3.3.1 Individual level factors: Maternal- and child-related risk factors.

In this review, thirty-four out of forty-eight articles, found an association between maternal characteristics and infant mortality, while twenty-two articles (approx. 45%) reported on factors related to the child. Maternal age at birth was the most reported risk factor among all maternal characteristics identified in the articles, such that 40% of the included articles reported an association between a mother’s age at birth and the risk of infant death. Other maternal factors reported in the reviewed articles include maternal BMI, maternal education, fertility rate, health literacy and self-efficacy etc. Among child-related factors, gender/sex of the child, child’s weight and immunisation were all factors considered to be strongly related to infant mortality.

3.3.2 Interpersonal factors: Socioeconomic status and socio-cultural factors.

At the interpersonal level, approximately 38% of included articles reported associations between socioeconomic status and infant mortality, while nine articles (18%) reported on socio-cultural factors. Majority of the reviewed articles that focused on socioeconomic factors at the interpersonal level associated household income/wealth index (n = 14) with infant mortality. Other factors identified include father’s occupation, paternal education, maternal occupation/employment, social support (from family) and household headship. Socio-cultural factors identified at the interpersonal level include marital status, religion and ethnicity.

3.3.3 Organizational level factors: Provision and utilization of health services, health workforce, hospital resources and access to healthcare services.

Twenty-four articles associated provision and utilisation of health services factors with infant mortality, while other articles focused on access to healthcare services and hospital resources. As seen in Table 3, key factors associated with the utilisation of health services include antenatal (ANC) attendance, place of delivery (including traditional birth centres (TBCs) and mission houses) and mode of delivery (c-section vs vaginal). With regards to health workforce factors, the skill of birth attendants was the most reported factor affecting infant mortality in Nigeria. The factors identified under the theme of hospital resources were the adequacy of neonatal transport vehicles and referral services and the availability of appropriate medical facilities and equipment. Proximity/distance of health facility and out-of-pocket spending/private health finance were the two factors identified under the theme of access to health service.

3.3.4 Community level: Housing and environmental factors.

Approximately 40% of the reviewed articles reported housing-related factors as a significant community-level factor compared to five articles that reported on environmental factors. Place of residence, which refers to either urban or rural areas of residence, was the most reported factor among the community-level factors examined in the included articles. The studies that reported on environmental factors at the community level linked crude oil pollution, Carbon IV oxide (CO₂) emission, drinking water source, type of cooking fuel and conflict within 25km of residence to infant mortality in Nigeria.

3.3.5 Public policy level: Governmental factors.

At the public policy level, the only theme that emerged was governmental factors. Six articles reported on factors related to the government, with public health expenditure being the most reported factor associated with infant mortality. The other factors identified are GDP per capita and external health resources/aid.

4.1.1 Maternal-related factors.

Maternal age and maternal BMI. Studies in this review reported that mother’s proximate factors such as age and BMI were significant risk factors for infant death. Maternal age was observed to have a U-shaped relationship with early neonatal and infant mortality such that infant deaths were more common among younger mothers (< 20 years) and older mothers (> 35 years) [28, 30, 52]. Similar results were reported in Korea [72] Brazil [73] and also in a cross-sectional study that evaluated the association between maternal age and infant mortality across 55 LMICs, indicating that an elevated risk of infant mortality was observed among women below age 18 [74]. These findings can be attributed to the fact that adolescent girls are physically and psychologically immature to handle childbearing and are less likely to make contact with a health clinic [75] while older mothers tend to be more predisposed to hypertension and preeclampsia leading to adverse neonatal outcomes [2, 76]. According to the 2018 DHS report, Nigeria’s rate of adolescent pregnancies among girls aged 15 to 19 was approximately 23% [77] which is significantly higher than that of Ethiopia reported to be 13% [78]. Early marriage in Nigeria is synonymous with early childbearing where 4 in 10 women get married before 18 years and 31% have already given birth by age 18 [77].

