher levels of health investment in children. The
authors showed that monogamously married women had higher odds of investment than non-
married women in all health-seeking behaviours (Uggla & Mace, 2016). Therefore, the
Botswana Government policy makers should work in order to improve the primary health care
system, promote early health seeking behavior, and support to the single mothers in terms of
education and income resources so that these mothers could make sound decisions towards
their under-5 children. Contributing Causes of Stunting Key study variables identified as
contributing to stunting are specific to food and nutrition (food security, food diversity), health
seeking practices (breastfeeding behaviors), family structure, and emotional care and family
support systems. 33 Food insecurity and diversity: Food availability at the household level
plays a crucial role in childhood health. Mechanisms through which food insecurity leads to
stunting are inequality of food consumption in terms of quality, quantity, and continuity; and
inadequacy of infant and child feeding which in turn leads to food consumption issues (FAO,
2017). In their study, Shinsugi et al. (2015) opined that stunting and severe stunting were
associated with food insecurity. The authors found that moderately food insecure households
were more likely to have stunted children as compared to food secure households (OR = 1.26;
95% BCI: 1.04–1.51). Moreover, children who lived in moderately food insecure households
were more likely to be severely stunted than children living in food secure households (OR =
1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these findings are in accord with
previous works, Hagos et al. (2017) noted that wealth played no role in the relationship between
food insecurity and stunting. The late introduction of additional foods, the use of nutrient-poor
foods, or limited access to essential nutrients especially vitamin A, iron, and zinc were
recognized as contributing to higher stunting risk in food-insecure households (Moradi et al.,
2019). While some authors found that stunting was associated with household food insecurity,
Motbainor et al. (2015) documented that food security was not the only criterion to determine
childhood nutritional status, but rather other factors such as mother’s knowledge of child
nutrition and health care practices, maternal nutritional status, intrahousehold food allocation
and utilization practices, and access to health services and healthy environmental conditions
were to be taken into account. 34 Also of concern is food/dietary diversity to ensure an adequate
and balanced diet for energy, growth and development. Food diversity is the number of varied
foods or food groups consumed over a given reference period (Bukania et al., 2014).
Researchers in Ghana have explored the relationship between dietary diversity and child health
outcomes and found a general inclination of food diversification with increased child age
(Frempong & Annim, 2017). They noted a significant relationship between food diversity in
the health of children born to educated mothers (Frempong & Annim, 2017). Also, food
diversity and number of meals the child ate per day were significantly associated with stunting
(β = 0.039, p < 0.01) and underweight (β = 0.035, p < 0.05) respectively (Motbainor et al.,
2015). Food security and nutrition status of household members have been found to be
dependent on culture and institutional frameworks in either male headed household or female
headed household (Mikalitsa, 2015). Several authors found higher incidences of malnutrition
and food poverty in the de jure female-headed households as compared to male headed
households and de facto female headed households (Mikalitsa, 2015; Ragasa et al., 2019).
Ragasa et al. (2019) examined whether providing agriculture and nutritional information to
both mother and men had an impact on household food security. The authors found that the
quantity and quality of household food availability was determined by agricultural production
and income. Moreover, Ragasa et al. (2019) noted that in male-only and female-only headed
household’s food insecurity was on the high. Formal education and literacy levels and nutrition
education appeared to lead to improved food security in households with sole male adults (
headed by male) while food security in de 35 jure female headed households was driven by
landholdings, livestock units (especially poultry) having other assets, planting tree and root
crops, and nutrition-related informationResearchers have documented poor school uptake
among stunted children and a high risk of mortality and susceptibility to infections (Tariku et
al., 2017). Stunting has important economic consequences for both sexes at the individual,
household, and community level. Authors have shown an association between shorter adult
stature and labor-market outcomes such as lower earnings and poorer productivity (de Onis &
Branca, 2016). Increases in morbidity, mortality, and health expenditure and subsequent
reductions in human capital investment, physical capital investment, and labor supply are all
various pathways through which undernutrition could affect aggregate economic growth
(McGovern et al., 2017). Ill health leads to reductions 25 in productivity and could negatively
affect individuals’ and society’s economic growth, especially when a significant proportion of
a country’s population is affected by stunting (McGovern et al., 2017). Underlying Causes of
Stunting The key study variables identified as underlying causes for stunting are discussed in
terms of the study population’s economic resources, social determinants of health, and
environmental conditions. Parental Employment Parental employment status or the
occupational standing of the household head has been associated with child's nutrition status
(stunting). Rashad and Sharaf (2019) conducted a study to estimate the causal impact of
women’s work on child nutritional status as measured by the height-for-age score. The authors
found a negative correlation between under-5 nutritional status and their mother’s employment
and reduced odds of stunting for children whose parents (households’ heads) were farmers or
self-employed relative to households headed by housewives. Rashad and Sharaf (2019) argued
that it is because couples’ income combined could procure better services than those gained by
a single parent (mothers). Parental Education Parental education has been linked to childhood
mortality through different pathways and is often associated with the socioeconomic status of
the household. Frost et al. (2005) mentioned five pathways through which the linear
relationship between parental education and mortality among under-five children occur: (a)
socioeconomic 26 status, (b) health knowledge, (c) attitudes towards health care, (d) female
autonomy, and (e) reproductive behaviors. These pathways operate at both individual and
community levels. Mosley and Chen (1984), in their seminal work, observed a strong linear
relationship between mother’s educational achievement and reduction in infant and under-5
mortality in South Saharan African countries. Likewise, researchers found that high literacy
levels among mothers in a given country of sub-Saharan Africa promoted the utilization of
health care facilities and the lowering of risk for under-5 child mortality (Anyamele et al., 2017;
McTavish et al., 2010). Nkurunziza et al. (2017) found that children whose mothers had no
education were more likely to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely
stunted (cOR= 2.0; 95% CI: 1.3-2.9; p < 0.001) relative to those whose mothers who completed
secondary school and above. Likewise, Acharya et al. (2018), in their analysis of determinants
of childhood stunting in Democratic Republic of Congo, found a negative association between
level of education and childhood stunting. Similarly, Birhanu et al. (2017) found an increased
mortality rate for infants and under-5 children whose parents were illiterate as well as a strong
association between stunting and parental education level. The number of stunted children was
found to be four times higher among illiterate mothers as compared to those from educated
mothers (Birhanu et al., 2017). Similarly, the likelihood of childhood stunting from illiterate
fathers was eight times higher as compared to those children who had educated fathers (Birhanu
et al., 2017). Even though the pathways are clearly understood, authors agree that parental
education is a key determinant of childhood nutritional status as it plays a central role by
empowering both women and men to offer 27 adequate childcare in terms of health, child
feeding, and child education (Geberselassie et al., 2018). Researchers noted a protective effect
against stunting in children whose fathers completed secondary school as compared to children
whose fathers completed primary school (Geberselassie et al., 2018). However, this protective
effect might be missing in single mother households. Clark and Hamplová (2013) explored
single motherhood and child mortality in sub-Saharan Africa and found that children born to
never-married single mothers were significantly more likely to die before age five in six
countries as compared with children whose parents were married (odds ratios range from 1.36
in Nigeria to 2.61 in Zimbabwe). In nine countries, they found children of formerly married
mothers were at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to
1.75 in Kenya) relative to children having married parents (Clark & Hamplová, 2013). These
findings are supported by Härkönen (2018) who found a universal growing negative
educational gradient of single motherhood across countries. The negative educational gap
promotes social inequality among single mothers and their children who live in a cycle of
poverty, and experience health related risks, and poor educational achievement. Therefore,
policy makers, public health organizations and country governments should empower single
mothers with better education and environmental incentives to help ensure positive child health
outcomes both at the community and individual levels (Frost et al., 2005). Birth Interval 28
The relationship of birth interval and malnutrition has been linked to maternal education. The
general consensus among researchers is that chronic malnutrition (stunting) is promoted by
large sibling numbers and by later birth order. Researchers in sub-Saharan Africa found boys
to be more affected by chronic malnutrition (stunting) as opposed to girls. For example,
Guilbert and Marazyan (2018) found that among mothers who were single, first-born boys were
more exposed to childhood mortality and not the first born girls’ counterpart. They found that
sister and not brother played a protective role towards lowering childhood mortality rate for
second- born children. Similarly, Bukusuba et al. (2017) in their case control study among
children aged six and 59 months in Uganda found that boys were significantly more likely to
be stunted compared to girls (OR= 2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al.
(2018) found increased odds of childhood stunting for children born within a preceding birth
interval less than 24 months. Equally, Liwin and Houle (2019) found the risk of dying more
than five times higher among children aged 12–23 months in the interval of zero to 11 months
between the index child and the succeeding while for a succeeding birth interval of 0–23
months the risk of dying among children aged 24–59 months increased by two-fold (Liwin &
Houle, 2019). Economic Resources and Living Conditions Linked to education level of parents
and opportunity for employment is the potential for improved household income which also
plays a pivotal role in determining child health outcome. Household income is tied to parental
education and child health 29 and impacts food intake through improved food purchasing
power, quality of medication via its access and affordability, and household sanitation (Umar
Farooq et al., 2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014)
found higher probabilities of childhood stunting in single mother households being
significantly affected by economic resources and parental education. Moreover, Mikalitsa
(2015) while examining the intra household allocation, household headship and nutrition of
under-fives in western Kenya argued that better nutrition status of children was not always
linked to high income in a household, but rather to who controls the income. The author found
a linear association between improved childhood nutritional status in households and active
participation of women in decision making (Mikalitsa, 2015). Households with uneducated
parents were inclined to have low-income levels. These parents spent less on nutritional
adequate foods and their children were more susceptible to growth failure due to lack of access
to sufficient food of adequate quality as well as poor living conditions, lack of access to basic
health care services and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019).
Thow et al. (2016) in their discourse on effect of remittances on diet and nutrition found that
remittances could lead to household food security and underweight reduction even though
remittances were found to play a little effect on chronic undernourishment. In addition to the
economic resources available to a household, where individuals in a household reside the effect
of the environs are important considerations to stunting outcomes. In particular, is the
recognized rural-urban inequality on childhood stunting and that malnutrition in under-five
children has been variably distributed according to 30 geographic regions. Kismul et al. (2018)
when analyzing the determinants of stunting in the Democratic Republic of Congo found a
significant variability in the prevalence of stunting between rural and urban regions with rural
areas having a larger percentage of children living with stunting than children living urban
areas. In another study on environmental predictors of stunting among under-5 children in
Somalia, authors found variation in stunting distribution spatially and temporally due to rains
seasons and vegetation (Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with
previous findings showed that rural populations were characterized as having worse nutrition
outcomes than those of urban populations. However, they demonstrated that the nutritional
inequality that exists between rural-urban regions stems from differences in wealth, education,
health, and non-road infrastructure services across rural and urban areas (Stifel et al., 2018).
