This article reports findings of a study that investigated the factors that contribute to high infant mortality rates in rural areas in sub-Saharan Africa, focusing in the Kingdom of Eswatini. Infant mortality rate is a key indicator used to measure progress in the development of any country. Global development is currently underpinned by the Sustainable Development Goals (SDGs). The third SDG is to ensure healthy lives and well-being for all at all ages. One of the specific targets of this SDG is that “by 2030, end preventable deaths of new-borns and children under 5 years of age, with all countries aiming to reduce neonatal mortality to at least as low as 12 per 1,000 live births and under-5 mortality to at least as low as 25 per 1,000 live births” (https://sustainabledevelopment.un.org/sdg3). Interestingly, progress is being made towards this target because the global number of under-5 deaths continues to drop, having dropped to 5.4 million in 2017 from 9.8 million in 2000 (ibid). However, in developing countries, the number of under-5 deaths, more especially in the first few months of life (infant mortality), remains high. Globally, 2.6 million newborns died in 2016, with sub-Saharan Africa constituting 38 per cent [1]. Mostly from preventable causes such as pneumonia, diarrhea and malaria. In almost half of such cases, malnutrition plays a role, while unsafe water, sanitation and hygiene are also signi?cant contributing factors [2]. This indicates that progress towards attaining SDG3 is linked to progress made in all other SDGs focused on social development: SGD 1 “no poverty”, SDG2 “zero hunger”, SDG6 “clean water and sanitation”, SDG4 “quality education”, SDG5 “gender equality” and SDG10 “reduced inequalities”. Recent evaluation and monitoring of progress reports indicate that the performance of different countries on the implementation of these SDGs differs widely depending on their diverse policy approaches and resource endowment [3]. The Kingdom of Eswatini, like all other countries, is making progress in the implementation of some of the SDGs but lags far behind in others. Eswatini is a small landlocked country in southern Africa with high levels of rural poverty and its economy relies mainly on sugar cane production as the largest industry [4]. The current population of Eswatini stands at 1,093,238 people [5]. As per the Sustainable Development Report, Eswatini is making progress in addressing hunger (SDG2) but is still struggling to end poverty (SGD 1) as only 20.7% of the population is reported to be undernourished, whilst 40.5% of the population is in poverty [6]. Measures to address hunger appear to be less aligned to the long-term goal of eradicating poverty. Over half of the population in Eswatini required livelihood support, mainly in the form of food aid due to the on-going El Niño drought [7]. States that “the persistence of the food crisis in Eswatini has prompted concerns that prolonged provision of food aid could be developing dependency tendencies” [8]. Temporary measures to fight hunger such as provision of food aid during drought periods fail to reduce maternal mortality rate, which stands at 389/100, 000 live births [6]. Despite substantial investment in the health sector over the years, Eswatini has struggled to keep pace with the rising disease burden [9]. According to Mhlongo et al. limited progress has been made in the reduction of infant mortality rates in the country. Nonetheless, changes in socio-economic, environmental, and maternal conditions in any given country have long been known to influence infant mortality rates [10, 11]. Using Zombodze Emuva as a case study, this article investigates the factors contributing to high infant mortality in rural areas in the Kingdom of Eswatini. Specifically, the paper explores the socio-economic status of households, to identify maternal and environmental factors impacting the living standards of households, and to ascertain the impact of these factors on the likelihood of infant mortality in the area. Structurally, the article first provides an overview of literature on socio-economic, environmental and maternal factors influencing infant mortality in developing countries. These determinants of infant mortality do not operate independently, but are intertwined and reinforce each other, resulting in high cumulative effect. The article moves on to provide a detailed description of the methodological approach in the study, including a brief description of the study area. It then presents and discusses the findings of the study in relation to the literature. The article concludes by highlighting potential determinants of infant mortality in Zombodze Emuva and then makes recommendations on how they can be addressed.

Infant mortality, just like any other concept, has been defined differently by different scholars. Defines infant mortality as the number of deaths of infants less than one year of age per 1,000 live births. Infant mortality is defined as the total number of infant deaths per 1,000 live births in a specified period of time [12, 13]. To Sullivan and Tureeva  infant mortality is the probability of dying between birth and the first birthday[14]. Infant mortality is here defined as the number of children born in a country and dying prior to their first birthday. It is determined by dividing the total number of new-borns by the total population of that particular country and multiplying that by a 1,000 [15]. The factors influencing infant mortality are diverse and they differ between and within countries. This diversity is made easy by classifying factors based on their similarities and differences. The literature reviewed highlighted three main classifications of factors influencing infant mortality, namely, socio-economic factors, environmental factors, and maternal factors. As defined from a sociological perspective, socio-economic factors are those that relate to the social standing or class position of an individual or group of people within a given society [16]. Maternal factors entail internal (endogenous) personal characteristics of an individual. On the contrary, environmental factors entail conditions of the external (exogenous) surroundings of an individual that have an impact on his/her daily living. The section below discusses few selected determinants of infant mortality under each of these three broad classifications. However, it should be emphasized that there is an overlap between these three classifications of the determinants of infant mortality because the factors are connected and they reinforce each other.

Socio-economic factors such as maternal education, household income, place of residence and place of delivery influence infant mortality in most countries of the global South. Maternal education is a social determinant of infant mortality in that access to education is a fundamental human right that needs not to be denied. Apart from this legal context of education, literate women are knowledgeable about proper childcare and feeding practices. They also have proper knowledge and understanding of the importance of prenatal care and immunization services for the health and wellbeing of their infants [17]. Contrary to this, being illiterate has a negative impact on infant mortality in that the women are not knowledgeable about hygiene, and as such, they end up raising their children in unhygienic conditions which eventually lead to their death [17]. The highest levels of infant mortality in most counties in the global South are recorded on mothers with no formal education than on educated mothers [18]. In Eswatini, mothers who obtained secondary education had a reduction of 59% chances of infant mortality, while those who had obtained tertiary education had an average reduction of 74% [19].