In the case of BMI, studies in this review reported that babies born to obese mothers (BMI ≥30 kg/m2) had a greater risk of dying than mothers with low BMI [28, 44, 65]. Comparable results were documented in studies conducted in California [2] and Sweden [79]. Maternal obesity has been suggested to lead to higher risks of macrosomia, stillbirth, and neonatal admission to intensive care unit [80].

Maternal education and health literacy. From the review, maternal educational status or level of schooling was recognised as a protective factor for infant mortality due to its impact on the mother’s ability to access knowledge that will help promote her baby’s health [26, 27, 35, 57]. Similar results were reported in Ethiopia [81], Indonesia [82], and Denmark [83] where better maternal education was associated with a reduction in Infant mortality. In Nigeria, 35% of women (age 15–49) had no education and only about 11% had more than secondary school education in 2018; 88% of the women with more than secondary school education were likely to deliver at a health facility than their counterparts [77]. This suggests that mothers with increased levels of education actively seek health-related knowledge and health system services compared to those with lower or no educational attainment [83]. Education and health literacy help to promote self-efficacy and confidence, allowing adherence to doctors’ advice, and increased chances of infant survival [61].

Parity, fertility rate and gestation age. Three studies in our review found high parity to be associated with IMR [9, 42, 59], such that babies born to mothers who had many children (grand multiparity) had higher chances of mortality. This is supported by the results of Kozuki et al. where increased mortality risk was observed among high parity infants, especially mothers having high fertility at the end of their reproductive years [76].

Although studies on fertility rate were limited in this study, one study emphasized that in the long run, total fertility rate significantly increased IMR in Nigeria [69]. As of 2018, women in Nigeria gave birth to approximately 5.3 children compared to 6.0 children recorded in 1990 [77]. This suggests that Nigeria’s fertility rate has barely changed in the past 33 years. One possible explanation for the high parity and slow decline in fertility rate among Nigerian women is that few women engage in family planning practices. According to the 2018 Nigerian DHS, only about 17% of married women used a family planning method [77], suggesting that the lack of proper family planning could be a potential cause for undesired births (increased fertility rate) and high parity. The likelihood of survival for infants born to high-parity women is severely impacted when there is poor access to adequate care, especially for those with low household wealth [9].

In addition to parity and fertility rate, gestation age which describes the duration of pregnancy before delivery, has been associated with infant mortality. Two studies in the review accentuated that preterm births (< 37 weeks) were a more significant contributor to IMR than term births [32, 46]. A Danish study demonstrated similar results wherein preterm babies were more likely to have an increased risk of death and was commonly observed among mothers with low education, adolescent mothers, and mothers with higher parity among other factors [83].

Preceding birth interval, multiple births, and birth order. Fourteen studies in this review demonstrated a significant association between infant mortality and preceding birth interval indicating that women with short birth intervals <18 months had a higher risk of infant deaths. Similar results were recorded in a study that compared birth histories from 66 different countries, revealing that neonatal, infant, and child mortality were strongly related to preceding birth spaces, where births within a short interval after previous birth increased mortality [84]. According to the 2018 NDHS report, 1 in 4 children are born less than two years following the birth of their siblings. Birth intervals of no less than 36 months should be maintained between successive births to lessen the likelihood of infant mortality [77].

Having more than one baby from a single pregnancy, known as multiple births, was reported to be a contributory factor to IMR. Infants from multiple births were considered twice as likely to die compared to singletons [30, 59]. This is comparable to a Korean study that found NMR and IMR to be significantly higher among twin and triplet births than in singleton births [85]. As suggested by Adeyinka et al. one possible explanation for this association is that multiple births tend to lead to preterm births and vulnerability to infection because of low birth weight.

Eight studies highlighted that having a fourth or higher birth order significantly increased the risk of postneonatal deaths [36, 44, 52]. Although similar results were found from research conducted in India, it was suggested that the relationship between birth order and NMR was not linear but J-shaped, signifying that the likelihood of mortality was greater among first and last-born infants compared to those born in the middle [86].