Thus, reduction of childhood stunting requires improvement of parental education,
improvement of and access to healthy diets, and improvement of household’s wealth as these
are identified in the literature as the major predictors of stunting (Headey et al., 2017). A
component of the environment is water, but also relevant to hygiene and sanitation and a key
nutrient for health is access to safe water in rural and urban communities. Childhood stunting
has been negatively associated with poor access to safe water and to hygienic toilet (Kismul et
al., 2018). While Freeman et al. (2014) estimated that fewer than one in five people globally
wash their hands with soap after defecation, Bain et al. (2014) demonstrated the presence of
disparities in access to safe water and sanitation between rural and urban communities with
access to both water and sanitation 31 services in rural generally much lower than in urban
areas, especially in low to middle income countries (LMIC; Bain et al., 2014). Cumming and
Cairncross (2016) documented complex direct biological routes and many broader, less direct
routes including the socioeconomic aspects such as accessibility and affordability of water
supplies and sanitation facilities through which water supply, sanitation, and hygiene (WASH)
could lead to stunting. Cumming et al. (2014) also had shown that over onethird of the world’s
population was deprived from domestic access to safe water and sanitation. Aheto et al. (2015)
in their study on determinants of malnutrition in Ghana found that absence of toilet facilities in
households was associated with increased risk of malnutrition. Similarly, Haile et al. (2016)
analyzed secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and
found lack of a better-quality latrine to be associated with high probabilities of stunting from
the community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information They argued that PEE is closely linked to stunting and is
characterized by the predominance of the inflammatory process dominated by villous atrophy,
crypt hyperplasia, increased permeability, inflammatory cell infiltrate, and modest
malabsorption caused by fecal bacteria ingested in large quantities by young children living in
conditions of poor sanitation, and hygiene is crucial in the understanding of childhood stunting
(Millward, 2017; Vonaesch et al., 2018). In their study Brown et al. (2015) concluded that
PEE/stunting in the first two years of life was characterized by enteric infections caused by
Shigella and enterotoxins. Long-term consequences related to childhood stunting syndrome are
varied and of massive severity. These include increased morbidity and mortality; reduced
physical, neurodevelopmental, and economic capacity, and an elevated risk of metabolic
disease into adulthood (de Onis & Branca, 2016). Researchers have documented poor school
uptake among stunted children and a high risk of mortality and susceptibility to infections
(Tariku et al., 2017). Stunting has important economic consequences for both sexes at the
individual, household, and community level. Authors have shown an association between
shorter adult stature and labor-market outcomes such as lower earnings and poorer productivity
(de Onis & Branca, 2016). Increases in morbidity, mortality, and health expenditure and
subsequent reductions in human capital investment, physical capital investment, and labor
supply are all various pathways through which undernutrition could affect aggregate economic
growth (McGovern et al., 2017). Ill health leads to reductions 25 in productivity and could
negatively affect individuals’ and society’s economic growth, especially when a significant
proportion of a country’s population is affected by stunting (McGovern et al., 2017).
Underlying Causes of Stunting The key study variables identified as underlying causes for
stunting are discussed in terms of the study population’s economic resources, social
determinants of health, and environmental conditions. Parental Employment Parental
employment status or the occupational standing of the household head has been associated with
child's nutrition status (stunting). Rashad and Sharaf (2019) conducted a study to estimate the
causal impact of women’s work on child nutritional status as measured by the height-for-age
score. The authors found a negative correlation between under-5 nutritional status and their
mother’s employment and reduced odds of stunting for children whose parents (households’
heads) were farmers or self-employed relative to households headed by housewives. Rashad
and Sharaf (2019) argued that it is because couples’ income combined could procure better
services than those gained by a single parent (mothers). Parental Education Parental education
has been linked to childhood mortality through different pathways and is often associated with
the socioeconomic status of the household. Frost et al. (2005) mentioned five pathways through
which the linear relationship between parental education and mortality among under-five
children occur: (a) socioeconomic 26 status, (b) health knowledge, (c) attitudes towards health
care, (d) female autonomy, and (e) reproductive behaviors. These pathways operate at both
individual and community levels. Mosley and Chen (1984), in their seminal work, observed a
strong linear relationship between mother’s educational achievement and reduction in infant
and under-5 mortality in South Saharan African countries. Likewise, researchers found that
high literacy levels among mothers in a given country of sub-Saharan Africa promoted the
utilization of health care facilities and the lowering of risk for under-5 child mortality
(Anyamele et al., 2017; McTavish et al., 2010). Nkurunziza et al. (2017) found that children
whose mothers had no education were more likely to be stunted (cOR = 2.3; 95% CI: 1.7-3; p
< 0.001) and severely stunted (cOR= 2.0; 95% CI: 1.3-2.9; p < 0.001) relative to those whose
mothers who completed secondary school and above. Likewise, Acharya et al. (2018), in their
analysis of determinants of childhood stunting in Democratic Republic of Congo, found a
negative association between level of education and childhood stunting. Similarly, Birhanu et
al. (2017) found an increased mortality rate for infants and under-5 children whose parents
were illiterate as well as a strong association between stunting and parental education level.
The number of stunted children was found to be four times higher among illiterate mothers as
compared to those from educated mothers (Birhanu et al., 2017). Similarly, the likelihood of
childhood stunting from illiterate fathers was eight times higher as compared to those children
who had educated fathers (Birhanu et al., 2017). Even though the pathways are clearly
understood, authors agree that parental education is a key determinant of childhood nutritional
status as it plays a central role by empowering both women and men to offer 27 adequate
childcare in terms of health, child feeding, and child education (Geberselassie et al., 2018).
Researchers noted a protective effect against stunting in children whose fathers completed
secondary school as compared to children whose fathers completed primary school
(Geberselassie et al., 2018). However, this protective effect might be missing in single mother
households. Clark and Hamplová (2013) explored single motherhood and child mortality in
sub-Saharan Africa and found that children born to never-married single mothers were
significantly more likely to die before age five in six countries as compared with children
whose parents were married (odds ratios range from 1.36 in Nigeria to 2.61 in Zimbabwe). In
nine countries, they found children of formerly married mothers were at a significantly higher
risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya) relative to children
having married parents (Clark & Hamplová, 2013). These findings are supported by Härkönen
(2018) who found a universal growing negative educational gradient of single motherhood
across countries. The negative educational gap promotes social inequality among single
mothers and their children who live in a cycle of poverty, and experience health related risks,
and poor educational achievement. Therefore, policy makers, public health organizations and
country governments should empower single mothers with better education and environmental
incentives to help ensure positive child health outcomes both at the community and individual
levels (Frost et al., 2005). Birth Interval 28 The relationship of birth interval and malnutrition
has been linked to maternal education. The general consensus among researchers is that chronic
malnutrition (stunting) is promoted by large sibling numbers and by later birth order.
Researchers in sub-Saharan Africa found boys to be more affected by chronic malnutrition
(stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found that among
mothers who were single, first-born boys were more exposed to childhood mortality and not
the first born girls’ counterpart. They found that sister and not brother played a protective role
towards lowering childhood mortality rate for second- born children. Similarly, Bukusuba et
al. (2017) in their case control study among children aged six and 59 months in Uganda found
that boys were significantly more likely to be stunted compared to girls (OR= 2.2, 95% CI:
1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of childhood stunting
for children born within a preceding birth interval less than 24 months. Equally, Liwin and
Houle (2019) found the risk of dying more than five times higher among children aged 12–23
months in the interval of zero to 11 months between the index child and the succeeding while
for a succeeding birth interval of 0–23 months the risk of dying among children aged 24–59
months increased by two-fold (Liwin & Houle, 2019). Economic Resources and Living
Conditions Linked to education level of parents and opportunity for employment is the
potential for improved household income which also plays a pivotal role in determining child
health outcome. Household income is tied to parental education and child health 29 and impacts
food intake through improved food purchasing power, quality of medication via its access and
affordability, and household sanitation (Umar Farooq et al., 2019). In their study in Cameroon
and DRC, Ntoimo and Odimegwu (2014) found higher probabilities of childhood stunting in
single mother households being significantly affected by economic resources and parental
education. Moreover, Mikalitsa (2015) while examining the intra household allocation,
household headship and nutrition of under-fives in western Kenya argued that better nutrition
status of children was not always linked to high income in a household, but rather to who
controls the income. The author found a linear association between improved childhood
nutritional status in households and active participation of women in decision making
(Mikalitsa, 2015). Households with uneducated parents were inclined to have low-income
levels. These parents spent less on nutritional adequate foods and their children were more
susceptible to growth failure due to lack of access to sufficient food of adequate quality as well
as poor living conditions, lack of access to basic health care services and greater exposure to
diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in their discourse on effect
of remittances on diet and nutrition found that remittances could lead to household food
security and underweight reduction even though remittances were found to play a little effect
on chronic undernourishment. In addition to the economic resources available to a household,
where individuals in a household reside the effect of the environs are important considerations
to stunting outcomes. In particular, is the recognized rural-urban inequality on childhood
stunting and that malnutrition in under-five children has been variably distributed according to
30 geographic regions. Kismul et al. (2018) when analyzing the determinants of stunting in the
Democratic Republic of Congo found a significant variability in the prevalence of stunting
between rural and urban regions with rural areas having a larger percentage of children living
with stunting than children living urban areas. In another study on environmental predictors of
stunting among under-5 children in Somalia, authors found variation in stunting distribution
spatially and temporally due to rains seasons and vegetation (Kinyoki et al., 2016). Similarly,
Stifel et al. (2018) in keeping with previous findings showed that rural populations were
characterized as having worse nutrition outcomes than those of urban populations. However,
they demonstrated that the nutritional inequality that exists between rural-urban regions stems
from differences in wealth, education, health, and non-road infrastructure services across rural
and urban areas (Stifel et al., 2018). Thus, reduction of childhood stunting requires
improvement of parental education, improvement of and access to healthy diets, and
improvement of household’s wealth as these are identified in the literature as the major
predictors of stunting (Headey et al., 2017). A component of the environment is water, but also
relevant to hygiene and sanitation and a key nutrient for health is access to safe water in rural
and urban communities. Childhood stunting has been negatively associated with poor access
to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al. (2014) estimated
that fewer than one in five people globally wash their hands with soap after defecation, Bain et
al. (2014) demonstrated the presence of disparities in access to safe water and sanitation
between rural and urban communities with access to both water and sanitation 31 services in
rural generally much lower than in urban areas, especially in low to middle income countries
(LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented complex direct
biological routes and many broader, less direct routes including the socioeconomic aspects
such as accessibility and affordability of water supplies and sanitation facilities through which
water supply, sanitation, and hygiene (WASH) could lead to stunting. Cumming et al. (2014)
also had shown that over onethird of the world’s population was deprived from domestic access
to safe water and sanitation. Aheto et al. (2015) in their study on determinants of malnutrition
in Ghana found that absence of toilet facilities in households was associated with increased risk
of malnutrition. Similarly, Haile et al. (2016) analyzed secondary data of the 2011 Ethiopian
Demographic and Health Survey (EDHS) and found lack of a better-quality latrine to be
associated with high probabilities of stunting from the community-level factors. Torlesse et al.