Rutstein  puts an emphasis on the importance of mothers’ education and the place of residence and cite these as some of the socio-economic determinants of infant and child survival[20]. Changes in a mother’s socio-economic status due to education may lead to good use of health centres, improved child care practices, better food preservation and household cleanliness, among others [20]. Education heightens the mother’s ability to make use of government and private health care resources and it may increase the autonomy necessary to advocate for her child in the household and the outside world [19]. Formal education enables a mother to earn an income and make good use of health care facilities at her disposal and it improves the mother’s self-esteem, coping ability and skills of resource mobilization for herself and her children [21]. It also equips the mother with increased powers of negotiation and ability to make independent decisions in matters of child bearing in general and child spacing in particular, both of which affect infant survival.

Directly linked to the impact of educational level on infant mortality is the employment status of women and the overall income level of their households. According to Mutunga, infants of wealthier families have better housing conditions, better feeding patterns and access to medical care, and as such, are expected to live longer than those of poor families [22]. Wealth status determines the availability of nutritional resources, which is especially important because once infants reach the age of 6 months, they can no longer depend on nourishment from breast milk alone [19]. In Kenya, lower mortality rates have been observed in households with iron sheets and tiles as roofing materials and those with household assets such as radio and television [22]. In Eswatini 2017 census shows that children born to working women experiences lower mortality of 98 per thousand than children born to non-working women which is 117 per 1, 000 [5]. Despite rural women in the global South countries bearing much of the burden of providing for their households, they are often excluded from gainful employment [23]. Women constitute over 50% of Africa’s population but their rate of unemployment is high compared to their male counterparts. In Eswatini, women constitute 24% of the total unemployment level, and rural unemployment stands 15.8% compared 6.8% in urban areas [5]. Furthermore, women in most countries of the global South disproportionately engage in unstable and risky informal sector activities such as street vending to earn an income and are predisposed to determinants of infant mortality.

It is undoubted that the level of income determines place of residence for most households, particularly those in the low income bracket. Although Africa is currently experiencing rapid rates of urbanization due to multiplicity of factors, but mainly due to rural-urban migration, its population is still predominately rural [24]. In Eswatini, as of 2017, about 76.2 percent of the population is rural but there are strong ties between rural and urban due to dependency on employment as the major source of income and livelihoods [25]. According to Yi et al. infant and child mortality rates are higher in rural areas than in urban areas because the former is characterized by poor health infrastructure and low levels of income [26]. It has also been observed that people in rural areas find it difficult to access health services when they are sick, hence increasing their chances of dying [27]. The World Health Organization recommended that pregnant women must have their first contact with their doctors in the first 12 weeks’ gestation, 20, 26, 30, 34, 36, 38 and 40 weeks’ gestation [28]. This is because it facilitates the uptake of preventative measures, timely detection risks, reduces complications and addresses health inequalities. These visits provide counselling about healthy eating, optimal nutrition and what vitamins or minerals women should take during pregnancy which helps in the development of the baby throughout the pregnancy. Infant mortality differentials by place of residence (rural-urban) are expected due to regional differences in health infrastructure, and communication and disease prevalence conditions [28].

Almost all African countries, including the Kingdom of Eswatini, have unequal spatial distribution of resources. This inequitable distribution of resources is a contributing factor to infant mortality rates. There is a shortage of health facilities in developing countries, particularly in rural and impoverished peri-urban and informal areas [29]. Although currently in Eswatini, 80 percent of the population is within a radius of 8 kilometres of health facility [24]. Transport and communication networks in rural areas are poor, making accessibility of such facilities difficult. In most cases, available facilities are not adequately equipped and they lack skilled workforce such as nurses, midwifes and pediatricians [30]. Although the numbers have declined in recent times, many women in rural Africa still give birth at home, using traditional methods that contribute to maternal and infant mortality rates [31]. In Eswatini, maternal mortality remains high at 593 per 100,000 live births as most health facilities do not have Ante Natal Care and Post Natal Care guidelines [24]. Children delivered in modern health facilities usually exhibit lower rates of mortality [28]. However, in some cases, mortality among children delivered in modern facilities is observed to be higher because mothers use these facilities mostly when they have pregnancy complications (ibid). This is related to level of education of expecting women, which determines their health behaviours such as in relation to pre and postnatal care, hygiene and other aspects [32].

Maternal factors entail personal characteristics such as mother’s age at first birth, birth order, preceding birth interval or sequencing of children, hypertensive disorders, obstructed labour, child’s sex, antenatal care, pre-natal care, and type of birth [33]. These factors play an important role in influencing infant mortality in developing countries. Age of the mother plays a major role in the survival of an infant. Despite progress made in inclusive education, school dropout due to adolescent childbearing is still high in developing countries [34,35]. In Eswatini, 87 of every 1000 school going girls fall pregnant and drop out of school every year [7]. Despite the introduction of free primary education in Eswatini over a decade ago, there still exist several inefficiencies within this sub-sector that tend to drive learners out of the system [24]. Young mothers aged below 20 years normally tend to have biological, emotional, social and economic problems that frustrate their entire childbearing process [38]. Likewise, women who began childbearing early would not tolerate the stresses of all activities that go with childbearing such as pregnancy, delivery and breastfeeding. Maternal age has long been noted to be greatly affecting the survival of a child, mostly during the first 12 months of life and this effect declines with the growth of the child [37]. Young mothers have difficulties in providing enough food for their children since their influence on allocation of household resources is very limited [36]. Full antenatal care has always been beneficial to motherhood; it safeguards the mother, keeps her healthy and thereby directly and indirectly leaves positive influence on the birth and survival of an infant [39].

Although fertility rates have been declining in recent decades, Pison reports that “in Africa, women have 4.5 children on average while in Asia the figure is 2.1 children, in Latin America 2.0, in North America 1.9 and in Europe 1.6”[40]. The more the number of children bore, the more infectious diseases and greater competition for resources and childcare within the household [41]. Short preceding birth intervals affect the survival of the index child due to maternal depletion [10]. The mother has not had enough time to recover from the previous birth and so, her next birth will be very weak with low birth weight [26]. Short birth intervals necessitate the premature weaning of the index child. The latter is exposed to malnutrition and increased probability of contracting infectious and parasitic diseases due to lack of attention from the mother [42]. Zwane notes that previous studies have shown that short birth intervals (less than or equal to 18 months), high parity (6 or more children), low maternal age (less than 20 years) and high maternal age (35 and more years) adversely impact infant and child survival [28]. 

