Determinants of undernutrition among children under-five years old in southern Ethiopia: does pregnancy intention matter? A community-based unmatched case-control study

Despite the considerable progress made during the past decade in reducing its burden, child undernutrition, which includes conditions such as stunting, underweight, and wasting, contributes to approximately one-half of all childhood mortalities. It is still a major public health problem, particularly in resource-poor countries [1, 2]. Stunting, or a small height for one’s age, describes decelerated or arrested linear growth, and it is an indicator of chronic malnutrition [3]. It mainly develops as a result of prolonged food deprivation, or due to a chronic disease or illness [1].

Regardless of the importance of early childhood nutrition for survival and long-term development, there is no consensus among the global nutrition communities on how to best combat child undernutrition [4, 5]. Tackling the challenge of undernutrition requires cross-sector collaboration, innovative approaches, and optimizing the use of all available interventions [6]. A large number of children under 5 years old die each year from malnutrition-related causes, representing nearly one-half of all the deaths in this age group [2]. Moreover, the survivors also exhibit impaired physical growth and intellectual development, which ultimately reduce their adulthood productivity. On a grander scale, this prevents some nations from achieving their full economic potentials [6].

Globally, an estimated 165 million children under 5 years old (26%) are stunted, 52 million (8%) exhibit wasting, and 100 million (16%) are underweight [7]. Malnutrition contributes to 11% of the disability-adjusted life years (healthy years of life) lost worldwide [8]. In Ethiopia, malnutrition in general, and malnutrition in children under 5 years old in particular, are significant health challenges. Moreover, 44% of the children are stunted, 10% exhibit wasting, and 29% are underweight in Ethiopia [9]. Stunting was estimated to cause a loss of approximately 44 billion Ethiopian Birr from 2005 to 2015 [10].

Although stunting often begins in utero, several research studies have identified socio-demographic, socioeconomic [11‐16], environmental [17, 18], dietary [12, 18, 19], parasitic infection and other related illness [11, 20], and psychological [19] correlations with stunting. One of the psychological factors that might be associated with children stunting is unintended pregnancy, either unwanted (the parent did not desire any or any more children) or mistimed (the pregnancy occurred earlier than desired) [16, 21, 22]. Evidence has suggested that family planning (FP), a proxy indicator of one’s pregnancy intention, can have a significant influence on achieving key nutrition outcomes. Previous research has shown that when planned pregnancies occur, the feeding practices (including breastfeeding) are improved, resulting in an improvement in the nutritional outcomes of the children [6]. There is also ample evidence of the link between unwanted pregnancies and adverse neonatal outcomes and behaviors, including a low birth weight, neonatal mortality, the absence of breast-feeding, and poor parental care [23, 24]. A study conducted by Marston and Cleland in Peru found a 15% greater risk of stunting among children from unwanted pregnancies when compared to those from wanted pregnancies. However, the findings of a study from Egypt were contrary, in that the likelihood of stunting was lower if a pregnancy was mistimed or unwanted than if it was wanted [24]. Barber and colleagues hypothesized that various pathways linked unwanted childbearing, child health, and mother-child relationships. They suggested that children who were unwanted at the time of conception may face more neglect and abuse than those who were wanted [25].

Even though there has been a trend toward a drop in the global pregnancy rates, the proportion of unintended pregnancies remains high, particularly in the developing world [6]. During 2012, approximately 40% of the pregnancies worldwide (or 85 million pregnancies) were estimated to be unintended. Moreover, studies conducted in different regions of Ethiopia revealed unintended pregnancy rates ranging between 27.9 and 42.4% [26‐28], and nearly one million unintended pregnancies are expected to occur each year in Ethiopia [29]. Unintended pregnancies are of concern from both a human rights perspective and a public health perspective, and the consequences of unintended pregnancies are serious. They impose considerable burdens on children, women, men, and families [30].