(2016) found a prevalence of 28.4% and 6.7% for stunting and severe stunting, respectively
among children in their study on determinants of stunting in Indonesia. After controlling for
potential covariates, they found an association between household sanitary facility and
household water treatment. They reported an increased adjusted odds of childhood stunting up
to more than three times in households that drank unprocessed water if the household utilized
a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-7.28, p < 0.001) as compared to households
that drank treated water and utilized unimproved latrine where the adjusted odds on child
stunting was not significantly higher (AOR = 1.27, 95 % CI: 0.99-1.63, p = 0.06; Torlesse et
al., 2016). 32 Health-seeking behavior hinges on the direct route between socio-economic
position (SEP) and health outcomes (Benova et al., 2014). Various factors that influence health
seeking behavior have been reported including socioeconomic and demographic
characteristics, perceived need, accessibility, and service availability (Sarker et al., 2016).
Uggla and Mace (2016) opined that parental investment in health-seeking behaviors was
independently influenced by the association between maternal factors (age, health, and marital
status) and child factors (birth order, health, sex, and age). The authors found higher odds of
investment among children with lower birth order, older mothers and mothers with better health
status. Similarly, scholars opined that higher maternal education and wealth, and better access
to health services were associated with higher levels of health investment in children. The
authors showed that monogamously married women had higher odds of investment than non-
married women in all health-seeking behaviours (Uggla & Mace, 2016). Therefore, the
Botswana Government policy makers should work in order to improve the primary health care
system, promote early health seeking behavior, and support to the single mothers in terms of
education and income resources so that these mothers could make sound decisions towards
their under-5 children. Contributing Causes of Stunting Key study variables identified as
contributing to stunting are specific to food and nutrition (food security, food diversity), health
seeking practices (breastfeeding behaviors), family structure, and emotional care and family
support systems. 33 Food insecurity and diversity: Food availability at the household level
plays a crucial role in childhood health. Mechanisms through which food insecurity leads to
stunting are inequality of food consumption in terms of quality, quantity, and continuity; and
inadequacy of infant and child feeding which in turn leads to food consumption issues (FAO,
2017). In their study, Shinsugi et al. (2015) opined that stunting and severe stunting were
associated with food insecurity. The authors found that moderately food insecure households
were more likely to have stunted children as compared to food secure households (OR = 1.26;
95% BCI: 1.04–1.51). Moreover, children who lived in moderately food insecure households
were more likely to be severely stunted than children living in food secure households (OR =
1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these findings are in accord with
previous works, Hagos et al. (2017) noted that wealth played no role in the relationship between
food insecurity and stunting. The late introduction of additional foods, the use of nutrient-poor
foods, or limited access to essential nutrients especially vitamin A, iron, and zinc were
recognized as contributing to higher stunting risk in food-insecure households (Moradi et al.,
2019). While some authors found that stunting was associated with household food insecurity,
Motbainor et al. (2015) documented that food security was not the only criterion to determine
childhood nutritional status, but rather other factors such as mother’s knowledge of child
nutrition and health care practices, maternal nutritional status, intrahousehold food allocation
and utilization practices, and access to health services and healthy environmental conditions
were to be taken into account. 34 Also of concern is food/dietary diversity to ensure an adequate
and balanced diet for energy, growth and development. Food diversity is the number of varied
foods or food groups consumed over a given reference period (Bukania et al., 2014).
Researchers in Ghana have explored the relationship between dietary diversity and child health
outcomes and found a general inclination of food diversification with increased child age
(Frempong & Annim, 2017). They noted a significant relationship between food diversity in
the health of children born to educated mothers (Frempong & Annim, 2017). Also, food
diversity and number of meals the child ate per day were significantly associated with stunting
(β = 0.039, p < 0.01) and underweight (β = 0.035, p < 0.05) respectively (Motbainor et al.,
2015). Food security and nutrition status of household members have been found to be
dependent on culture and institutional frameworks in either male headed household or female
headed household (Mikalitsa, 2015). Several authors found higher incidences of malnutrition
and food poverty in the de jure female-headed households as compared to male headed
households and de facto female headed households (Mikalitsa, 2015; Ragasa et al., 2019).
Ragasa et al. (2019) examined whether providing agriculture and nutritional information to
both mother and men had an impact on household food security. The authors found that the
quantity and quality of household food availability was determined by agricultural production
and income. Moreover, Ragasa et al. (2019) noted that in male-only and female-only headed
household’s food insecurity was on the high. Formal education and literacy levels and nutrition
education appeared to lead to improved food security in households with sole male adults (
headed by male) while food security in de 35 jure female headed households was driven by
landholdings, livestock units (especially poultry) having other assets, planting tree and root
crops, and nutrition-related information
Researchers have documented poor school uptake among stunted children and a high risk of
mortality and susceptibility to infections (Tariku et al., 2017). Stunting has important economic
consequences for both sexes at the individual, household, and community level. Authors have
shown an association between shorter adult stature and labor-market outcomes such as lower
earnings and poorer productivity (de Onis & Branca, 2016). Increases in morbidity, mortality,
and health expenditure and subsequent reductions in human capital investment, physical capital
investment, and labor supply are all various pathways through which undernutrition could
affect aggregate economic growth (McGovern et al., 2017). Ill health leads to reductions 25 in
productivity and could negatively affect individuals’ and society’s economic growth,
especially when a significant proportion of a country’s population is affected by stunting
(McGovern et al., 2017). Underlying Causes of Stunting The key study variables identified as
underlying causes for stunting are discussed in terms of the study population’s economic
resources, social determinants of health, and environmental conditions. Parental Employment
Parental employment status or the occupational standing of the household head has been
associated with child's nutrition status (stunting). Rashad and Sharaf (2019) conducted a study
to estimate the causal impact of women’s work on child nutritional status as measured by the
height-for-age score. The authors found a negative correlation between under-5 nutritional
status and their mother’s employment and reduced odds of stunting for children whose parents
(households’ heads) were farmers or self-employed relative to households headed by
housewives. Rashad and Sharaf (2019) argued that it is because couples’ income combined
could procure better services than those gained by a single parent (mothers). Parental Education
Parental education has been linked to childhood mortality through different pathways and is
often associated with the socioeconomic status of the household. Frost et al. (2005) mentioned
five pathways through which the linear relationship between parental education and mortality
among under-five children occur: (a) socioeconomic 26 status, (b) health knowledge, (c)
attitudes towards health care, (d) female autonomy, and (e) reproductive behaviors. These
pathways operate at both individual and community levels. Mosley and Chen (1984), in their
seminal work, observed a strong linear relationship between mother’s educational achievement
and reduction in infant and under-5 mortality in South Saharan African countries. Likewise,
researchers found that high literacy levels among mothers in a given country of sub-Saharan
Africa promoted the utilization of health care facilities and the lowering of risk for under-5
child mortality (Anyamele et al., 2017; McTavish et al., 2010). Nkurunziza et al. (2017) found
that children whose mothers had no education were more likely to be stunted (cOR = 2.3; 95%
CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI: 1.3-2.9; p < 0.001) relative to
those whose mothers who completed secondary school and above. Likewise, Acharya et al.
(2018), in their analysis of determinants of childhood stunting in Democratic Republic of
Congo, found a negative association between level of education and childhood stunting.
Similarly, Birhanu et al. (2017) found an increased mortality rate for infants and under-5
children whose parents were illiterate as well as a strong association between stunting and
parental education level. The number of stunted children was found to be four times higher
among illiterate mothers as compared to those from educated mothers (Birhanu et al., 2017).
Similarly, the likelihood of childhood stunting from illiterate fathers was eight times higher as
compared to those children who had educated fathers (Birhanu et al., 2017). Even though the
pathways are clearly understood, authors agree that parental education is a key determinant of
childhood nutritional status as it plays a central role by empowering both women and men to
offer 27 adequate childcare in terms of health, child feeding, and child education (Geberselassie
et al., 2018). Researchers noted a protective effect against stunting in children whose fathers
completed secondary school as compared to children whose fathers completed primary school
(Geberselassie et al., 2018). However, this protective effect might be missing in single mother
households. Clark and Hamplová (2013) explored single motherhood and child mortality in
sub-Saharan Africa and found that children born to never-married single mothers were
significantly more likely to die before age five in six countries as compared with children
whose parents were married (odds ratios range from 1.36 in Nigeria to 2.61 in Zimbabwe). In
nine countries, they found children of formerly married mothers were at a significantly higher
risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya) relative to children
having married parents (Clark & Hamplová, 2013). These findings are supported by Härkönen
(2018) who found a universal growing negative educational gradient of single motherhood
across countries. The negative educational gap promotes social inequality among single
mothers and their children who live in a cycle of poverty, and experience health related risks,
and poor educational achievement. Therefore, policy makers, public health organizations and
country governments should empower single mothers with better education and environmental
incentives to help ensure positive child health outcomes both at the community and individual
levels (Frost et al., 2005). Birth Interval 28 The relationship of birth interval and malnutrition
has been linked to maternal education. The general consensus among researchers is that chronic
malnutrition (stunting) is promoted by large sibling numbers and by later birth order.