According to the World Bank, environmental health risks fall into two broad categories. The first are the traditional hazards related to poverty and lack of development, such as lack of safe water, inadequate sanitation and waste disposal, indoor air pollution, and vector-borne diseases [36]. The second category is the modern hazards such as urban air pollution and exposure to agro-industrial chemicals and wastes that are caused by development that lacks environmental safeguards. According to the World Health Organization, estimates for 2020, about 8.9 per cent of people in the world openly defecate without any toilet or latrine; the global impact of poor sanitation on infant and child death and health is profound. In Eswatini, the percentage of the population with sustainable access to basic sanitation is 52.84 per cent in urban areas and 30.34 per cent in rural areas and 11 per cent nationally practice open defecation [43].

Environmental factors that contribute to infant mortality include the number of occupants per room, sources of water supply, cleanliness of feeding bottles and utensils, methods of defecation used and methods of solid waste disposal [44]. Infants born in households with access to piped water or a public tap have an infant mortality rate of 25 per cent lower than infants in households using surface water, open wells, lakes and rivers [45]. According to World Health Organization, households with piped water were associated with 35 per cent reduction in risk of infant live births compared to that in a house whose source of drinking water is from a river or stream [46]. An expecting mother that is exposed to environmental chemicals risks the foetus and infant to mortality because they are more susceptible to chemicals than adults due to their fast growth, immature and defenseless mechanisms [47]. During pregnancy, a placenta may accumulate toxic chemicals, which may result to reduced blood flow, deterioration of the transport of nutrients and fetal growth (ibid). Godson & Nnamdi found that unsterilized feeding bottles and utensils are significantly related to the high incidence of diarrhoea [48]. High infant mortality signifies the lack of proper childcare owing to societal preferences (such as the affinity for a male child) [49]. One reason for gender differences in child mortality is a preference for sons, and after the first month of life other factors come into play, including environmental and behavioural factors such as care-seeking practices [50]. Females are often brought to health facilities in more advanced stages of illness than a boy child (ibid).

Description of the Study Area and Research Design

The study was conducted in the rural village of Zombodze Emuva in the Shiselweni region of the Kingdom of Eswatini (Figure 1).

Figure 1: Map of Zombodze Emuva with sampled households.

According to the 2017 population and housing census, the current population of Eswatini stands at 1,093,238 people and the rural region of Shiselweni contributes 204,111 people (18.7%) of which 14231 are from Zombodze Emuva [5]. The Shiselweni region registered negative population growth of -0.2% compared to the 2007 census, indicating outmigration of people from this region in search for job opportunities, among other factors. The proportion of the national population defined as poor fell from 69% in 2000/01 to 63% in 2009/10 (Ministry of Tourism and Environmental Affairs [51]. This national decline of 6% in the poverty rating was not evenly distributed across the four administrative regions, meaning that levels of poverty remain extremely high for many Swazis in more rural regions. As of the 2017 census, 41.1% of the national population is considered not poor, 39.7% is moderately poor, and 20.1% is extremely poor [5]. There is also a gender disparity to poverty, with 67% of female-headed households (predominately in rural areas) living in poverty compared to 59% of male-headed households [52, 51]. In this situation of extreme poverty and regional inequalities, Zombodze Emuva and Shiselweni region, in general, experiences the highest infant mortality rate (76 deaths per 1,000 live births) and rates of infant deaths continue to display an upward trend as it is currently estimated at 81 deaths per 1,000 live births [17].

It is generally believed that behind high infant mortality in countries of the global South and in Eswatini, in particular are various socio-economic factors, environmental and maternal factors. While the major direct causes of child deaths are well known, our knowledge of the relative important underlying socio-economic, environmental and maternal determinants of infant mortality in Zombodze Emuva is less clear. The study, therefore, aimed to contribute by filling up this knowledge gap, and probably help uncover some factors that may be more specific (rather than general) to the study area. The study, therefore, employed the mixed-method research design by combining quantitative and qualitative methods. Mixed methods research has been widely used within healthcare research for a variety of reasons [53]. In the context of this study, it was used to fully understand the diversity of the determinants of infant mortality from the perspectives of rural households and health education practitioners in the area.

This study was targeted to regular households in Zombodze Emuva. A regular household in 2017 census referred to a person or group of persons who may be related (family) or un-related or both who live together and share meals (eat from the same pot), sharing of meals was the paramount criteria in identifying a household in a homestead [5]. The current population of Zombodze Emuva stands at 14231 people translating into 2700 regular households [5]. A sample size determination calculator was used to select a representative sample and the confidence level was set at 95% and 76 households were selected through simple random sampling. To select the sample, every individual household was assigned a number. Microsoft Excel (RAND function) was then used to generate random numbers which correspond with individual households in the selected study area. The GPS coordinates for all the randomly selected households are then loaded in a GPS which was used to locate the sampled households during the data collection exercise.

To increase access of the population to the health interventions, the Government of Eswatini has introduced the rural health motivator services to strengthen the outreach services into the communities [10]. The study also targeted health motivators in the Shiselweni region, particularly those servicing Zombodze Emuva. One female health motivator was purposively selected as she is better placed to provide valuable insights pertaining to the subject under investigation since she usually interacts with the households in different health issues including the wellbeing of infants. Ethical guidelines of the University of Eswatini regarding research involving human subjects were followed in conducting this study. A written permission was granted by the GEP Department. The purpose of the study was fully explained to each and every participant before s/he could grant his/her oral informed consent. All the participants were informed that their confidentiality and anonymity will be protected and that they can withdraw from the study at any time they wish.

Data Collection and Analysis

Although it is important and more authentic to understand infant mortality from the perspectives of mothers who actually experienced it, such an approach was considered to be very sensitivity. To minimize the sensitivity of this study, quantitative data were collected from a randomly selected heads of households using a structured questionnaire with closed-ended questions. Where the head is a male, his wife was asked to be present in the interview since some of the questions were focused on women. The head of household as decision-maker play a key role in the status of a household. However, the questions in the questionnaire were focused on childbirth by all women in a household not necessarily by the head (if is female) or his wife (if is a male). Hence, in cases were the household head or his wife was not available during the time of the visit, the questionnaire was administered to any member of the household who is above the age of 18. Qualitative data were collected through an in-depth interview with health motivator using a semi-structured questionnaire with open-ended questions. The semi-structured interview is commonly used in research projects to corroborate data emerging from other data sources [54]. The interview was captured using a voice recorder, after a verbal consent was granted by each participant. All the participants in this study were interviewed by the first author at their places of residence. Unstructured observations were also used to note factors that are likely to influence infant mortality in households that were visited. The data were analysed using quantitative and qualitative methods. Statistical Package for Social Sciences (SPSS) was used to analyse quantitative data from households. Descriptive statistics, such as frequency tables, cross-tabulations and bar charts were used to show the determinants of infant mortality in Zombodze Emuva. The qualitative data obtained from semi-structured interviews were analysed through a process of content analysis. This analytical approach is chosen because “content analysis is a useful technique for allowing us to discover and describe actions, context, people, places, events, etc.” [54].