Such high rates of unintended pregnancy and its possible series repercussion justify a study of their negative consequences on child nutrition and growth, especially when considering that adverse nutritional outcomes in a child are more likely to occur if the pregnancy was unintended. In addition, the study area is one of the areas with the highest burden of malnutrition [31] and very poor family planning utilization in the country [32]. This study hypothesized that high number of unintended pregnancies that could be resulted from poor family planning in the study area might be contributed for such significant burden of under-five malnutrition Therefore, the study was designed to assess the association of unintended pregnancies and other family and child characteristics with the nutritional status of children, particularly with regard to stunting among children under 5 years old. There are no studies showing such side of the factors contributing for the burden of malnutrition.

Apart from the usual predictors of nutritional status of children, reproductive health related factors particularly, pregnancy intention, was identified to significantly affect stunting among children. In this study, children whose conception had been unintended were significantly at greater risk of stunting than children whose conception was intended. Accordingly, children from unintended pregnancies were about three fold more likely to be stunted compared to those from intended pregnancies. This finding is also supported by finding from Bolivia [16], and finding from Bangladesh Demographic and Health Survey [22] which reported higher proportion of stunting among children born from unintended pregnancies. This might be due to the fact that, children born from unintended pregnancies were more likely to be exposed to conscious or unconscious neglect, resulting in inappropriate feeding, poor child-mother bond and poor health care seeking for the child. In addition, there might be poor economical and psychological preparation/capability for rearing and feeding of the child, which could substantially affect nutritional status of the child.

The household wealth of the family of the subject children was associated with stunted growth in this study. This finding is in line with other findings from analysis of the Mini Ethiopian Demographic and Health Survey data [15], a cross sectional study conducted on prevalence and determinants malnutrition in Bangladesh [13], and another study from Sindh, Pakistan [14]. The primary influence of household wealth is on child feeding practices related to an inadequate amount of food and the nutritional content of meals, which affect the nutritional status of children. This implies that economic empowerment for households in developing countries may be still an issue that need due attention.

Among the socio-economic factors, the educational status of the father of children was found to be correlated with stunting among children. Children of illiterate fathers were at more than three times higher risk of stunting compared to educated fathers. This finding is also in line with the findings of the study from Bangladesh [13] and other studies [32‐36]. Possibly, this might be due to the fact that households with educated fathers have a greater likelihood of having a higher income, since fathers are the main source of income for the households studied. This may also indicate educated fathers have a higher awareness and are more conscience of the nutritional needs of children. Other studies in other locations [11, 15] found that the educational status of the mother was associated with stunting. This was not the case in our study. This might be happened as a result of chance, otherwise we couldn’t find possible reason from the characteristics of the study population.

The other variable found as a predictor for stunting in this study was minimum meal frequency. Children who hadn’t received the recommended daily meal frequency were 4.5 times more likely to be stunted. This finding is also consistent with the result of a study from Burkina Faso [37]. This finding could be explained by the reality that if the child was not getting the complementary food (for breast-feeding children) or regular daily meals (for non-breast-feeding children) as recommended, the possibility of getting adequate nutrient intake will be diminished, which may significantly affect the child’s nutritional status. However, other studies, including those in other areas of Ethiopia, found no association between a child’s daily meal frequency and stunting [12, 38]. This could be due to the difference in the study population and period across the study areas and/or due to the difference in methods of assessment of feeding pattern between the studies.

This study tried to assess the effect of numerous variables, including reproductive health-related information, on stunting among children. As it is a community-based study, cases and controls were selected randomly from the community to address the issue of representation. Though the study has valuable strength, we admit certain important limitations. The main limitation of this study was that while assessing pregnancy intention, we only used information from the mother and didn’t use paternal data. Another limitation is that pregnancy intention was assessed based on retrospective recall of the pregnancy’s status, which could affect the mother’s response. The other considerable limitations of this study are its small sample size, single center (study area), and exclusion of some important variables like dietary diversity.

Untended pregnancy was found to be an important predictor of stunting in this study, where children from untended conception were at greater risk of stunting. Considerable number of stunted children were from untended pregnancy. Similarly the educational status of the father, the households’ wealth index and the daily meal frequency were significantly associated with children stunting. Children from a poorer households, children whose fathers were illiterate and children with inadequate minimum meal frequency were at greater risk of being stunted. Hence, preventing unintended pregnancy could be a valuable intervention modality for the prevention of stunting among children. Further studies to determine the mechanism through which unintended pregnancies could contribute for child malnutrition should be considered.