Researchers in sub-Saharan Africa found boys to be more affected by chronic malnutrition
(stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found that among
mothers who were single, first-born boys were more exposed to childhood mortality and not
the first born girls’ counterpart. They found that sister and not brother played a protective role
towards lowering childhood mortality rate for second- born children. Similarly, Bukusuba et
al. (2017) in their case control study among children aged six and 59 months in Uganda found
that boys were significantly more likely to be stunted compared to girls (OR= 2.2, 95% CI:
1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of childhood stunting
for children born within a preceding birth interval less than 24 months. Equally, Liwin and
Houle (2019) found the risk of dying more than five times higher among children aged 12–23
months in the interval of zero to 11 months between the index child and the succeeding while
for a succeeding birth interval of 0–23 months the risk of dying among children aged 24–59
months increased by two-fold (Liwin & Houle, 2019). Economic Resources and Living
Conditions Linked to education level of parents and opportunity for employment is the
potential for improved household income which also plays a pivotal role in determining child
health outcome. Household income is tied to parental education and child health 29 and impacts
food intake through improved food purchasing power, quality of medication via its access and
affordability, and household sanitation (Umar Farooq et al., 2019). In their study in Cameroon
and DRC, Ntoimo and Odimegwu (2014) found higher probabilities of childhood stunting in
single mother households being significantly affected by economic resources and parental
education. Moreover, Mikalitsa (2015) while examining the intra household allocation,
household headship and nutrition of under-fives in western Kenya argued that better nutrition
status of children was not always linked to high income in a household, but rather to who
controls the income. The author found a linear association between improved childhood
nutritional status in households and active participation of women in decision making
(Mikalitsa, 2015). Households with uneducated parents were inclined to have low-income
levels. These parents spent less on nutritional adequate foods and their children were more
susceptible to growth failure due to lack of access to sufficient food of adequate quality as well
as poor living conditions, lack of access to basic health care services and greater exposure to
diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in their discourse on effect
of remittances on diet and nutrition found that remittances could lead to household food
security and underweight reduction even though remittances were found to play a little effect
on chronic undernourishment. In addition to the economic resources available to a household,
where individuals in a household reside the effect of the environs are important considerations
to stunting outcomes. In particular, is the recognized rural-urban inequality on childhood
stunting and that malnutrition in under-five children has been variably distributed according to
30 geographic regions. Kismul et al. (2018) when analyzing the determinants of stunting in the
Democratic Republic of Congo found a significant variability in the prevalence of stunting
between rural and urban regions with rural areas having a larger percentage of children living
with stunting than children living urban areas. In another study on environmental predictors of
stunting among under-5 children in Somalia, authors found variation in stunting distribution
spatially and temporally due to rains seasons and vegetation (Kinyoki et al., 2016). Similarly,
Stifel et al. (2018) in keeping with previous findings showed that rural populations were
characterized as having worse nutrition outcomes than those of urban populations. However,
they demonstrated that the nutritional inequality that exists between rural-urban regions stems
from differences in wealth, education, health, and non-road infrastructure services across rural
and urban areas (Stifel et al., 2018). Thus, reduction of childhood stunting requires
improvement of parental education, improvement of and access to healthy diets, and
improvement of household’s wealth as these are identified in the literature as the major
predictors of stunting (Headey et al., 2017). A component of the environment is water, but also
relevant to hygiene and sanitation and a key nutrient for health is access to safe water in rural
and urban communities. Childhood stunting has been negatively associated with poor access
to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al. (2014) estimated
that fewer than one in five people globally wash their hands with soap after defecation, Bain et
al. (2014) demonstrated the presence of disparities in access to safe water and sanitation
between rural and urban communities with access to both water and sanitation 31 services in
rural generally much lower than in urban areas, especially in low to middle income countries
(LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented complex direct
biological routes and many broader, less direct routes including the socioeconomic aspects
such as accessibility and affordability of water supplies and sanitation facilities through which
water supply, sanitation, and hygiene (WASH) could lead to stunting. Cumming et al. (2014)
also had shown that over onethird of the world’s population was deprived from domestic access
to safe water and sanitation. Aheto et al. (2015) in their study on determinants of malnutrition
in Ghana found that absence of toilet facilities in households was associated with increased risk
of malnutrition. Similarly, Haile et al. (2016) analyzed secondary data of the 2011 Ethiopian
Demographic and Health Survey (EDHS) and found lack of a better-quality latrine to be
associated with high probabilities of stunting from the community-level factors. Torlesse et al.
(2016) found a prevalence of 28.4% and 6.7% for stunting and severe stunting, respectively
among children in their study on determinants of stunting in Indonesia. After controlling for
potential covariates, they found an association between household sanitary facility and
household water treatment. They reported an increased adjusted odds of childhood stunting up
to more than three times in households that drank unprocessed water if the household utilized
a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-7.28, p < 0.001) as compared to households
that drank treated water and utilized unimproved latrine where the adjusted odds on child
stunting was not significantly higher (AOR = 1.27, 95 % CI: 0.99-1.63, p = 0.06; Torlesse et
al., 2016). 32 Health-seeking behavior hinges on the direct route between socio-economic
position (SEP) and health outcomes (Benova et al., 2014). Various factors that influence health
seeking behavior have been reported including socioeconomic and demographic
characteristics, perceived need, accessibility, and service availability (Sarker et al., 2016).
Uggla and Mace (2016) opined that parental investment in health-seeking behaviors was
independently influenced by the association between maternal factors (age, health, and marital
status) and child factors (birth order, health, sex, and age). The authors found higher odds of
investment among children with lower birth order, older mothers and mothers with better health
status. Similarly, scholars opined that higher maternal education and wealth, and better access
to health services were associated with higher levels of health investment in children. The
authors showed that monogamously married women had higher odds of investment than non-
married women in all health-seeking behaviours (Uggla & Mace, 2016). Therefore, the
Botswana Government policy makers should work in order to improve the primary health care
system, promote early health seeking behavior, and support to the single mothers in terms of
education and income resources so that these mothers could make sound decisions towards
their under-5 children. Contributing Causes of Stunting Key study variables identified as
contributing to stunting are specific to food and nutrition (food security, food diversity), health
seeking practices (breastfeeding behaviors), family structure, and emotional care and family
support systems. 33 Food insecurity and diversity: Food availability at the household level
plays a crucial role in childhood health. Mechanisms through which food insecurity leads to
stunting are inequality of food consumption in terms of quality, quantity, and continuity; and
inadequacy of infant and child feeding which in turn leads to food consumption issues (FAO,
2017). In their study, Shinsugi et al. (2015) opined that stunting and severe stunting were
associated with food insecurity. The authors found that moderately food insecure households
were more likely to have stunted children as compared to food secure households (OR = 1.26;
95% BCI: 1.04–1.51). Moreover, children who lived in moderately food insecure households
were more likely to be severely stunted than children living in food secure households (OR =
1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these findings are in accord with
previous works, Hagos et al. (2017) noted that wealth played no role in the relationship between
food insecurity and stunting. The late introduction of additional foods, the use of nutrient-poor
foods, or limited access to essential nutrients especially vitamin A, iron, and zinc were
recognized as contributing to higher stunting risk in food-insecure households (Moradi et al.,
2019). While some authors found that stunting was associated with household food insecurity,
Motbainor et al. (2015) documented that food security was not the only criterion to determine
childhood nutritional status, but rather other factors such as mother’s knowledge of child
nutrition and health care practices, maternal nutritional status, intrahousehold food allocation
and utilization practices, and access to health services and healthy environmental conditions
were to be taken into account. 34 Also of concern is food/dietary diversity to ensure an adequate
and balanced diet for energy, growth and development. Food diversity is the number of varied
foods or food groups consumed over a given reference period (Bukania et al., 2014).
Researchers in Ghana have explored the relationship between dietary diversity and child health
outcomes and found a general inclination of food diversification with increased child age
(Frempong & Annim, 2017). They noted a significant relationship between food diversity in
the health of children born to educated mothers (Frempong & Annim, 2017). Also, food
diversity and number of meals the child ate per day were significantly associated with stunting
(β = 0.039, p < 0.01) and underweight (β = 0.035, p < 0.05) respectively (Motbainor et al.,
2015). Food security and nutrition status of household members have been found to be
dependent on culture and institutional frameworks in either male headed household or female
headed household (Mikalitsa, 2015). Several authors found higher incidences of malnutrition
and food poverty in the de jure female-headed households as compared to male headed
households and de facto female headed households (Mikalitsa, 2015; Ragasa et al., 2019).
Ragasa et al. (2019) examined whether providing agriculture and nutritional information to
both mother and men had an impact on household food security. The authors found that the
quantity and quality of household food availability was determined by agricultural production
and income. Moreover, Ragasa et al. (2019) noted that in male-only and female-only headed
household’s food insecurity was on the high. Formal education and literacy levels and nutrition
education appeared to lead to improved food security in households with sole male adults (
headed by male) while food security in de 35 jure female headed households was driven by
landholdings, livestock units (especially poultry) having other assets, planting tree and root
crops, and nutrition-related informationResearchers have documented poor school uptake
among stunted children and a high risk of mortality and susceptibility to infections (Tariku et
al., 2017). Stunting has important economic consequences for both sexes at the individual,
household, and community level. Authors have shown an association between shorter adult
stature and labor-market outcomes such as lower earnings and poorer productivity (de Onis &
Branca, 2016). Increases in morbidity, mortality, and health expenditure and subsequent
reductions in human capital investment, physical capital investment, and labor supply are all
various pathways through which undernutrition could affect aggregate economic growth
(McGovern et al., 2017). Ill health leads to reductions 25 in productivity and could negatively
affect individuals’ and society’s economic growth, especially when a significant proportion of
a country’s population is affected by stunting (McGovern et al., 2017). Underlying Causes of
Stunting The key study variables identified as underlying causes for stunting are discussed in
terms of the study population’s economic resources, social determinants of health, and
environmental conditions. Parental Employment Parental employment status or the
occupational standing of the household head has been associated with child's nutrition status
(stunting). Rashad and Sharaf (2019) conducted a study to estimate the causal impact of
women’s work on child nutritional status as measured by the height-for-age score. The authors
found a negative correlation between under-5 nutritional status and their mother’s employment
and reduced odds of stunting for children whose parents (households’ heads) were farmers or
self-employed relative to households headed by housewives. Rashad and Sharaf (2019) argued
that it is because couples’ income combined could procure better services than those gained by
a single parent (mothers). Parental Education Parental education has been linked to childhood
mortality through different pathways and is often associated with the socioeconomic status of
the household. Frost et al. (2005) mentioned five pathways through which the linear
relationship between parental education and mortality among under-five children occur: (a)
socioeconomic 26 status, (b) health knowledge, (c) attitudes towards health care, (d) female
autonomy, and (e) reproductive behaviors. These pathways operate at both individual and
community levels. Mosley and Chen (1984), in their seminal work, observed a strong linear
relationship between mother’s educational achievement and reduction in infant and under-5
mortality in South Saharan African countries. Likewise, researchers found that high literacy
levels among mothers in a given country of sub-Saharan Africa promoted the utilization of
health care facilities and the lowering of risk for under-5 child mortality (Anyamele et al., 2017;
McTavish et al., 2010). Nkurunziza et al. (2017) found that children whose mothers had no
education were more likely to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely
stunted (cOR= 2.0; 95% CI: 1.3-2.9; p < 0.001) relative to those whose mothers who completed
secondary school and above. Likewise, Acharya et al. (2018), in their analysis of determinants
of childhood stunting in Democratic Republic of Congo, found a negative association between
level of education and childhood stunting. Similarly, Birhanu et al. (2017) found an increased
mortality rate for infants and under-5 children whose parents were illiterate as well as a strong
association between stunting and parental education level. The number of stunted children was
found to be four times higher among illiterate mothers as compared to those from educated
mothers (Birhanu et al., 2017). Similarly, the likelihood of childhood stunting from illiterate
fathers was eight times higher as compared to those children who had educated fathers (Birhanu
et al., 2017). Even though the pathways are clearly understood, authors agree that parental
education is a key determinant of childhood nutritional status as it plays a central role by
empowering both women and men to offer 27 adequate childcare in terms of health, child
feeding, and child education (Geberselassie et al., 2018). Researchers noted a protective effect
against stunting in children whose fathers completed secondary school as compared to children
whose fathers completed primary school (Geberselassie et al., 2018). However, this protective
effect might be missing in single mother households. Clark and Hamplová (2013) explored
single motherhood and child mortality in sub-Saharan Africa and found that children born to
never-married single mothers were significantly more likely to die before age five in six
countries as compared with children whose parents were married (odds ratios range from 1.36
in Nigeria to 2.61 in Zimbabwe). In nine countries, they found children of formerly married
mothers were at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to
1.75 in Kenya) relative to children having married parents (Clark & Hamplová, 2013). These
findings are supported by Härkönen (2018) who found a universal growing negative
educational gradient of single motherhood across countries. The negative educational gap
promotes social inequality among single mothers and their children who live in a cycle of
poverty, and experience health related risks, and poor educational achievement. Therefore,
policy makers, public health organizations and country governments should empower single
mothers with better education and environmental incentives to help ensure positive child health
outcomes both at the community and individual levels (Frost et al., 2005). Birth Interval 28
The relationship of birth interval and malnutrition has been linked to maternal education. The
general consensus among researchers is that chronic malnutrition (stunting) is promoted by
large sibling numbers and by later birth order. Researchers in sub-Saharan Africa found boys
to be more affected by chronic malnutrition (stunting) as opposed to girls. For example,
Guilbert and Marazyan (2018) found that among mothers who were single, first-born boys were
more exposed to childhood mortality and not the first born girls’ counterpart. They found that
sister and not brother played a protective role towards lowering childhood mortality rate for
second- born children. Similarly, Bukusuba et al. (2017) in their case control study among
children aged six and 59 months in Uganda found that boys were significantly more likely to
be stunted compared to girls (OR= 2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al.