DemographicCharacteristics of the Heads of Households

The study set out to determine the factors that contribute to infant mortality in the rural settlement of Zombodze Emuva in the Shiselweni region. Quantitative data were collected from a randomly selected sample of 76 heads of households (or their adult representatives) using a standardized questionnaire. Demographic characteristics of the heads of households were regarded as important to the study because they would give an indication of the socio-economic status of the households as this is one of the most significant determinants of infant mortality.  The literature has shown that all other determinants of infant mortality are linked to the socio-economic status of the household. In terms of marital status, the findings in (Table 1).

Table 1: Marital status of heads of households.

Reveal that the majority (65.8%) of the sampled heads of households were married. This is an indication of the relative importance attached to the social institution of marriage by the Swazi society. However, a startlingly finding is the high number of the heads of households (23.7%) who are widowed. This figure appears to suggest that mortality rate in the area is likely to be high across ages. Interestingly, the numbers of the heads of households who are either divorced or separated are relatively low and this appears to suggest the integrity of marriage as a social institution.

Given the high number of widows, it was necessary to look at the gender differences by cross-tabulating marital status with gender of the heads of households as shown in below (Table 2).

Table 2: Distribution of marital status of heads of households by gender.

As shown in table 2, the majority (36.8%) of male heads of households were married compared to the female counterparts (28.9%). On the contrary, the majority (17.1%) of female heads of households were widowed. This appears to suggest that women tend to outlive their husbands due to a number of reasons which include among others, women tend to be married to older men and they also take their health seriously by seeking medical attention compared to their male counterparts. Alternatively, a high number of households in Zombodze Emuva are headed by single women who are vulnerable to social and material deprivations, given the patriarchal nature of the Swazi society. The challenges faced by (women-headed) rural households are implicitly reflected by the educational levels and employment status of the heads of households (Table 3).

Table 3: Distribution of educational levels of heads of households by gender.

Table 3 shows that the majority (63.2%) of the heads of households had secondary education with only 5 (6.6%) having not attended school at all, and as such, suggesting high levels of literacy in the area. However, a closer look at the statistics indicates that the majority of female heads of households have lower educational levels compared to the male counterparts.  A larger proportion (13.2%) of those with primary education level and all (6.6%) heads of households who have never attended school are women. Consequently, as shown in table 4 below, the level of education of the heads of households appears to be related to their employment status one way or the other (Table 4).

Table 4: Distribution of employment status of heads of households by gender

Table 4 shows that 61.8% of the heads of households were employed whereby the majority (34.2%) of them was formally employed and 28.9% were informally employed. Admittedly, the level of unemployment is also high in the area given that the total number of the heads of households who never worked and those who were unemployed (28%). A closer look at the statistics indicates that the level of unemployment and working precarious conditions are high among female heads of households compared to their male counterparts. Women constitute the majority of those who are informally employed (22.4%) and those who never worked in their lives (5.3%), and as such, female headed households have low levels of income. It is unquestionable therefore that the majority of (female-headed) households in Zombodze Emuva are generally poor and susceptible to a wide range of factors that determine infant mortality as shown below. 

It is essential to reiterate that the Kingdom of Eswatini has one of the highest infant mortalities in sub-Saharan Africa (WHO, 2012). The country has four administrative regions and infant mortality varies by region as shown in below (Figure 2).

Figure 2: Distribution of infant mortality by region in the Kingdom of Eswatini 2010-2015.

            Source: Central Statistics Office (2016).

The more rural region of Shiselweni where Zombodze Emuva is located is leading in terms of infant mortality (86%) followed by the more diverse but rapidly urbanizing region of Manzini (82%) due to increasing levels of poverty. The predominantly sugarcane producing rural region of Lubombo has slightly low infant mortality (73%) due to many factors but mainly because of improving access to health facilities provided by the private sugarcane industry [55]. Infant mortality is relatively low in the Hhohho region due to easy access (better transport and communication links) to better health facilities located in the country’s capital city of Mbabane.This variation in infant mortality in the Kingdom of Eswatini, with Shiselweni region being on the lead is a clear indication of inequitable spatial distribution of resources in the country. This situation necessitates an area specific study on the determinants of infant mortality rather than the generalized causes of infant deaths in the country. Zombodze Emuva has the highest rate in Shiselweni region, thus the study aimed to assess determinants of infant mortality in this area. The study set out to investigate the occurrence of these factors in the 76 sampled households.

As noted above, public facilities and health facilities in particular in the Kingdom of Eswatini are unevenly distributed and skewed towards urban areas at the expense of rural areas. Hence, among the main issues that the study investigated was the popular place where women in Zombodze Emuva give birth, since the place of birth is a contributing factor to infant mortality. Interestingly, table 5 below demonstrates that the majority (92.1%) of the heads of households reported that expecting women in their households normally give birth in health facilities and only a fewer (7.9%) reported that childbirth in their households normally happens at home (Table 5).

Table 5: Popular place of delivery in Zombodze Emuva.

The literature states that places of delivery are an important determinant of infant mortality, particularly neonatal mortality. This study found out that a majority of women in Zombodze Emuva deliver in maternal health facilities. A study conducted in Eswatini by Zwane reveals that children delivered in modern health facilities usually exhibit lower rates of mortality. Although majority of women in Zombodze Emuva are reported to deliver in health facilities, some women still give birth at home, a practice which increases the risk of neonatal mortality [28]. The practice of delivering at home may, therefore, be considered as one of the factors that are likely to be contributing to high infant mortality in the area. This is supported by Zwane who warned that home delivery exhibits higher rates of mortalities than health facility delivery [28]. To account for the fact that the literature considers maternal education to be influencing the decision on where to deliver [56, 18, 17], the place of delivery was cross-tabulated with the educational levels of households heads. This was to try and determine if the educational levels of households heads as the main decision-makers has an influence on the places of delivery by women in their own households (Table 6).