Breastfeeding duration and maternal outcome. Breastfeeding was considered by two studies in the review as a protective factor against IMR [25, 27]. Although these studies are old, there is reason to demonstrate the association between IMR and breastfeeding. Findings in the review suggest that a mother’s occupation affects IMR through breastfeeding wherein the intensity of breastfeeding matters more than its duration, indicating that the more an infant is breastfed the lower the susceptibility to the risk of death [25]. Another study conducted in Nigeria in 2002 supported these findings, indicating that the higher the level of education and employment of a mother, the higher her chances of breastfeeding her child [87]. Recent research in the U.S. found a significant reduction in postneonatal death across all seven regions in America as a result of breastfeeding [88]. The WHO advocates for exclusive breastfeeding for the initial six months. In Nigeria, 97% of children have been breastfed at some point, with a notable improvement in the practice of exclusive breastfeeding from 17% of infants exclusively breastfed in 2013 to 29% in 2018 [77].

Only one study in the review reported on the association between maternal outcome and infant mortality, emphasizing that mothers who died intrapartum (during childbirth) had an odds ratio of 15.28 when assessing their potential role in predicting infant death [46]. Although it is not clear why studies on this factor were limited, maternal mortality is an important predictor of infant survival and persists as a public health issue. In Nigeria, for every 100,000 live births that occurred in 2018, there were 512 maternal deaths [77], a slight reduction from 576 maternal deaths per 100,000 live births in 2013. A similar study conducted in the Western Pacific Region on maternal and newborn health suggested that providing high-quality care during the period of labour and immediately after childbirth is essential for ensuring the survival of both the mother and the newborn baby [89].

4.1.2 Child-related factors.

Child’s sex and child’s weight. Child sex emerged as one of the significant predictors of IMR among individual-level factors. Thirteen studies revealed that male infants were more likely to experience mortality compared to female infants. This is consistent with studies from Ethiopia [90] and Angola [12]. Also, recent results from a Brazilian study revealed that male neonates from twin births were at a significantly higher risk of death than females despite weighing heavier at birth [91]. This disparity in IMRs between male and female infants has been ascribed to differences in genetic and biological composition, predisposing boys to disease susceptibility and premature death [53, 90].

Birth weight as a predictor of infant death was highlighted by fifteen articles in the review. Babies with low birth weight (LBW) were highly associated with an increased risk of neonatal death especially for preterm babies. This result is supported by findings from other countries like the U.S. [2], Brazil [92] and Ethiopia [93]. In Nigeria, data on birthweight are not reliable considering that most births occur outside a health facility and data from surveys are based on the mother’s assessment of the child’s weight [77]. LBW has been attributed to factors like a child’s birth order, born to a younger mother, inadequate nutrient intake during pregnancy and poor quality of maternal care [41, 86].

Immunisation. Only two studies reported on the protective role of immunisation on infant mortality. These studies indicated that the number of infants immunised against infections and diseases significantly reduced the likelihood of infant mortality [51, 71]. Similarly, a scoping review conducted in Brazil reported the implementation of immunisation programs as a significant determinant of IMR [92]. Even though the rate of basic vaccination has improved in Nigeria, from 13% in 2003 to 31% in 2018, the number of children who have not received any vaccination has only reduced slightly to 19% in 2018 from 21% in 2013 [77]. Studies have shown that mother’s education, literacy level, place of residence, access, and proximity to health facilities were substantive factors associated with immunisation coverage [63, 77, 94].

4.2.1 Socioeconomic status.

Household wealth index. Results from fourteen studies demonstrated that households with lower wealth indexes have higher IMRs. Similar trends have been reported by other studies [11, 95, 96]. In Nigeria, household wealth has been shown to influence so many other factors such as education, immunisation, access to antenatal and postnatal care, residence, fertility etc. For instance, when compared to the fertility rate, women from the poorest households give birth to an average of 6.7 children, whereas women from the wealthiest households give birth to approximately 3.8 children [77]. People in poor households are more likely to put off the use of health services as they consider the cost of transportation, medicines, and tests, subsequently leading to a higher risk of mortality for their infants [70].