(2018) found increased odds of childhood stunting for children born within a preceding birth
interval less than 24 months. Equally, Liwin and Houle (2019) found the risk of dying more
than five times higher among children aged 12–23 months in the interval of zero to 11 months
between the index child and the succeeding while for a succeeding birth interval of 0–23
months the risk of dying among children aged 24–59 months increased by two-fold (Liwin &
Houle, 2019). Economic Resources and Living Conditions Linked to education level of parents
and opportunity for employment is the potential for improved household income which also
plays a pivotal role in determining child health outcome. Household income is tied to parental
education and child health 29 and impacts food intake through improved food purchasing
power, quality of medication via its access and affordability, and household sanitation (Umar
Farooq et al., 2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014)
found higher probabilities of childhood stunting in single mother households being
significantly affected by economic resources and parental education. Moreover, Mikalitsa
(2015) while examining the intra household allocation, household headship and nutrition of
under-fives in western Kenya argued that better nutrition status of children was not always
linked to high income in a household, but rather to who controls the income. The author found
a linear association between improved childhood nutritional status in households and active
participation of women in decision making (Mikalitsa, 2015). Households with uneducated
parents were inclined to have low-income levels. These parents spent less on nutritional
adequate foods and their children were more susceptible to growth failure due to lack of access
to sufficient food of adequate quality as well as poor living conditions, lack of access to basic
health care services and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019).
Thow et al. (2016) in their discourse on effect of remittances on diet and nutrition found that
remittances could lead to household food security and underweight reduction even though
remittances were found to play a little effect on chronic undernourishment. In addition to the
economic resources available to a household, where individuals in a household reside the effect
of the environs are important considerations to stunting outcomes. In particular, is the
recognized rural-urban inequality on childhood stunting and that malnutrition in under-five
children has been variably distributed according to 30 geographic regions. Kismul et al. (2018)
when analyzing the determinants of stunting in the Democratic Republic of Congo found a
significant variability in the prevalence of stunting between rural and urban regions with rural
areas having a larger percentage of children living with stunting than children living urban
areas. In another study on environmental predictors of stunting among under-5 children in
Somalia, authors found variation in stunting distribution spatially and temporally due to rains
seasons and vegetation (Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with
previous findings showed that rural populations were characterized as having worse nutrition
outcomes than those of urban populations. However, they demonstrated that the nutritional
inequality that exists between rural-urban regions stems from differences in wealth, education,
health, and non-road infrastructure services across rural and urban areas (Stifel et al., 2018).
Thus, reduction of childhood stunting requires improvement of parental education,
improvement of and access to healthy diets, and improvement of household’s wealth as these
are identified in the literature as the major predictors of stunting (Headey et al., 2017). A
component of the environment is water, but also relevant to hygiene and sanitation and a key
nutrient for health is access to safe water in rural and urban communities. Childhood stunting
has been negatively associated with poor access to safe water and to hygienic toilet (Kismul et
al., 2018). While Freeman et al. (2014) estimated that fewer than one in five people globally
wash their hands with soap after defecation, Bain et al. (2014) demonstrated the presence of
disparities in access to safe water and sanitation between rural and urban communities with
access to both water and sanitation 31 services in rural generally much lower than in urban
areas, especially in low to middle income countries (LMIC; Bain et al., 2014). Cumming and
Cairncross (2016) documented complex direct biological routes and many broader, less direct
routes including the socioeconomic aspects such as accessibility and affordability of water
supplies and sanitation facilities through which water supply, sanitation, and hygiene (WASH)
could lead to stunting. Cumming et al. (2014) also had shown that over onethird of the world’s
population was deprived from domestic access to safe water and sanitation. Aheto et al. (2015)
in their study on determinants of malnutrition in Ghana found that absence of toilet facilities in
households was associated with increased risk of malnutrition. Similarly, Haile et al. (2016)
analyzed secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and
found lack of a better-quality latrine to be associated with high probabilities of stunting from
the community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information
They argued that PEE is closely linked to stunting and is characterized by the predominance of
the inflammatory process dominated by villous atrophy, crypt hyperplasia, increased
permeability, inflammatory cell infiltrate, and modest malabsorption caused by fecal bacteria
ingested in large quantities by young children living in conditions of poor sanitation, and
hygiene is crucial in the understanding of childhood stunting (Millward, 2017; Vonaesch et al.,
2018). In their study Brown et al. (2015) concluded that PEE/stunting in the first two years of
life was characterized by enteric infections caused by Shigella and enterotoxins. Long-term
consequences related to childhood stunting syndrome are varied and of massive severity. These
include increased morbidity and mortality; reduced physical, neurodevelopmental, and
economic capacity, and an elevated risk of metabolic disease into adulthood (de Onis & Branca,
2016). Researchers have documented poor school uptake among stunted children and a high
risk of mortality and susceptibility to infections (Tariku et al., 2017). Stunting has important
economic consequences for both sexes at the individual, household, and community level.
Authors have shown an association between shorter adult stature and labor-market outcomes
such as lower earnings and poorer productivity (de Onis & Branca, 2016). Increases in
morbidity, mortality, and health expenditure and subsequent reductions in human capital
investment, physical capital investment, and labor supply are all various pathways through
which undernutrition could affect aggregate economic growth (McGovern et al., 2017). Ill
health leads to reductions 25 in productivity and could negatively affect individuals’ and
society’s economic growth, especially when a significant proportion of a country’s population
is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The key study
variables identified as underlying causes for stunting are discussed in terms of the study
population’s economic resources, social determinants of health, and environmental conditions.
Parental Employment Parental employment status or the occupational standing of the
household head has been associated with child's nutrition status (stunting). Rashad and Sharaf
(2019) conducted a study to estimate the causal impact of women’s work on child nutritional
status as measured by the height-for-age score. The authors found a negative correlation
between under-5 nutritional status and their mother’s employment and reduced odds of stunting
for children whose parents (households’ heads) were farmers or self-employed relative to
households headed by housewives. Rashad and Sharaf (2019) argued that it is because couples’
income combined could procure better services than those gained by a single parent (mothers).
Parental Education Parental education has been linked to childhood mortality through different
pathways and is often associated with the socioeconomic status of the household. Frost et al.
(2005) mentioned five pathways through which the linear relationship between parental
education and mortality among under-five children occur: (a) socioeconomic 26 status, (b)
health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e) reproductive
behaviors. These pathways operate at both individual and community levels. Mosley and Chen
(1984), in their seminal work, observed a strong linear relationship between mother’s
educational achievement and reduction in infant and under-5 mortality in South Saharan
African countries. Likewise, researchers found that high literacy levels among mothers in a
given country of sub-Saharan Africa promoted the utilization of health care facilities and the
lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information
Researchers have documented poor school uptake among stunted children and a high risk of
mortality and susceptibility to infections (Tariku et al., 2017). Stunting has important economic
consequences for both sexes at the individual, household, and community level. Authors have
shown an association between shorter adult stature and labor-market outcomes such as lower
earnings and poorer productivity (de Onis & Branca, 2016). Increases in morbidity, mortality,
and health expenditure and subsequent reductions in human capital investment, physical capital
investment, and labor supply are all various pathways through which undernutrition could
affect aggregate economic growth (McGovern et al., 2017). Ill health leads to reductions 25 in
productivity and could negatively affect individuals’ and society’s economic growth,
especially when a significant proportion of a country’s population is affected by stunting
(McGovern et al., 2017). Underlying Causes of Stunting The key study variables identified as
underlying causes for stunting are discussed in terms of the study population’s economic
resources, social determinants of health, and environmental conditions. Parental Employment
Parental employment status or the occupational standing of the household head has been
associated with child's nutrition status (stunting). Rashad and Sharaf (2019) conducted a study
to estimate the causal impact of women’s work on child nutritional status as measured by the
height-for-age score. The authors found a negative correlation between under-5 nutritional
status and their mother’s employment and reduced odds of stunting for children whose parents
(households’ heads) were farmers or self-employed relative to households headed by
housewives. Rashad and Sharaf (2019) argued that it is because couples’ income combined
could procure better services than those gained by a single parent (mothers). Parental Education
Parental education has been linked to childhood mortality through different pathways and is
often associated with the socioeconomic status of the household. Frost et al. (2005) mentioned
five pathways through which the linear relationship between parental education and mortality
among under-five children occur: (a) socioeconomic 26 status, (b) health knowledge, (c)
attitudes towards health care, (d) female autonomy, and (e) reproductive behaviors. These
pathways operate at both individual and community levels. Mosley and Chen (1984), in their
seminal work, observed a strong linear relationship between mother’s educational achievement
and reduction in infant and under-5 mortality in South Saharan African countries. Likewise,
researchers found that high literacy levels among mothers in a given country of sub-Saharan
Africa promoted the utilization of health care facilities and the lowering of risk for under-5
child mortality (Anyamele et al., 2017; McTavish et al., 2010). Nkurunziza et al. (2017) found
that children whose mothers had no education were more likely to be stunted (cOR = 2.3; 95%
CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI: 1.3-2.9; p < 0.001) relative to
those whose mothers who completed secondary school and above. Likewise, Acharya et al.
(2018), in their analysis of determinants of childhood stunting in Democratic Republic of
Congo, found a negative association between level of education and childhood stunting.
Similarly, Birhanu et al. (2017) found an increased mortality rate for infants and under-5
children whose parents were illiterate as well as a strong association between stunting and
parental education level. The number of stunted children was found to be four times higher
among illiterate mothers as compared to those from educated mothers (Birhanu et al., 2017).