Table 6: Level of education and place of delivery as determinants of infant mortality.

Table 6 below illustrates that all the heads of households with secondary education and above appears to be influencing the decision for their women to deliver in modern health facilities. Whereas the few heads of households that reported that childbirth in their households normally happens at home are those who have never gone to school or have only gone up to primary level.Using the educational levels of heads of households as a proxy as shown in Table 6, it appears the level of education influences the choice of place of delivery by women in Zombodze Emuva. The majority of the heads of households who have attained, at least secondary education reported that childbirth in their households happens in health facilities. This confirms the fact that literate women (and literate people more generally) are knowledgeable in making use of government and private health care resources in reducing the rate of infant mortality [28]. As widely documented in the literature it is likely that the highest levels of infant mortality in Zombodze Emuva are recorded in households with educated heads or mothers compared to households with uneducated heads or mothers [56,18]. However, it was also essential to establish from the heads of households if the age of the expectant mothers in their households had an influence in their choice of the place of delivery (Table 7).

Table 7: Age of mothers and place of delivery as determinants of infant mortality.

Table 7 demonstrates that the majority (88%) of women that are reported by the heads of households to have delivered their children in health facilities are young mothers aged below 35. Eight percent of women in this age bracket are reported to have delivered their children at home. Surprisingly, none of the women aged above 35 years was reported to have delivered their children at home. Surprisingly, none of the women aged above 35 years was reported to have delivered at home. A closer look at this statistics reveals three possible reasons for these findings. First, the educational levels of young mothers and their knowledge about the use of healthcare facilities appear to be influencing the decisions taken at a household level on the place of their childbirth. Second, it appears the heads of households were reluctant to reveal the place of delivery of their own children, hence, no woman aged above 35 years was reported to have given birth at home. Third, the few young mothers reported to have given birth at home might have lacked the autonomy necessary to advocate for their children in the household context [28]. They are the most vulnerable to infant mortality and this is worsened by the practice of home delivery. Generally, young mothers have difficulties in providing enough food for their children since their influence on allocation of household resources is very limited [38]. And are likely to experience high infant mortality rates compared to women of high maternal age [28]. Although the high number of women giving birth at modern health facilities is impressive, the study wanted to understand the reasons why there are some women who still give birth at home. The few (7.9%) heads of households that revealed that childbirth normally happens at home cited two interconnected reasons for this, namely, households’ lack of financial resources (40%) and the long distance household members had to travel to access the nearest health facility (60.0%). This is in line with the literature which shows that lack of access to health facilities due to their non-availability and financial constraints contributes to infant mortality rates. For this reason, poverty coupled with inaccessibility of maternal health facilities might be considered as factors that are more likely to be contributing to the rate of infant deaths in Zombodze Emuva.  The literature shows that the time upon which women go to health facilities to give birth also contributes to infant mortality. This study inquired from the heads of households who reported that childbirth happens in health facilities the time their women normally visit these facilities (Figure 3).

Figure 3: The time women in in Zombodze Emuva go to health facility for delivering.

Figure 3 shows that the majority (90.8%) of the heads of households in Zombodze Emuva stated that the women in their households visit the health facility when they are already in labour. The number of the heads of households who indicated that the women visit the health facilities before labour or when they are experiencing complications of pregnancy is surprisingly low (9.2%).  It appears that a huge number of women in Zombodze Emuva go to health facilities when they are already in labour and this is likely to be a contributing factor to infant mortality in the area. As observed by Zwane, mortality among children delivered in modern facilities is higher because mothers use these facilities mostly when they have pregnancy complications [28]. As noted above, lack of financial resources was cited as one the reasons for childbirth happening at home.  Correspondingly, the study sought to identify if the level of household income influences the time of visit to health facilities for those women who are reported to give birth in hospitals. For this reason, the time of visit to health facility by expecting mothers was cross-tabulated with the employment status of the heads of household as an indicator of the level of household income. The assumption was that the income level will be high in households with the heads who are formally employed, moderate in households with heads who are self-employed (informal sector) and low in households with heads who are not employed (Figure 4).

Figure 4: Occupation of head of household and time of visiting health facility by women as determinants of infant mortality in Zombodze Emuva.

Figure 4 demonstrates that the majority (92%) of the heads of households reported that women in their households go to maternal health facilities when they are already in labour despite the income level of their households. An unsurprising finding in figure 6 is the slight difference in the use of maternal health facilities between the households of formally employed heads and households of unemployed heads. Whilst 4% formally employed heads of households reported that their women make use of health facilities before labour, the same number of unemployed heads of households reported that their women visit health facilities when they are experiencing some complications of pregnancy. Formally employed heads of households are more likely to afford transport costs associated with pre and postnatal care compared to unemployed heads. These findings appear to suggest that expecting women from low income households in Zombodze Emuva visit health facilities only when they are facing what Blum calls the life-threatening complications of pregnancy[34].  Consequently, complications of pregnancy and reactive and infrequent visits to health facilities are some of the factors that are likely to be contributing to infant mortality in Zombodze Emuva. Although women in Zombodze Emuva infrequently visit health facilities, the health motivator spoke positively about their knowledge about natal care and highlighted transport challenges:

Even though the women here are knowledgeable about the importance of visiting health facilities when they are pregnant for check-ups, this area lacks good roads and there is no ambulance to furry women already in labour to deliver in hospital on time.

This extract from the healthcare officer clearly illustrates that poor access to transport and communication networks is one of the likely determinants of high infant deaths in the area. However, the study sought to get an understanding from the household heads (and their wives) the rate at which women in Zombodze Emuva attend pre-natal care despite transport challenges. In contrast to the findings on the time women visit health presented above (Figure 6).

Figure 5: Pre-natal visits as determinants of infant mortality in Zombodze Emuva.