Father’s occupation and paternal education. Studies on the relationship between IMR and a father’s occupation and education were quite limited in this review. These factors were also reported in a study done in Pakistan to understand the paternal factors associated with neonatal deaths [97]. Akinyemi et al. stressed that the type of father’s occupation significantly influenced infant mortality such that infants born to fathers whose occupation was agriculture and labour had a higher likelihood of dying during infancy than fathers who were professionals [49]. One possible explanation for the association between father’s occupation and IMR is that fathers who are exposed to certain types of chemicals and pesticides are more likely to have babies with adverse birth outcomes [97].

Regarding paternal education, children whose fathers had no formal education were more likely to die compared to those born to fathers with formal education [53]. This finding is contrary to an earlier study that found no significance in the role of a mother’s partner’s education and infant mortality [96]. However, in terms of teenage pregnancy, one study found that adolescents whose fathers had no formal education had a higher chance of early childbearing compared to those whose fathers received an education [81]. These results underscore the significant role that a father’s level of education plays in mitigating early child marriages, adolescent pregnancies and maternal deaths, which, in turn, can reduce infant mortality rates (IMR).

Maternal occupation/employment. Only four studies examined the impact of mothers’ occupation on infant mortality. It was revealed that working mothers were more likely than non-working mothers to experience infant deaths [29, 49]. This result contrasts with a study from Japan that showed no link between the likelihood of infant death and the mother’s employment, rather, concluded that mothers’ occupation impacts infant health outcomes only when they are pregnant [98]. Alternatively, the findings from our review that suggest that maternal employment negatively impacts child survival could be interpreted with reference to the type of occupation. Mothers with low-status occupations like farming and petty trading may experience a high risk of infant death due to low income and unfavourable work conditions that promote poor health outcomes for mother and child compared to mothers in professional and administrative roles [99].

Social support and household headship. Support from family was shown by only one study in the review to be a crucial factor in the well-being and survival of infants. Social support from family members was defined as the frequency of receiving assistance and encouragement from family members to comply with health counsel for infant survival [61]. This is supported by studies that associated social support during pregnancy with birth size [100, 101]. It is suggested that psychosocial distress and depression during pregnancy could cause the production of higher levels of cortisol that could lead to low birth weight and preterm birth. Therefore, social support was found to be a protective factor that acts as a buffer from stress hormones thereby promoting healthy embryonic growth and infant outcomes [102].

Regarding household headship and structure, one study found female household headship to be associated with a reduction in the risk of infant mortality [56]. This is consistent with findings from a study conducted in India where children from female-headed households were found to have a greater chance of survival than children from male household headship regardless of place of residence and child’s sex [103]. In Nigeria, the proportion of female-headed households has remained practically the same since 2013 with less than 20% of households headed by women and only about 34% of married women being able to make decisions about their healthcare, paying a visit to relatives and making purchases [77]. It is therefore imperative to acknowledge the role women play in the family and encourage the enhancement of a woman’s autonomy and empowerment to enable her to make better decisions regarding utilising health services for both her and the baby, thus, contributing to the reduction of infant deaths in Nigeria [56].

4.2.2 Sociocultural factors.

Marital status/marital type. Results from this review showed that maternal marital status was strongly associated with infant mortality, which is consistent with other studies [104, 105]. Infants born to mothers who were married or in a cohabiting relationship displayed a decreased mortality risk in comparison to those whose mothers were previously married (divorced, widowed, or separated) [45]. Also, children born into polygynous marriages (married to more than one wife) were discovered to have a higher risk of infant mortality [56]. This finding is supported by Smith-Greenaway and Trinitapoli (2014), who confirmed the survival disadvantage of infants born to polygynous marriages irrespective of a higher or lower household wealth index [106]. In Nigeria, the effect of marital type on IMR has been observed to co-exist with the region of residence; in 2013, 36% of women in the northwest were in polygynous unions compared to only 9% of women in the southeast [107]. Polygyny is linked to high parity and high fertility and has been found to impact infant survival through resource limitations and poor living conditions [56].