Similarly, the likelihood of childhood stunting from illiterate fathers was eight times higher as
compared to those children who had educated fathers (Birhanu et al., 2017). Even though the
pathways are clearly understood, authors agree that parental education is a key determinant of
childhood nutritional status as it plays a central role by empowering both women and men to
offer 27 adequate childcare in terms of health, child feeding, and child education (Geberselassie
et al., 2018). Researchers noted a protective effect against stunting in children whose fathers
completed secondary school as compared to children whose fathers completed primary school
(Geberselassie et al., 2018). However, this protective effect might be missing in single mother
households. Clark and Hamplová (2013) explored single motherhood and child mortality in
sub-Saharan Africa and found that children born to never-married single mothers were
significantly more likely to die before age five in six countries as compared with children
whose parents were married (odds ratios range from 1.36 in Nigeria to 2.61 in Zimbabwe). In
nine countries, they found children of formerly married mothers were at a significantly higher
risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya) relative to children
having married parents (Clark & Hamplová, 2013). These findings are supported by Härkönen
(2018) who found a universal growing negative educational gradient of single motherhood
across countries. The negative educational gap promotes social inequality among single
mothers and their children who live in a cycle of poverty, and experience health related risks,
and poor educational achievement. Therefore, policy makers, public health organizations and
country governments should empower single mothers with better education and environmental
incentives to help ensure positive child health outcomes both at the community and individual
levels (Frost et al., 2005). Birth Interval 28 The relationship of birth interval and malnutrition
has been linked to maternal education. The general consensus among researchers is that chronic
malnutrition (stunting) is promoted by large sibling numbers and by later birth order.
Researchers in sub-Saharan Africa found boys to be more affected by chronic malnutrition
(stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found that among
mothers who were single, first-born boys were more exposed to childhood mortality and not
the first born girls’ counterpart. They found that sister and not brother played a protective role
towards lowering childhood mortality rate for second- born children. Similarly, Bukusuba et
al. (2017) in their case control study among children aged six and 59 months in Uganda found
that boys were significantly more likely to be stunted compared to girls (OR= 2.2, 95% CI:
1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of childhood stunting
for children born within a preceding birth interval less than 24 months. Equally, Liwin and
Houle (2019) found the risk of dying more than five times higher among children aged 12–23
months in the interval of zero to 11 months between the index child and the succeeding while
for a succeeding birth interval of 0–23 months the risk of dying among children aged 24–59
months increased by two-fold (Liwin & Houle, 2019). Economic Resources and Living
Conditions Linked to education level of parents and opportunity for employment is the
potential for improved household income which also plays a pivotal role in determining child
health outcome. Household income is tied to parental education and child health 29 and impacts
food intake through improved food purchasing power, quality of medication via its access and
affordability, and household sanitation (Umar Farooq et al., 2019). In their study in Cameroon
and DRC, Ntoimo and Odimegwu (2014) found higher probabilities of childhood stunting in
single mother households being significantly affected by economic resources and parental
education. Moreover, Mikalitsa (2015) while examining the intra household allocation,
household headship and nutrition of under-fives in western Kenya argued that better nutrition
status of children was not always linked to high income in a household, but rather to who
controls the income. The author found a linear association between improved childhood
nutritional status in households and active participation of women in decision making
(Mikalitsa, 2015). Households with uneducated parents were inclined to have low-income
levels. These parents spent less on nutritional adequate foods and their children were more
susceptible to growth failure due to lack of access to sufficient food of adequate quality as well
as poor living conditions, lack of access to basic health care services and greater exposure to
diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in their discourse on effect
of remittances on diet and nutrition found that remittances could lead to household food
security and underweight reduction even though remittances were found to play a little effect
on chronic undernourishment. In addition to the economic resources available to a household,
where individuals in a household reside the effect of the environs are important considerations
to stunting outcomes. In particular, is the recognized rural-urban inequality on childhood
stunting and that malnutrition in under-five children has been variably distributed according to
30 geographic regions. Kismul et al. (2018) when analyzing the determinants of stunting in the
Democratic Republic of Congo found a significant variability in the prevalence of stunting
between rural and urban regions with rural areas having a larger percentage of children living
with stunting than children living urban areas. In another study on environmental predictors of
stunting among under-5 children in Somalia, authors found variation in stunting distribution
spatially and temporally due to rains seasons and vegetation (Kinyoki et al., 2016). Similarly,
Stifel et al. (2018) in keeping with previous findings showed that rural populations were
characterized as having worse nutrition outcomes than those of urban populations. However,
they demonstrated that the nutritional inequality that exists between rural-urban regions stems
from differences in wealth, education, health, and non-road infrastructure services across rural
and urban areas (Stifel et al., 2018). Thus, reduction of childhood stunting requires
improvement of parental education, improvement of and access to healthy diets, and
improvement of household’s wealth as these are identified in the literature as the major
predictors of stunting (Headey et al., 2017). A component of the environment is water, but also
relevant to hygiene and sanitation and a key nutrient for health is access to safe water in rural
and urban communities. Childhood stunting has been negatively associated with poor access
to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al. (2014) estimated
that fewer than one in five people globally wash their hands with soap after defecation, Bain et
al. (2014) demonstrated the presence of disparities in access to safe water and sanitation
between rural and urban communities with access to both water and sanitation 31 services in
rural generally much lower than in urban areas, especially in low to middle income countries
(LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented complex direct
biological routes and many broader, less direct routes including the socioeconomic aspects
such as accessibility and affordability of water supplies and sanitation facilities through which
water supply, sanitation, and hygiene (WASH) could lead to stunting. Cumming et al. (2014)
also had shown that over onethird of the world’s population was deprived from domestic access
to safe water and sanitation. Aheto et al. (2015) in their study on determinants of malnutrition
in Ghana found that absence of toilet facilities in households was associated with increased risk
of malnutrition. Similarly, Haile et al. (2016) analyzed secondary data of the 2011 Ethiopian
Demographic and Health Survey (EDHS) and found lack of a better-quality latrine to be
associated with high probabilities of stunting from the community-level factors. Torlesse et al.
(2016) found a prevalence of 28.4% and 6.7% for stunting and severe stunting, respectively
among children in their study on determinants of stunting in Indonesia. After controlling for
potential covariates, they found an association between household sanitary facility and
household water treatment. They reported an increased adjusted odds of childhood stunting up
to more than three times in households that drank unprocessed water if the household utilized
a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-7.28, p < 0.001) as compared to households
that drank treated water and utilized unimproved latrine where the adjusted odds on child
stunting was not significantly higher (AOR = 1.27, 95 % CI: 0.99-1.63, p = 0.06; Torlesse et
al., 2016). 32 Health-seeking behavior hinges on the direct route between socio-economic
position (SEP) and health outcomes (Benova et al., 2014). Various factors that influence health
seeking behavior have been reported including socioeconomic and demographic
characteristics, perceived need, accessibility, and service availability (Sarker et al., 2016).
Uggla and Mace (2016) opined that parental investment in health-seeking behaviors was
independently influenced by the association between maternal factors (age, health, and marital
status) and child factors (birth order, health, sex, and age). The authors found higher odds of
investment among children with lower birth order, older mothers and mothers with better health
status. Similarly, scholars opined that higher maternal education and wealth, and better access
to health services were associated with higher levels of health investment in children. The
authors showed that monogamously married women had higher odds of investment than non-
married women in all health-seeking behaviours (Uggla & Mace, 2016). Therefore, the
Botswana Government policy makers should work in order to improve the primary health care
system, promote early health seeking behavior, and support to the single mothers in terms of
education and income resources so that these mothers could make sound decisions towards
their under-5 children. Contributing Causes of Stunting Key study variables identified as
contributing to stunting are specific to food and nutrition (food security, food diversity), health
seeking practices (breastfeeding behaviors), family structure, and emotional care and family
support systems. 33 Food insecurity and diversity: Food availability at the household level
plays a crucial role in childhood health. Mechanisms through which food insecurity leads to
stunting are inequality of food consumption in terms of quality, quantity, and continuity; and
inadequacy of infant and child feeding which in turn leads to food consumption issues (FAO,
2017). In their study, Shinsugi et al. (2015) opined that stunting and severe stunting were
associated with food insecurity. The authors found that moderately food insecure households
were more likely to have stunted children as compared to food secure households (OR = 1.26;
95% BCI: 1.04–1.51). Moreover, children who lived in moderately food insecure households
were more likely to be severely stunted than children living in food secure households (OR =
1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these findings are in accord with
previous works, Hagos et al. (2017) noted that wealth played no role in the relationship between
food insecurity and stunting. The late introduction of additional foods, the use of nutrient-poor
foods, or limited access to essential nutrients especially vitamin A, iron, and zinc were
recognized as contributing to higher stunting risk in food-insecure households (Moradi et al.,
2019). While some authors found that stunting was associated with household food insecurity,
Motbainor et al. (2015) documented that food security was not the only criterion to determine
childhood nutritional status, but rather other factors such as mother’s knowledge of child
nutrition and health care practices, maternal nutritional status, intrahousehold food allocation
and utilization practices, and access to health services and healthy environmental conditions
were to be taken into account. 34 Also of concern is food/dietary diversity to ensure an adequate
and balanced diet for energy, growth and development. Food diversity is the number of varied
foods or food groups consumed over a given reference period (Bukania et al., 2014).
Researchers in Ghana have explored the relationship between dietary diversity and child health
outcomes and found a general inclination of food diversification with increased child age
(Frempong & Annim, 2017). They noted a significant relationship between food diversity in
the health of children born to educated mothers (Frempong & Annim, 2017). Also, food
diversity and number of meals the child ate per day were significantly associated with stunting
(β = 0.039, p < 0.01) and underweight (β = 0.035, p < 0.05) respectively (Motbainor et al.,
2015). Food security and nutrition status of household members have been found to be
dependent on culture and institutional frameworks in either male headed household or female
headed household (Mikalitsa, 2015). Several authors found higher incidences of malnutrition
and food poverty in the de jure female-headed households as compared to male headed
households and de facto female headed households (Mikalitsa, 2015; Ragasa et al., 2019).