Figure 5 demonstrates that the majority (64.9%) of the participants indicated that expecting women in their households always visit health facilities for prenatal check-ups and only 3.5% indicated that they never visit. Even though the women are knowledgeable about the importance of visiting health facilities as claimed by the health motivator, the rate at which they do not attend pre-natal care is worrisome. In this regard, lack of adequate knowledge about the importance of prenatal (and postnatal) care is likely to be one of the main determinates of high infant mortality rate in Zombodze Emuva.The literature indicates that following the visits to health facilities during pregnancy is the importance of postnatal care of infants themselves [47, 17]. The literature identifies methods of baby feeding as contributing significantly to infant mortality. This study therefore identified the types of feeding that the mothers in Zombodze Emuva use. Table 8 illustrates that the majority of household heads (73.7%) reported that the infants in their households are exclusively breast fed and only 9.2% reported that the infants are only bottle fed. The number of households where infants are fed both breast and bottled milk is significantly

Table 8: Baby food in Zombodze Emuva.

These findings reveal that the majority of women in Zombodze Emuva use the form of baby feeding recommended by World Health Organization, which is exclusive breastfeeding. Exclusive breastfeeding is not only good nutrition but it also eradicates infant mortality. On the contrary, bottle and mixed feeding contribute to infant deaths since utensils used in preparing the milk are rarely sterilised and might cause illness and even lead to ultimate death of infants.  Even though most infants in Zombodze Emuva are exclusively breast fed, the few that are bottle fed (exclusively and mixed) are exposed to the risk of infections and are more vulnerable to death. For this reason, exclusive bottle feeding coupled with mixed feeding can be considered as some of the main factors that are likely to contribute to the high infant mortality in Zombodze Emuva. Conversely, it transpired from the interview with the health motivator that bottle feeding is actually saving the lives of many infants who are left under the care of their grandparents:

We do our level best to educate young mothers about the importance of breastfeeding their babies during the first few months of birth and they do understand us very well. But what I can say to you is that after giving birth most of these young mothers simply leave the babies with their parents. They now have mouths to feed, buy cloths and other essentials, so they go and look for jobs wherever they can get them. What do you expect in such a situation? Grandparents are left with no choice but to use bottles to save lives.

The important role played by grandparents in bringing up their grandchildren as explained by the health motivator in this extract was observed in some of the households that were visited. Some grandparents were found to be taking care of young children in the absence of their mothers. This clearly demonstrates that grandparents particularly grandmothers are an important social support system in childcare giving in rural areas of Eswatini and in other African countries.   Although bottle feeding is not a recommended method, it is clear from the above extract that poor households in Zombodze Emuva are forced to use it due to the situation they are in. A closer look at this extract indicates that the use of bottle feeding in the context of Zombodze Emuva is not necessarily a problem but it is a consequence of high unemployment and lack of access to basic necessities of life such as sufficient food and clothing by poor households. Hence, the outmigration of young mothers resulting to infants being left under the care of their grandparents might be considered as a factor behind high infant mortality rate in the area

The previous section shed light on some of the socio-economic and maternal factors that are more likely to be contributing to the high infant mortality rate in Zombodze Emuva. It is clear that such factors are intertwined and they cannot be understood separately from each other. Similarly, those factors cannot be detached from the environment within which they operate. For this reason and in line with the literature, this study sought to identify environmental factors that are likely to be contributing to infant mortality in the area. The World Bank reveals that environmental factors impacting on health and infant mortality in particular are too broad [36]. This study focused on environmental factors the World Bank classifies as traditional hazards such as access to water, sanitation and hygiene, methods of solid waste disposal, and sources of energy.  The study found out that the main source of water for all the 76 households (100%) that were interviewed is piped communal taps and none of the households had its own water connection. The participants indicated that getting water from the communal taps is a challenge in terms of the distance they had to travel on daily basis and constant breakdown of the taps themselves. Similarly, the main system of sanitation available to most of the households was found to be pit latrines and others use communal forests. In this situation of poor access to water and sanitation, it is unquestionable that the level of hygiene in most of the households is highly compromised. Likewise, the study found out that the majority of households (96%) burn their litter in pits dug within their homesteads while very few (4%) said that they dump it outside their homesteads. In responding to the question that sought to identify other aspects they cover when educating mothers on the health and welfare of their infants, the health motivator stated that:

There is a lot we are covering. For example, we teach them about the importance of cleanliness and hygiene, not only of themselves but also of their environment. Waste management is one of the aspects we emphasize on because an untidy environment is dangerous to the health of babies. Disease carrying rodents and mosquitoes are attracted by waste and this is bad, not only for the babies but also for all community members. We encourage them to safely dispose of their waste and most of them heed to our call...

Although the health motivator did not specify the safe methods of solid waste disposal they encourage households in Zombodze Emuva to use, the popularity of disposing waste through burning is a clear indication that it is likely to be one of the disposal methods they emphasize. This popularity of burning of litter is also an indication that the area might be lacking a site or landfill designated for solid waste disposal. However, burning of solid waste is likely to be contributing to high infant mortality in the area as documented in the literature. In a study carried out in Kenya, Apunda asserts that burning litter contributes to high infant mortality as it produces toxic materials that are harmful to infants and may cause them to suffocate to death. In addition to this, all the households in Zombodze Emuva were found to be using fuel wood as their main source of energy for cooking but some also supplement it with electricity [27]. Using fuel wood for cooking contributes to indoor air pollution that is harmful to the life of infants. A study carried out in Nigeria affirm that electrification reduces infant mortality compared to fuel wood as fuel wood produces harmful gases that are dangerous to the lives of infants [57]. Nonetheless, it was observed that almost all the households visited have a separate outdoor area for cooking and this is likely to minimize infant mortality due to inhalation of harmful gases.

