Coverage of the WHO’s four essential elements of newborn care and their association with neonatal survival in southern Nepal

Between 1990 and 2017 the global under-five mortality rate declined by 58% and the neonatal mortality rate by 51% [1]. Despite this progress, neonatal mortality continues to decline at a slower rate than under-five mortality and both measures remain well above the Sustainable Development Goal (SDG) targets for 2030 [1, 2]. Nepal was one of the few countries that achieved the antecedent Millennium Development Goal’s (MDG) fourth target, with the under 5 mortality rate declining from 162 to 38 deaths per 1000 live births. During the same time period, neonatal mortality in Nepal declined from 39 to 23 deaths per 1000 live births [3, 4].

In an effort to continue the progress made in reducing maternal, child and neonatal mortality, the global community has recognized that in going forward with the SDGs there needs to be a focus on quality of care measurement [5]. For this purpose, the World Health Organization (WHO) developed and released a list of global indicators for maternal and newborn care in 2014, one of which is the “proportion of newborns who received all four elements of essential care”, which are as follows: immediate and thorough drying, immediate skin-to-skin contact, delayed cord clamping and initiation of breast feeding within the first hour [6].

Immediate and thorough drying (ITD) of the infant serves two purposes in reducing neonatal mortality: aiding in stimulation/resuscitation and thermal regulation of infants [7]. Roughly 10 million babies annually will require simple stimulation, which includes ITD, in order to begin breathing [8]. Neonatal hypothermia has been shown to be associated with higher risk of mortality and morbidity, particularly in preterm and low birthweight infants [9‐11]. The primary source of heat loss for a newborn is the evaporation of amniotic fluid from the baby’s skin. Often, a newborn’s temperature drops 3–4 °C within minutes after being born [12]. This almost immediate change in temperature highlights the importance of ITD to reduce the presence of amniotic fluid on the baby’s body.

Skin to skin contact (SSC) is important for establishing breastfeeding practices, thermal regulation of a newborn, and transfer of normal skin flora from mother to infant. Infants who experienced SSC were more likely to be breastfed one to four months postpartum, to be exclusively breastfed at six weeks to six months, and breastfeed more effectively [13, 14]. The second most common source of heat loss for a newborn is via conduction, or being placed on a cold surface [12]. Immediate SSC ensures that the baby is placed on a warm surface, their mother’s chest. SSC is recommended for hypothermia prevention and has been shown to be at least as effective in temperature maintenance in newborns as conventional incubators [12, 15].

The third element in the WHO’s package of essential elements for newborn care is delayed cord clamping (DCC), which is defined as the severing of the umbilical cord at least one minute after delivery or when the umbilical cord is no longer pulsating [16]. DCC allows the newborn to continue receiving fetal blood from the placenta after delivery, which can result in 60% more red blood cells and a 30% increase in blood volume [17]. Other suggested benefits of DCC include increased hemoglobin concentrations, reduced rates of anemia, improved cardiopulmonary adaptation, and better iron status in infants six months postpartum [17, 18].

Early initiation of breastfeeding (EIB) is defined as initiation of breastfeeding within the first hour after delivery. There is substantial evidence demonstrating that breastfeeding, specifically exclusive breastfeeding for six months, is a cheap and effective way to reduce all-cause and infection-related infant mortality [19]. Recently, there has been a shift in focus to the timing of breastfeeding initiation. Early initiation has been shown to result in an increased duration of and exclusivity of breastfeeding [20, 21]. Furthermore, beginning to breastfeed within the first hour also ensures that the baby is exposed to colostrum, which is associated with improved development of the intestinal mucosal and immunological protections [22].

The logic is that neonates who receive all four essential elements are more likely to survive than those who do not, given the pre-existing evidence base for the individual elements. As the indicator was recently developed, there has been limited research conducted looking at the four elements together. This paper will examine the current coverage of each individual element, distribution of number of elements received, associated characteristics with receipt of each element, and the association(s) between the element(s) and neonatal mortality in Sarlahi district of Nepal.

This analysis demonstrated that the greater the number of WHO’s elements of essential newborn care that an infant receives, the lower their associated risk of mortality. The analysis also showed that despite these elements being recommended by the WHO, their coverage is quite variable in this population in rural Southern Nepal.