Religion and ethnicity. Religion and ethnicity were identified as predictors of infant mortality. Similar findings were reported in a study in India where infant mortality was strongly associated with religion and caste [86]. Results from this review suggest that children from households that practice Islam had a higher rate of mortality [55, 65] compared to other religions, and children born to the Hausa/Fulani ethnic group were more exposed to death during infancy [9, 42]. This could be attributed to the existence of polygamous marriages, teenage marriages, early childbearing, and high-risk birth behaviours (such as short birth intervals and high parity) among mothers from the Hausa tribe [9].

4.3.1 Provision and utilisation of health services.

ANC attendance and prenatal care. Findings from this review demonstrated the association between increased infant mortality and lack of ANC attendance [29, 33, 46, 50], which is also similar to findings from a Kenyan study where a greater risk of death was observed among neonates delivered to mothers who did not utilise ANC services during pregnancy [108]. Health facilities provide ANC services as a way of monitoring and supervising pregnancies and utilising this service is seen as a protective factor that offers the opportunity to avert, detect, and possibly treat illnesses or problems that increase the likelihood of infant death [32]. In Nigeria, the percentage of women who attend four or more ANC visits has improved, from 44.1% in 2008 to 57% in 2018 [77], however, the majority of these women were from the urban areas while 47% of rural women were found not to have received any antenatal care. Female education and wealth index are precursory factors that influence ANC attendance [107], therefore, the quality of prenatal care and intrapartum supervision offered during ANC visits proves to be essential factors in mitigating the risks of stillbirths and other adverse birth outcomes [33].

Place of delivery, mode of delivery and time of birth. Correspondingly, delivery conditions such as birthplace, time of birth, and mode of delivery were found to influence infant mortality [31, 53]. Mothers who delivered their babies outside the hospital (at traditional birth centres and mission houses), were reported to experience more infant deaths than mothers whose delivery occurred at a health facility [26, 69]. This result is corroborated by a study done in Bangladesh [109]. In Nigeria, the percentage of health facility deliveries has barely improved, where 35% of births were recorded in 2008 compared to 39% in 2018 and over 59% of births are still delivered at home. Place of delivery is largely driven by the level of female education attained and wealth index [77]. One study in the review examined the connection between place of delivery and social class and revealed that home births and deliveries in unsupervised settings were more frequent among women from lower socioeconomic groups than others [68]. This finding demonstrates that mothers from poor backgrounds would seek cheap and substandard healthcare services rather than pay for quality services due to financial limitations [71].

Neonates delivered by caesarean section have a significantly higher risk of death than those delivered normally [38]. This finding is in line with a study done in California [2]. However, it has been suggested that having a skilled professional or birth attendant during delivery at a health facility, especially for neonates born through caesarean section, helps to avert or resolve birth complications [53, 64].

About the time of birth, studies in the review emphasised that women who delivered during the daytime at a health facility, as opposed to nighttime (between the hours of 8 p.m. and 8 a.m.), had a reduced risk of infant mortality [31, 48]. In contrast, a study conducted in Spain found no distinction in terms of survival between births that occurred during the day and those that took place at night [110]. The only plausible reason attributed to the negative relationship between nighttime delivery and IMR was that the lack of proper/facility care at the time of birth increases the risk of neonatal deaths [48].

Provision of 24-hour service coverage and delay in healthcare delivery. An increase in the rate of mortality was predominantly observed in areas that lacked access to 24-hour health facilities [48]. Studies in high-income countries have demonstrated that having facilities with medical personnel available 24/7 is a key factor in reducing infant mortality [110, 111]. The importance of providing round-the-clock services could be linked to birth timing, such that women who go into labour at night are given the same attention as those for daytime delivery.