Ragasa et al. (2019) examined whether providing agriculture and nutritional information to
both mother and men had an impact on household food security. The authors found that the
quantity and quality of household food availability was determined by agricultural production
and income. Moreover, Ragasa et al. (2019) noted that in male-only and female-only headed
household’s food insecurity was on the high. Formal education and literacy levels and nutrition
education appeared to lead to improved food security in households with sole male adults (
headed by male) while food security in de 35 jure female headed households was driven by
landholdings, livestock units (especially poultry) having other assets, planting tree and root
crops, and nutrition-related informationResearchers have documented poor school uptake
among stunted children and a high risk of mortality and susceptibility to infections (Tariku et
al., 2017). Stunting has important economic consequences for both sexes at the individual,
household, and community level. Authors have shown an association between shorter adult
stature and labor-market outcomes such as lower earnings and poorer productivity (de Onis &
Branca, 2016). Increases in morbidity, mortality, and health expenditure and subsequent
reductions in human capital investment, physical capital investment, and labor supply are all
various pathways through which undernutrition could affect aggregate economic growth
(McGovern et al., 2017). Ill health leads to reductions 25 in productivity and could negatively
affect individuals’ and society’s economic growth, especially when a significant proportion of
a country’s population is affected by stunting (McGovern et al., 2017). Underlying Causes of
Stunting The key study variables identified as underlying causes for stunting are discussed in
terms of the study population’s economic resources, social determinants of health, and
environmental conditions. Parental Employment Parental employment status or the
occupational standing of the household head has been associated with child's nutrition status
(stunting). Rashad and Sharaf (2019) conducted a study to estimate the causal impact of
women’s work on child nutritional status as measured by the height-for-age score. The authors
found a negative correlation between under-5 nutritional status and their mother’s employment
and reduced odds of stunting for children whose parents (households’ heads) were farmers or
self-employed relative to households headed by housewives. Rashad and Sharaf (2019) argued
that it is because couples’ income combined could procure better services than those gained by
a single parent (mothers). Parental Education Parental education has been linked to childhood
mortality through different pathways and is often associated with the socioeconomic status of
the household. Frost et al. (2005) mentioned five pathways through which the linear
relationship between parental education and mortality among under-five children occur: (a)
socioeconomic 26 status, (b) health knowledge, (c) attitudes towards health care, (d) female
autonomy, and (e) reproductive behaviors. These pathways operate at both individual and
community levels. Mosley and Chen (1984), in their seminal work, observed a strong linear
relationship between mother’s educational achievement and reduction in infant and under-5
mortality in South Saharan African countries. Likewise, researchers found that high literacy
levels among mothers in a given country of sub-Saharan Africa promoted the utilization of
health care facilities and the lowering of risk for under-5 child mortality (Anyamele et al., 2017;
McTavish et al., 2010). Nkurunziza et al. (2017) found that children whose mothers had no
education were more likely to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely
stunted (cOR= 2.0; 95% CI: 1.3-2.9; p < 0.001) relative to those whose mothers who completed
secondary school and above. Likewise, Acharya et al. (2018), in their analysis of determinants
of childhood stunting in Democratic Republic of Congo, found a negative association between
level of education and childhood stunting. Similarly, Birhanu et al. (2017) found an increased
mortality rate for infants and under-5 children whose parents were illiterate as well as a strong
association between stunting and parental education level. The number of stunted children was
found to be four times higher among illiterate mothers as compared to those from educated
mothers (Birhanu et al., 2017). Similarly, the likelihood of childhood stunting from illiterate
fathers was eight times higher as compared to those children who had educated fathers (Birhanu
et al., 2017). Even though the pathways are clearly understood, authors agree that parental
education is a key determinant of childhood nutritional status as it plays a central role by
empowering both women and men to offer 27 adequate childcare in terms of health, child
feeding, and child education (Geberselassie et al., 2018). Researchers noted a protective effect
against stunting in children whose fathers completed secondary school as compared to children
whose fathers completed primary school (Geberselassie et al., 2018). However, this protective
effect might be missing in single mother households. Clark and Hamplová (2013) explored
single motherhood and child mortality in sub-Saharan Africa and found that children born to
never-married single mothers were significantly more likely to die before age five in six
countries as compared with children whose parents were married (odds ratios range from 1.36
in Nigeria to 2.61 in Zimbabwe). In nine countries, they found children of formerly married
mothers were at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to
1.75 in Kenya) relative to children having married parents (Clark & Hamplová, 2013). These
findings are supported by Härkönen (2018) who found a universal growing negative
educational gradient of single motherhood across countries. The negative educational gap
promotes social inequality among single mothers and their children who live in a cycle of
poverty, and experience health related risks, and poor educational achievement. Therefore,
policy makers, public health organizations and country governments should empower single
mothers with better education and environmental incentives to help ensure positive child health
outcomes both at the community and individual levels (Frost et al., 2005). Birth Interval 28
The relationship of birth interval and malnutrition has been linked to maternal education. The
general consensus among researchers is that chronic malnutrition (stunting) is promoted by
large sibling numbers and by later birth order. Researchers in sub-Saharan Africa found boys
to be more affected by chronic malnutrition (stunting) as opposed to girls. For example,
Guilbert and Marazyan (2018) found that among mothers who were single, first-born boys were
more exposed to childhood mortality and not the first born girls’ counterpart. They found that
sister and not brother played a protective role towards lowering childhood mortality rate for
second- born children. Similarly, Bukusuba et al. (2017) in their case control study among
children aged six and 59 months in Uganda found that boys were significantly more likely to
be stunted compared to girls (OR= 2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al.
(2018) found increased odds of childhood stunting for children born within a preceding birth
interval less than 24 months. Equally, Liwin and Houle (2019) found the risk of dying more
than five times higher among children aged 12–23 months in the interval of zero to 11 months
between the index child and the succeeding while for a succeeding birth interval of 0–23
months the risk of dying among children aged 24–59 months increased by two-fold (Liwin &
Houle, 2019). Economic Resources and Living Conditions Linked to education level of parents
and opportunity for employment is the potential for improved household income which also
plays a pivotal role in determining child health outcome. Household income is tied to parental
education and child health 29 and impacts food intake through improved food purchasing
power, quality of medication via its access and affordability, and household sanitation (Umar
Farooq et al., 2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014)
found higher probabilities of childhood stunting in single mother households being
significantly affected by economic resources and parental education. Moreover, Mikalitsa
(2015) while examining the intra household allocation, household headship and nutrition of
under-fives in western Kenya argued that better nutrition status of children was not always
linked to high income in a household, but rather to who controls the income. The author found
a linear association between improved childhood nutritional status in households and active
participation of women in decision making (Mikalitsa, 2015). Households with uneducated
parents were inclined to have low-income levels. These parents spent less on nutritional
adequate foods and their children were more susceptible to growth failure due to lack of access
to sufficient food of adequate quality as well as poor living conditions, lack of access to basic
health care services and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019).
Thow et al. (2016) in their discourse on effect of remittances on diet and nutrition found that
remittances could lead to household food security and underweight reduction even though
remittances were found to play a little effect on chronic undernourishment. In addition to the
economic resources available to a household, where individuals in a household reside the effect
of the environs are important considerations to stunting outcomes. In particular, is the
recognized rural-urban inequality on childhood stunting and that malnutrition in under-five
children has been variably distributed according to 30 geographic regions. Kismul et al. (2018)
when analyzing the determinants of stunting in the Democratic Republic of Congo found a
significant variability in the prevalence of stunting between rural and urban regions with rural
areas having a larger percentage of children living with stunting than children living urban
areas. In another study on environmental predictors of stunting among under-5 children in
Somalia, authors found variation in stunting distribution spatially and temporally due to rains
seasons and vegetation (Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with
previous findings showed that rural populations were characterized as having worse nutrition
outcomes than those of urban populations. However, they demonstrated that the nutritional
inequality that exists between rural-urban regions stems from differences in wealth, education,
health, and non-road infrastructure services across rural and urban areas (Stifel et al., 2018).
Thus, reduction of childhood stunting requires improvement of parental education,
improvement of and access to healthy diets, and improvement of household’s wealth as these
are identified in the literature as the major predictors of stunting (Headey et al., 2017). A
component of the environment is water, but also relevant to hygiene and sanitation and a key
nutrient for health is access to safe water in rural and urban communities. Childhood stunting
has been negatively associated with poor access to safe water and to hygienic toilet (Kismul et
al., 2018). While Freeman et al. (2014) estimated that fewer than one in five people globally
wash their hands with soap after defecation, Bain et al. (2014) demonstrated the presence of
disparities in access to safe water and sanitation between rural and urban communities with
access to both water and sanitation 31 services in rural generally much lower than in urban
areas, especially in low to middle income countries (LMIC; Bain et al., 2014). Cumming and
Cairncross (2016) documented complex direct biological routes and many broader, less direct
routes including the socioeconomic aspects such as accessibility and affordability of water
supplies and sanitation facilities through which water supply, sanitation, and hygiene (WASH)
could lead to stunting. Cumming et al. (2014) also had shown that over onethird of the world’s
population was deprived from domestic access to safe water and sanitation. Aheto et al. (2015)
in their study on determinants of malnutrition in Ghana found that absence of toilet facilities in
households was associated with increased risk of malnutrition. Similarly, Haile et al. (2016)
analyzed secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and
found lack of a better-quality latrine to be associated with high probabilities of stunting from
the community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related informationResearchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information Researchers have documented poor school uptake among stunted
children and a high risk of mortality and susceptibility to infections (Tariku et al., 2017).
Stunting has important economic consequences for both sexes at the individual, household, and
community level. Authors have shown an association between shorter adult stature and labor-
market outcomes such as lower earnings and poorer productivity (de Onis & Branca, 2016).
Increases in morbidity, mortality, and health expenditure and subsequent reductions in human
capital investment, physical capital investment, and labor supply are all various pathways
through which undernutrition could affect aggregate economic growth (McGovern et al.,
2017). Ill health leads to reductions 25 in productivity and could negatively affect individuals’
and society’s economic growth, especially when a significant proportion of a country’s
population is affected by stunting (McGovern et al., 2017). Underlying Causes of Stunting The
key study variables identified as underlying causes for stunting are discussed in terms of the
study population’s economic resources, social determinants of health, and environmental
conditions. Parental Employment Parental employment status or the occupational standing of
the household head has been associated with child's nutrition status (stunting). Rashad and
Sharaf (2019) conducted a study to estimate the causal impact of women’s work on child
nutritional status as measured by the height-for-age score. The authors found a negative
correlation between under-5 nutritional status and their mother’s employment and reduced odds
of stunting for children whose parents (households’ heads) were farmers or self-employed
relative to households headed by housewives. Rashad and Sharaf (2019) argued that it is
because couples’ income combined could procure better services than those gained by a single
parent (mothers). Parental Education Parental education has been linked to childhood mortality
through different pathways and is often associated with the socioeconomic status of the
household. Frost et al. (2005) mentioned five pathways through which the linear relationship
between parental education and mortality among under-five children occur: (a) socioeconomic
26 status, (b) health knowledge, (c) attitudes towards health care, (d) female autonomy, and (e)
reproductive behaviors. These pathways operate at both individual and community levels.
Mosley and Chen (1984), in their seminal work, observed a strong linear relationship between
mother’s educational achievement and reduction in infant and under-5 mortality in South
Saharan African countries. Likewise, researchers found that high literacy levels among mothers
in a given country of sub-Saharan Africa promoted the utilization of health care facilities and
the lowering of risk for under-5 child mortality (Anyamele et al., 2017; McTavish et al., 2010).