This study investigated socio-economic status, maternal and environmental factors that appear to be detrimental to the health and wellbeing of infants in rural households in Zombodze Emuva. The main target population was household heads as key decision makers at the household level. The majority of the participants were married females and most of them were unemployed. The evidence from the study revealed a multiplicity of interconnected factors that are likely to be contributing to high infant mortality rate in Zombodze Emuva. The findings of the study revealed that the low level of both formal and maternal education is likely to be a determinate of high infant deaths experienced in the area. This in line with previous studies [56, 18, 28]. That highlights the highest levels of infant mortality to be recorded on mothers with no formal education than on educated mothers. Moreover, the study found out that maternal age and place of delivery are likely to be contributing to the causes of high infant deaths in Zombodze Emuva. Interestingly, very few women in the area are reported to have given birth at home and this is highly commendable. However, the majority of women who deliver at modern health facilities visit such facilities when they are already in labour or experiencing other complications of pregnancy. Several factors were identified to be leading to this situation among them being poor transport and communication networks. High levels of household poverty and inadequate knowledge about prenatal care are also contributing factors to the infrequent utilization of health facilities, hence high mortality rates. Nevertheless, the study found out that the majority of women in the area use exclusive breastfeeding, which is considered to be healthy for infants. However, bottle feeding is used by few households but mainly in the absence of mothers who are forced by the challenging rural situation to migrate in search of employment opportunities. This situation of impoverishment of households is pronounced by environmental factors such as sources of water, systems of sanitation, methods of solid waste disposal, and sources of energy. The low quality of these factors is likely to be contributing to infant mortality rate in the area. The households are trying to improve their livelihoods despite the challenges they are faced with. Infant mortality rate is a key indicator used to measure progress in the development of any country and it is currently addressed by the third Sustainable Development Goal (SDG3). Although Eswatini is investing substantially in health, the findings of this study demonstrate that a lot still needs to be done in particularly in impoverished rural areas such as Zombodze Emuva. There is an urgent need for the Government of Eswatini to improve infrastructure in marginalized areas such as the Shiselweni region and Zombodze Emuva by building health facilities that will be easily accessible so as to enhance the quality of life of poor households. The results also provide insights into the need to upscale public education on infant mortality and it is recommended that campaigns be channeled on empowering women with maternal education. Above all, the results of this study reflect lack of opportunities in Zombodze Emuva and it is recommended that the Government and its development partners should pursue a progressive rural development strategy that is geared towards creating livelihood opportunities in the area.

Acknowledgments

Authors wish to acknowledge the respondents in Zombodze Emuva for their valuable time and input to this study.

Conflict Of Interest
The author declares that there is no existing conflict of interest in this paper.

Ethical approval and consent to participate

The study was conducted in full compliance with the ethical guidelines of the University of Eswatini as approved by the institution.