Only one article reported on the delays in healthcare delivery and NMR, however, the findings revealed that delay in reaching a health facility (level-2 delay) was the most significant factor in NMR compared to the delay in receiving healthcare (level-3 delay) [39]. This is somewhat different from the study led by Upadhyay et al. where level-3 delay was shown to be the most important contributory factor to NMR [14]. In Nigeria, delay in healthcare services is mostly related to lack of transportation to and from the facility whereas delay in healthcare delivery was commonly observed among tertiary and secondary health institutions than primary and private hospitals [39].

4.3.2 Access to healthcare services.

Proximity to health facilities and out-of-pocket spending. The influence of household proximity to health facilities was investigated by two studies as a potential factor associated with NMR. Neonates born to mothers residing close to a hospital exhibited a reduction in the likelihood of neonatal mortality compared to neonates born to mothers residing far away [68]. A similar study has also reported accessibility to health facilities as a determinant of IMR [92]. In Nigeria, access to health facilities was reported by 52% of women as a major concern in 2018, particularly women from rural areas (60%), while 26% of women reported concerns about the distance of the health facility [77].

In the review, studies showed that out-of-pocket spending and dependence on private health finance are associated with IMR and are a major deterrent to seeking healthcare services [58, 71]. This contrasts with another study that found no significant association between private health expenditure and neonatal or infant mortality [112]. Over 97.4% of Nigerian women reported not having any health insurance, relative to 1.9% of women with employer-based health insurance [113]. This suggests that many Nigerians pay for health services out of their pocket and thus calls for the re-evaluation of the National Health Insurance Scheme (NHIS) in Nigeria.

4.3.3 Health workforce.

Availability of obstetricians and doctors/patient ratio. Fawole et al. highlighted the positive effect of the availability of obstetricians on the reduction of perinatal death [33]. The presence of an obstetrician is essential for providing quality obstetric care for both mother and child. In addition, findings from the review showed that an increase in the number of doctors per 1,000 patients caused a reduction in IMR [63]. These findings are corroborated by a Vietnamese study that confirmed the importance of the availability and density of doctors in improving health outcomes [114]. In 2019, Nigeria’s doctor per 10,000 population ratio was 3.67, compared to other African countries like Libya (21.57) and Algeria (17.32) [115]. This suggests that the number of doctors available to treat patients in Nigeria is significantly low and contributes to limited access to obstetric and postnatal care, consequently leading to increased mortality rates [63].

Skill of birth attendants and knowledge level of midwives. Studies revealed that NMR and IMR mortalities were highest among women whose prenatal care was provided by an unskilled attendant [38, 43]. A similar study demonstrated that having auxiliary nurses as birth attendants compared to doctors was a predictor of IMR [15]. The utilization of skilled birth assistance in Nigeria has slightly improved to 43% in 2018 from 32% in 1990 [77]. Furthermore, knowledge and understanding of midwives on evidence-based newborn care practices play a pivotal role in the reduction of neonatal mortality rates, particularly in Nigeria, where a significant proportion of childbirths (32%) are attended to by a nurse/midwife [77]. It is therefore imperative to empower and upskill birth attendants on best practices offered during childbirth and immediate newborn care, as they serve as essential support in preventing complications and ensuring survival during the birthing process [60].

4.3.4 Hospital resources.

Adequacy of neonatal transport vehicles and appropriate medical facilities and equipment. Results from this review suggest that inadequate neonatal transport services are barely considered to influence infant mortality in Nigeria as seen in the dearth of studies on the topic [40]. A study conducted in India supports this factor, where a substantial reduction was observed in neonatal and infant mortality due to the provision of emergency neonatal transport services [116]. NMR appears to be closely associated with the quality of care provided during pregnancy and delivery, thus, having adequate transport and referral services for neonates is critical, especially for transferring high-risk infants to higher-level facilities [39].