Nkurunziza et al. (2017) found that children whose mothers had no education were more likely
to be stunted (cOR = 2.3; 95% CI: 1.7-3; p < 0.001) and severely stunted (cOR= 2.0; 95% CI:
1.3-2.9; p < 0.001) relative to those whose mothers who completed secondary school and
above. Likewise, Acharya et al. (2018), in their analysis of determinants of childhood stunting
in Democratic Republic of Congo, found a negative association between level of education and
childhood stunting. Similarly, Birhanu et al. (2017) found an increased mortality rate for infants
and under-5 children whose parents were illiterate as well as a strong association between
stunting and parental education level. The number of stunted children was found to be four
times higher among illiterate mothers as compared to those from educated mothers (Birhanu et
al., 2017). Similarly, the likelihood of childhood stunting from illiterate fathers was eight times
higher as compared to those children who had educated fathers (Birhanu et al., 2017). Even
though the pathways are clearly understood, authors agree that parental education is a key
determinant of childhood nutritional status as it plays a central role by empowering both
women and men to offer 27 adequate childcare in terms of health, child feeding, and child
education (Geberselassie et al., 2018). Researchers noted a protective effect against stunting in
children whose fathers completed secondary school as compared to children whose fathers
completed primary school (Geberselassie et al., 2018). However, this protective effect might
be missing in single mother households. Clark and Hamplová (2013) explored single
motherhood and child mortality in sub-Saharan Africa and found that children born to never-
married single mothers were significantly more likely to die before age five in six countries as
compared with children whose parents were married (odds ratios range from 1.36 in Nigeria to
2.61 in Zimbabwe). In nine countries, they found children of formerly married mothers were
at a significantly higher risk of dying (odds ratios range from 1.29 in Zambia to 1.75 in Kenya)
relative to children having married parents (Clark & Hamplová, 2013). These findings are
supported by Härkönen (2018) who found a universal growing negative educational gradient
of single motherhood across countries. The negative educational gap promotes social inequality
among single mothers and their children who live in a cycle of poverty, and experience health
related risks, and poor educational achievement. Therefore, policy makers, public health
organizations and country governments should empower single mothers with better education
and environmental incentives to help ensure positive child health outcomes both at the
community and individual levels (Frost et al., 2005). Birth Interval 28 The relationship of birth
interval and malnutrition has been linked to maternal education. The general consensus among
researchers is that chronic malnutrition (stunting) is promoted by large sibling numbers and by
later birth order. Researchers in sub-Saharan Africa found boys to be more affected by chronic
malnutrition (stunting) as opposed to girls. For example, Guilbert and Marazyan (2018) found
that among mothers who were single, first-born boys were more exposed to childhood mortality
and not the first born girls’ counterpart. They found that sister and not brother played a
protective role towards lowering childhood mortality rate for second- born children. Similarly,
Bukusuba et al. (2017) in their case control study among children aged six and 59 months in
Uganda found that boys were significantly more likely to be stunted compared to girls (OR=
2.2, 95% CI: 1.1-4.2; P < .05). Moreover, Kismul et al. (2018) found increased odds of
childhood stunting for children born within a preceding birth interval less than 24 months.
Equally, Liwin and Houle (2019) found the risk of dying more than five times higher among
children aged 12–23 months in the interval of zero to 11 months between the index child and
the succeeding while for a succeeding birth interval of 0–23 months the risk of dying among
children aged 24–59 months increased by two-fold (Liwin & Houle, 2019). Economic
Resources and Living Conditions Linked to education level of parents and opportunity for
employment is the potential for improved household income which also plays a pivotal role in
determining child health outcome. Household income is tied to parental education and child
health 29 and impacts food intake through improved food purchasing power, quality of
medication via its access and affordability, and household sanitation (Umar Farooq et al.,
2019). In their study in Cameroon and DRC, Ntoimo and Odimegwu (2014) found higher
probabilities of childhood stunting in single mother households being significantly affected by
economic resources and parental education. Moreover, Mikalitsa (2015) while examining the
intra household allocation, household headship and nutrition of under-fives in western Kenya
argued that better nutrition status of children was not always linked to high income in a
household, but rather to who controls the income. The author found a linear association
between improved childhood nutritional status in households and active participation of
women in decision making (Mikalitsa, 2015). Households with uneducated parents were
inclined to have low-income levels. These parents spent less on nutritional adequate foods and
their children were more susceptible to growth failure due to lack of access to sufficient food
of adequate quality as well as poor living conditions, lack of access to basic health care services
and greater exposure to diseases (Mikalitsa, 2015; Ragasa et al., 2019). Thow et al. (2016) in
their discourse on effect of remittances on diet and nutrition found that remittances could lead
to household food security and underweight reduction even though remittances were found to
play a little effect on chronic undernourishment. In addition to the economic resources
available to a household, where individuals in a household reside the effect of the environs are
important considerations to stunting outcomes. In particular, is the recognized rural-urban
inequality on childhood stunting and that malnutrition in under-five children has been variably
distributed according to 30 geographic regions. Kismul et al. (2018) when analyzing the
determinants of stunting in the Democratic Republic of Congo found a significant variability
in the prevalence of stunting between rural and urban regions with rural areas having a larger
percentage of children living with stunting than children living urban areas. In another study
on environmental predictors of stunting among under-5 children in Somalia, authors found
variation in stunting distribution spatially and temporally due to rains seasons and vegetation
(Kinyoki et al., 2016). Similarly, Stifel et al. (2018) in keeping with previous findings showed
that rural populations were characterized as having worse nutrition outcomes than those of
urban populations. However, they demonstrated that the nutritional inequality that exists
between rural-urban regions stems from differences in wealth, education, health, and non-road
infrastructure services across rural and urban areas (Stifel et al., 2018). Thus, reduction of
childhood stunting requires improvement of parental education, improvement of and access to
healthy diets, and improvement of household’s wealth as these are identified in the literature
as the major predictors of stunting (Headey et al., 2017). A component of the environment is
water, but also relevant to hygiene and sanitation and a key nutrient for health is access to safe
water in rural and urban communities. Childhood stunting has been negatively associated with
poor access to safe water and to hygienic toilet (Kismul et al., 2018). While Freeman et al.
(2014) estimated that fewer than one in five people globally wash their hands with soap after
defecation, Bain et al. (2014) demonstrated the presence of disparities in access to safe water
and sanitation between rural and urban communities with access to both water and sanitation
31 services in rural generally much lower than in urban areas, especially in low to middle
income countries (LMIC; Bain et al., 2014). Cumming and Cairncross (2016) documented
complex direct biological routes and many broader, less direct routes including the
socioeconomic aspects such as accessibility and affordability of water supplies and sanitation
facilities through which water supply, sanitation, and hygiene (WASH) could lead to stunting.
Cumming et al. (2014) also had shown that over onethird of the world’s population was
deprived from domestic access to safe water and sanitation. Aheto et al. (2015) in their study
on determinants of malnutrition in Ghana found that absence of toilet facilities in households
was associated with increased risk of malnutrition. Similarly, Haile et al. (2016) analyzed
secondary data of the 2011 Ethiopian Demographic and Health Survey (EDHS) and found lack
of a better-quality latrine to be associated with high probabilities of stunting from the
community-level factors. Torlesse et al. (2016) found a prevalence of 28.4% and 6.7% for
stunting and severe stunting, respectively among children in their study on determinants of
stunting in Indonesia. After controlling for potential covariates, they found an association
between household sanitary facility and household water treatment. They reported an increased
adjusted odds of childhood stunting up to more than three times in households that drank
unprocessed water if the household utilized a unimproved latrine (AOR = 3.47, 95 % CI: 1.73-
7.28, p < 0.001) as compared to households that drank treated water and utilized unimproved
latrine where the adjusted odds on child stunting was not significantly higher (AOR = 1.27, 95
% CI: 0.99-1.63, p = 0.06; Torlesse et al., 2016). 32 Health-seeking behavior hinges on the
direct route between socio-economic position (SEP) and health outcomes (Benova et al., 2014).
Various factors that influence health seeking behavior have been reported including
socioeconomic and demographic characteristics, perceived need, accessibility, and service
availability (Sarker et al., 2016). Uggla and Mace (2016) opined that parental investment in
health-seeking behaviors was independently influenced by the association between maternal
factors (age, health, and marital status) and child factors (birth order, health, sex, and age). The
authors found higher odds of investment among children with lower birth order, older mothers
and mothers with better health status. Similarly, scholars opined that higher maternal education
and wealth, and better access to health services were associated with higher levels of health
investment in children. The authors showed that monogamously married women had higher
odds of investment than non-married women in all health-seeking behaviours (Uggla & Mace,
2016). Therefore, the Botswana Government policy makers should work in order to improve
the primary health care system, promote early health seeking behavior, and support to the single
mothers in terms of education and income resources so that these mothers could make sound
decisions towards their under-5 children. Contributing Causes of Stunting Key study variables
identified as contributing to stunting are specific to food and nutrition (food security, food
diversity), health seeking practices (breastfeeding behaviors), family structure, and emotional
care and family support systems. 33 Food insecurity and diversity: Food availability at the
household level plays a crucial role in childhood health. Mechanisms through which food
insecurity leads to stunting are inequality of food consumption in terms of quality, quantity,
and continuity; and inadequacy of infant and child feeding which in turn leads to food
consumption issues (FAO, 2017). In their study, Shinsugi et al. (2015) opined that stunting and
severe stunting were associated with food insecurity. The authors found that moderately food
insecure households were more likely to have stunted children as compared to food secure
households (OR = 1.26; 95% BCI: 1.04–1.51). Moreover, children who lived in moderately
food insecure households were more likely to be severely stunted than children living in food
secure households (OR = 1.31; 95% BCI: 1.04–1.64; Shinsugi et al., 2015). While these
findings are in accord with previous works, Hagos et al. (2017) noted that wealth played no
role in the relationship between food insecurity and stunting. The late introduction of additional
foods, the use of nutrient-poor foods, or limited access to essential nutrients especially vitamin
A, iron, and zinc were recognized as contributing to higher stunting risk in food-insecure
households (Moradi et al., 2019). While some authors found that stunting was associated with
household food insecurity, Motbainor et al. (2015) documented that food security was not the
only criterion to determine childhood nutritional status, but rather other factors such as
mother’s knowledge of child nutrition and health care practices, maternal nutritional status,
intrahousehold food allocation and utilization practices, and access to health services and
healthy environmental conditions were to be taken into account. 34 Also of concern is
food/dietary diversity to ensure an adequate and balanced diet for energy, growth and
development. Food diversity is the number of varied foods or food groups consumed over a
given reference period (Bukania et al., 2014). Researchers in Ghana have explored the
relationship between dietary diversity and child health outcomes and found a general
inclination of food diversification with increased child age (Frempong & Annim, 2017). They
noted a significant relationship between food diversity in the health of children born to
educated mothers (Frempong & Annim, 2017). Also, food diversity and number of meals the
child ate per day were significantly associated with stunting (β = 0.039, p < 0.01) and
underweight (β = 0.035, p < 0.05) respectively (Motbainor et al., 2015). Food security and
nutrition status of household members have been found to be dependent on culture and
institutional frameworks in either male headed household or female headed household
(Mikalitsa, 2015). Several authors found higher incidences of malnutrition and food poverty in
the de jure female-headed households as compared to male headed households and de facto
female headed households (Mikalitsa, 2015; Ragasa et al., 2019). Ragasa et al. (2019)
examined whether providing agriculture and nutritional information to both mother and men
had an impact on household food security. The authors found that the quantity and quality of
household food availability was determined by agricultural production and income. Moreover,
Ragasa et al. (2019) noted that in male-only and female-only headed household’s food
insecurity was on the high. Formal education and literacy levels and nutrition education
appeared to lead to improved food security in households with sole male adults ( headed by
male) while food security in de 35 jure female headed households was driven by landholdings,
livestock units (especially poultry) having other assets, planting tree and root crops, and
nutrition-related information