In the adjusted models, skin to skin contact, delayed cord clamping, and early initiation of breastfeeding were all associated with lower risks of mortality, whereas immediate and thorough drying was associated with a slight increased risk of mortality. As ITD is often included in a package of practices for neonatal resuscitation, there are few papers examining this single element’s impact on neonatal mortality. One study that examined the 2006 WHO recommendations for newborn care using data from the 2011 Bangladesh DHS also found an increased, but statistically insignificant, risk of neonatal mortality associated with drying within five minutes of delivery (OR = 1.64, 0.75, 3.58) [25]. A possible explanation for the increased associated risk of death for this element is that infants who were having difficulty breathing were more likely to receive ITD as part of the neonatal resuscitation procedure. Infants with issues breathing initially have been shown to be at a higher risk for mortality, confounding the impact for this element [26]. In the stratified analysis, ITD was associated with an insignificant reduction in mortality for preterm infants. As demonstrated by Mullany et al., preterm infants are at the highest risk of hypothermia-related mortality which is the other pathway through which ITD is hypothesized to offer a protective effect [26].

Skin to skin contact and early initiation of breastfeeding were consistently associated with a reduction in neonatal mortality in this analysis. There is a substantial amount of evidence supporting EIB as a simple, effective tool to reduce neonatal mortality, consistent with the findings presented here [22]. Skin to skin contact as part of kangaroo mother care has been shown to reduce mortality in low birthweight infants as well as improve breastfeeding practices [13, 27, 28]. In the model that included all four elements, both SSC and EIB demonstrated a significant reduction in risk of neonatal mortality (defined as death between three hours and 28 days postpartum) after controlling for the other elements and covariates, indicating that despite the relationship between the two elements, they both offer protective effects independent of the other.

The association between risk of mortality and delayed cord clamping was inconsistent across the models. This is congruent with other findings in both term and preterm infants, none of which found significant impacts on mortality associated with delayed cord clamping. Those analyses reported that in preterm infants, delayed cord clamping was associated with a significant decrease in the need for transfusions, intraventricular hemorrhage and necrotizing enterocolitis [17, 29]. These outcomes were not assessed in this study, but may be considered in recommendations in the future.

The study’s strengths are the size, completeness, short recall period and regular follow-up schedule of the parent trial. Additionally, the amount of information collected allowed for adjusting for multiple potentially confounding variables. Furthermore, surveys were conducted as soon as appropriate after births and deaths of infants to reduce potential recall bias.

The data for this analysis was collected using primarily maternal recall (few instances of family member recall for sections of the newborn verbal autopsy questionnaire), which is a potential limitation because maternal recall of events around labor and delivery has been shown to have varied validity. Previous validation studies that assessed maternal recall of three elements (DCC not examined) did not have high accuracy (AUC > 0.70); the AUCs ranged from 0.50–0.63, indicating poor to moderate validity [30‐32]. These studies were conducted in facilities where the validity of these recall measures may have been worse than for home births where the mother may have had more ability to see and know what was happening to the infant right after birth. In this study, the elements DCC and ITD had the highest numbers of “don’t know” responses (Supplemental Table 8), both of which a woman is likely to know if it occurred if she is told by a health care provider or family member present at delivery. However, there was no assessment of the validity of the maternal recall for home versus facility births in the parent study.

According to two reports, information on these elements collected in household and facility surveys to measure coverage is sparse, if collected at all [6, 33]. The fact that the parent trial collected information on these elements allowed us to demonstrate that the four elements together have the potential to make a sizable impact on neonatal mortality. Even though the magnitude of the reduction was not necessarily significant for all elements individually, there was a demonstrated dose response relationship between the number of the four elements received and reduced risk of mortality. Therefore, routine data on the coverage of the four elements, individually and the number received per infant, should be collected to inform programmatic efforts and health systems. This routine reporting of the four elements can act as a status update for health system functioning and goals for coverage can be set in order to reach the SDG’s neonatal mortality target of 12 deaths per 1000 live births [2].

As the majority of the 75 countries did not reach the MDG goals for child survival, it is apparent that further action is needed to achieve the SDGs. Determining how interventions interact, rather than exclusively studying them in isolation, can inform design and implementation of health programs. The results of this study demonstrate that these four interventions are associated with improved outcomes for newborns in a dose response way. Furthermore, these findings support designing future programs for newborn care with a multidimensional approach in order to improve newborn and child survival.