Likewise, the absence of essential medical equipment, such as incubators for the management of low-birth-weight infants, and inadequate power supply to sustain the limited incubators are all contributory factors to IMR [62]. This finding is also supported by a study in Bangladesh [109]. Since comprehensive data on neonatal transport services in Nigeria is limited, the provision of neonatal emergency transport services and medical equipment should be considered paramount for the reduction of IMR.

4.4.1 Housing/neighbourhood factors.

Place of residence and region of residence. At the community level, place of residence and geopolitical region were found to be associated with IMR, with most of the studies reporting on both factors concurrently. These results are in line with other studies that have reported an association between place and region of residence and infant mortality [12, 90, 117]. Studies that assessed the association between place of residence found inequalities in NMR and IMR across all the populations examined. Infants born to mothers residing in rural areas reportedly experienced higher IMRs than mothers who lived in urban areas [26, 35, 44, 57]. In Nigeria, urban areas consistently exhibit lower mortality rates in comparison to their rural counterparts as captured in the Nigeria DHS report. In 2013, IMR in urban areas was 60 deaths per 1,000 live births compared to 86 deaths per 1,000 live births in rural areas [107], while, in 2018, there was a slight reduction in IMR for urban areas (56 deaths per 1,000 live births) and rural areas (74 deaths per 1,000 live births) [77]. Evidence suggests that the reason for this urban-rural disparity in IMRs could be attributed to limited access to healthcare services and lower socioeconomic status among the inhabitants of rural areas [90, 92].

IMRs were observed to be different across the six geopolitical regions. For instance, some studies reported increased IMR in the North-West region compared to other regions [42, 57], while others reported a combination of two or more regions associated with increased risk, such as North-West and North-East regions [45, 64]. A plausible explanation for the variations in IMRs across different regions as reported by the included articles could be accredited to sampling sizes, the region assessed, and the possibility of under-reporting from data sources used. However, it was noted that mothers who reside in northern Nigeria face a higher risk of infant mortality than those who live in the south. According to the 2018 NDHS report, the northwest had an IMR of 80 deaths per 1,000 live births compared to 43 deaths per 1,000 live births in the southwest [77]. This has been attributed to differences in Nigeria’s regional environment regarding socioeconomic levels, health-seeking behaviours, and cultural customs [36]. For instance, the northern region of Nigeria is predominantly faced with child marriages, teenage mothers, insurgency, low levels of education and Islamic populations [118], while the southern regions are well known for having higher levels of education, utilising health services, having better socioeconomic development and being Christians.

Toilet facility and access to electricity. Access to proper sanitation facilities has been shown to influence infant mortality [43], which is comparable to the results from a multi-country level study [119]. In Nigeria, 56% of households had improved sanitation facilities in 2018 compared to 25% of households with no sanitation facilities [77]. Inadequate sanitation can contribute to the spread of infection which can lead to an increase in the risk of infant deaths [43].

Access to electricity has been shown in this review to play a crucial role in reducing IMR [41]. Similarly, in Ghana, it was observed that a 10% improvement in electricity accessibility reduced IMR, particularly within regions characterized by high mortality rates [120]. In 2018, 59% of households in Nigeria had access to electricity, urban households had more access (83%) than rural households (39%) [77]. Electricity is very important in healthcare as it contributes to the provision of quality care in health facilities through the optimization of medical facilities and equipment [41].

4.4.2 Environmental factors.

Crude oil pollution and CO₂ emissions. In the review, three studies demonstrated the association between environmental pollution and IMR such that mothers who resided in areas exposed to crude oil spills and CO₂ emissions were found to be at a higher risk of experiencing preterm birth and NMR [54, 67, 69]. A study conducted in Ecuador reported that women who lived in communities exposed to oil pollutants faced a heightened risk of spontaneous abortion [121]. Even though there is insufficient data on crude oil pollution and mortality rates in Nigeria, most oil-producing regions, specifically the south-south region, often suffer from environmental pollution due to oil exploration and oil spills and evidence suggests that exposure to hydrocarbons poses a significant risk to neonatal development and other adverse health outcomes [54].