1. Hug L, Sharrow D,You D. Levels & Trends in Child Mortality. United Nations Children’s Fund. 2017.
2. Alemu AM. To what extent does access to improved sanitation explain the observed differences in infant mortality in Africa? African journal of primary health care & family medicine. 2017; 9.
3. Federal Republic of Nigeria Implementation of the SDGs: A National Voluntary Review. Senior Special Assistant to the President on SDGs. Nigeria. 2017.
4. Anderson BD. Factors Driving Sugar Cane Production in the Kingdom of Eswatini. Theses and Dissertations Retrieved. 2018.
5. The Population and Housing Census: Preliminary Results. Mbabane: Central Statistics Office.2017.
6. Sachs J, Schmidt-Traub G, Kroll C, Lafortune G, Fuller G. Sustainable Development Report 2019. New York: Bertelsmann Stiftung and Sustainable Development Solutions Network (SDSN). 2019.
7. Ndongo-She N. Demystifying the Sextual and Domestic Violence Act (SODV) 2018. An event organized to educate, inform and motivate students across the University of Eswatini on the widely talked about Sexual and Domestic Violence Act 2018. United Nations – UN Eswatini.2020.
8. Keregero KJB. Food aid in Swaziland: Emerging lessons and alternative strategies. Swaziland Journal of Sustainable Development. 2012; 1: 91-114.
9. Ministry of Health Essential Health Care Package for Swaziland. Mbabane, Swaziland. 2010.
10. Mhlongo V, Mndzebele L, Ndlela P, Shabangu S, Nyoni D, Sardiwal D, et al. Swaziland Millennium Development Goals Report. Mbabane: Ministry of Economic Planning. 2010.
11. Mosley WH, Chen L. An analytical framework for the study of child survival in developing countries. Population and development review.1984; 25- 45.
12. Osawe OW. Determinant of Infant Mortality Rate: A Panel Data Analysis of African Countries. Developing Country Studies. 2014; 4.
13. Kurinczuk JJ, Hollowell J, Brocklehurst P, Gray R. Inequalities in Infant Mortality Project Briefing Paper 1. Oxford: National Perinatal Epidemiology Unit. 2009.
14. Sullivan, J.M. and N.K. Tureeva, Infant and Child Mortality, Ch. 8 in Ministry of Health of the Republic of Uzbekistan et al. Uzbekistan Health Examination Survey 2002. Tashkent: Ministry of Health. 2004; 77-85.
15. Nath S, Hardelid P, Zylbersztejn A. Are infant mortality rates increasing in England? The effect of extreme prematurity and early neonatal deaths. Journal of Public Health. 2020; 43: 241-550.
16. Galobardes B, Shaw M, Lawlor DA, Lynch JW, Davey Smith G. Indicators of socioeconomic position (part 1). Journal of epidemiology and community health.2006; 60: 7-12.
17. Susuman SA. Is Swaziland on Track with the 2015 Millennium Development Goals? Journal of Asian and African Studies.2017; 52: 1117 -1123.
18. Mustafa HE and Odimwengu C. Socioeconomic determinants of infant mortality in Kenya; Analysis of KDHS. Journal of Humanities and Social sciences. 2008; 2: 7-13.
19. Rutstein SO. Factors associated with trends in infant and child mortality in developing countries during the 1990s. Bulletin of the World Health Organization. 2000; 78: 1256-1270.
20. Pena R. Wall S, Pearson LA. The Effect of Poverty, Social Inequity, and Maternal Education on Infant Mortality in Nicaragua, 1988-1993. American Journal of Public Health. 2006; 90: 64-69.
21. Mutunga CJ. Environmental Determinants of infant mortality in Kenya. Nairobi: Institute for Public Policy Analysis. 2007
22. Hommann K, Lall SV. Which Way to Livable and Productive Cities? A Road Map for Sub-Saharan Africa. International Development in Focus. Washington, DC: World Bank. 2019.
23. International Labour Organization. World Employment Social Outlook: Trends for Women 2018 - Global Snapshot.
24. Ministry of Economic Planning and Development (MEPD, 2019) National Development Plan 2019/20 – 2021/22- Towards Economic Recovery. The Kingdom of Eswatini.
25. Yi B, Wu L, Liu H. Rural-urban differences of neonatal mortality in a poorly developed province of China. BMC Public Health.2011; 11: 477.
26. Madise N. Banda E, Benaya K. Infant Mortality in Zambia: Socioeconomic and Demographic Correlates. Social Biology. 2003.
27. World Health Organization. Trends in Maternal Mortality: 2010 to 2015. Geneva: World Health Organization. 2016.
28. Zwane E. Socioeconomic and Maternal Determinants of Infant Mortality: An Analysis Using the Swaziland Demographic Health Survey 2007. The Internet Journal of Epidemiology.2012; 10.
29. Banerjee A, Jadhav S, Khedkar D, Bhawalkar J, Rathod H. Access to health services among slum dwellers in an industrial township and surrounding rural areas: a rapid epidemiological assessment: J Fam Med Primary Care. 2012; 1: 20-26.
30. Safi N, Naeem A, Khalil M, Anwari P, Gedik G. Addressing health workforce shortages and maldistribution in Afghanistan. East Mediterr Health J. 2018; 24: 951-958.
31. Adatara P, Strumpher J, Ricks E, Mwini-Nyaledzigbor PP. Cultural beliefs and practices of women influencing home births in rural Northern Ghana: International Journal of Womens Health. 2019; 11: 353.
32. Uddin G, Kabir M. Factors associated with child health in urban areas of Bangladesh, working paper at the University of Jagannath, Department of Demography and statistics, Dhaka, Bangladesh.2006; 6-8.
33. Alvarez JL, Gil R, Hernandez V, Gil A. Factors associated with maternal mortality in Sub-Saharan Africa: an ecological study. 2009
34. Blum RW. Youth in Sub-Saharan Africa: Journal of Adolescent Health. 2007; 41: 230-238.
35. Singh S. Adolescent Childbearing in Developing Countries: A Global Review. Studies in Family Planning, 1998; 29: 117-136.
36. World Bank. Globalization, Growth and Poverty. Washington DC: World Bank.2000.
37. Galway K, Wolff B, Sturgis R. Child Survival: Risks and the Road to Health. Institute for Resource Development. Westinghouse.1987.
38. Ikamari LDE. Explaining Regional Variation on Infant and Child Mortality in Kenya. A working paper in Demography, Research School of Social Sciences. The Australian National University. 1995
39. Banerjee S. Major Determinants of Infant Mortality: District-level Evidences from Odisha. Journal of Health Management. 2018; 20:345-362.
40. Pison G. Why African families are larger than those of other continents.2017.
41. DaVanzo J, Burtz WP, Habitcht JP. How Biological and Behavioural influences on mortality in Malaysia vary during first year of life: Population Studies. 1983; 37: 381-402.
42. Palloni A. Milman S. Effects of inter-birth intervals and breast feeding on infant and early childhood mortality. Population Studies.1986; 40: 234-300.
43. Water, sanitation and hygiene (WASH): Improving water, sanitation and hygiene in Eswatini.2020.
44. Adekola PO. Allen AA, Akintunde EA. Environmental Factors Affecting Infant Mortality in Ibadan North Local: African Journal of Social Sciences. 2014; 4: 53-67.
45. Schell CO, Reilly M, Rosling H, Peterson S, Ekstrom AM. Socioeconomic determinants of infant mortality: a worldwide study of 152 low, middle, and high-income countries. Scand J Publ Health 2007; 35: 288-297
46. World Health Organization. Trends in Maternal Mortality: 1990 to 2010. Geneva: World Health Organization Press.2012.
47. Genowska A, Jamiolkowski J, Szafraniec K, Stepaniak U, Szpak A, Pajak A. Environmental and socio-economic determinants of infant mortality in Poland: an ecological study. Environmental Health. 2015; 14.
48. Godson MC. Nnamdi MJ. Environmental Determinants of Child Mortality in Nigeria. Journal of Sustainable Development. 2012; 5.
49. Zakir M, Wunnava PV. Factors Affecting Infant Mortality Rates: Evidence from Cross-Sectional Data. Applied Economics Letters. 1999; 6: 271-273.
50. Claeson M, Bos ER, Mawji T. Pathmanathan I. Reducing Child Mortality in India in the New Millennium:World Health Organization.2000;78: 1192-1199.
51. Ministry of Tourism and Environmental Affairs (MTEA2016). The Kingdom of Swaziland’s Third National Communication to the United Nations Framework Convention on Climate Change (UNFCCC). Mbabane, Swaziland.
52. Swaziland Millennium Development Goals Progress Report. Swaziland on the Road to Development. Ministry of Economic Planning and Development. 2015.
53. Ostlund U, Kidd L, Wengstrom Y, Rowa-Dewar N. Combining qualitative and quantitative research within mixed method research designs: A methodological review. International Journal of Nursing Studies. 2011; 48: 369-383
54. Maree K. First Steps in Research. Second Edition. Van Schaik Publishers. 2016.
55. Richardson-Ngwenya P, Richardson B. Aid for Trade and African Agriculture: the bittersweet case of Swazi sugar. Review of African Political Economy. 2014; 41: 201-215.
56. Amouzou A, Hill K. Child Mortality and Socioeconomic Status in Sub-Saharan Africa. African Population Studies. 2004; 19:1-12.
57. Ezeh O, Agho KE, Dibley MJ, Hall H, Nicholas A. Determinantsof Neonatal Mortality in Nigeria: Evidence from the 2008 Demographic Health Survey. Abuja: BioMed Central Ltd. 2014.
58. Madise NJ. Infant Mortality in Zambia: Socio-economic and demographic Correlates. Social biology.2003; 50: 148-166.
59. Parnell S, Pieterse E. Africas Urban Revolution. Cape Town: UCT Press.2014.
60. UNICEF and World Health Organization. Progress on Drinking Water and Sanitation: 2012 Update. New York: WHO Library Cataloguing-in-Publication Data.2012.
61. World Economic Forum. Poor sanitation is one of the biggest causes of child mortality. So what can we do about it? 2020.
62. World Health Organization. Complementary Feeding and Summary of Guiding Principles. Geneva: World Health Organization.2000.
63. World Health Organizaton. Children: Improving survival and well-being 2020.
64. World Health Organization : Sanitation. 2020.