Similarly, CO₂ emissions from cement production and gas flaring activities were revealed to significantly increase neonatal and infant deaths in Nigeria [69]. Similar studies have also identified a positive correlation between exposure to carbonaceous pollutants and increased infant mortality [122, 123] this is attributed to the fact that carbonaceous emissions have the potential to induce inflammation and oxidative stress that can impact the gestation period and foetal growth during pregnancy [123].

Drinking water source and type of cooking fuel. The importance of drinking water sources was emphasized in two studies [38, 43] such that access to an improved source of drinking water in the household was associated with survival advantage among infants. Other studies have noted similar findings on the effect of improved water sources on mothers and infants [93, 124]. As reported in the Multiple Indicator Cluster Survey (MICS) in 2021, approximately 79% of Nigerian households enjoyed access to an enhanced drinking water source, while 15% of households relied on unimproved water sources, reflecting a noteworthy decline from the previous rate of 25% recorded in 2018 [113].

Additionally, only one study demonstrated that women who used biomass/unclean fuel experienced higher neonatal and infant deaths than those using clean fuel [43]. In Nigeria, the percentage of households using solid fuel for cooking (including charcoal, wood, coal/lignite, straw/shrubs/grass, and animal dung) was approximately 65% compared to 27% of households using clean fuel for cooking (including electricity, natural gas, and biogas) [113]. This suggests that more families in Nigeria are exposed to high levels of indoor fumes and air pollution which lead to respiratory problems and other health implications for infants and mothers. Poverty and low wealth index appear to be the main factors influencing the use of solid/unclean fuels [43].

4.5.1 Governmental factors.

Public health expenditure and GDP per capita. The main governmental factor reported in the review was public health expenditure. Similar to studies done in India [125] and sub-Saharan Africa [113], this review found a negative relationship between public health expenditure and infant mortality, such that a percentage increase in government health spending led to a decrease in Nigeria’s infant mortality rate [51, 69]. Although the recommended healthcare budget by the WHO in 2009 was pegged at 15% of a country’s annual budget, Nigeria’s allocation to the health sector dropped from 8.2% in 2014 to 3.9% in 2018 [63]. The effect of this inadequate budgetary allocation to the health sector explains why the Nigerian healthcare system has been in a dysfunctional state, particularly the inadequate provision of health services, the shortage of medical equipment and facilities, wide gaps in supply and demand of skilled health workers, and the over-dependence of the masses on out-of-pocket payments for health services [126].

Likewise, studies on GDP per capita [63, 69, 71] suggest that lower GDP per capita negatively affects access to quality health services and leads to people depending on inadequate, cheap and traditional health services [63]. In 2020, Nigeria recorded a GDP per capita of $2,074.6, a significant decline from $3,201 in 2014. Countries with similar and even lower GDP per capita have recorded lower IMRs compared to Nigeria. For instance, in 2020, Bangladesh with a GDP per capita of $2,233.3 had an IMR of 24.06 deaths per 1,000 live births while India recorded a GDP per capita of $1,913.2 and an IMR of 26.82 deaths per 1,000 live births [3, 127]. This disparity in IMRs demonstrates the complexity between economic development and basic infrastructure and therefore suggests that Nigeria’s GDP has not translated into an improvement in healthcare provision and infant survival.

External health resources/aid. This review found the effect of external health resources to influence infant mortality directly and indirectly [51]. Comparably, results from a recent study revealed that health aid showed a favourable and significant impact on the health outcomes of low-income countries [128]. In Nigeria, external health aid from donor agencies and non-governmental organisations was reported to significantly affect infant mortality when directly invested in the health care system. Likewise, external aid in the form of aid projects such as agricultural, energy generation, water, sanitation, and health projects was shown to have an indirect effect on IMR [51]. The likelihood of death is lower for children born in areas close to an ongoing aid project than those without an aid project. Therefore, the allocation and effectiveness of external health resources in Nigeria tend to vary across different regions, and the impact of such aid on IMR is mostly influenced by governance and implementation capacity [47].