Impact of timing of breastfeeding initiation on neonatal mortality in India

Globally, around 5.6 million children died before reaching their fifth birthday, of those, 2.6 million (or 46%) died in the first 30 days of life [1]. Approximately 7000 newborns died every day, most of which occurred within first 7 days after birth, with about 1 million dying on the first day and close to 1 million dying within the next 6 days in 2016 [2]. Most of the neonates died in Southern Asia (39%), followed by sub-Saharan Africa (38%). Half of all newborn deaths occurred in the following five countries: India, Pakistan, Nigeria, the Democratic Republic of the Congo and Ethiopia [1]. Over the past 25 years, the age under-five mortality rate dropped from 93 deaths per 1000 live births in 1990 to 41 in 2016. In India, in the year 2015, infant mortality accounts for 37 infant deaths per 1000 live births, of which 67.8% infants (25 per 1000 live births) died in the first month of births [3]. In 2013, India recorded the highest absolute number of neonatal deaths of any country, nearly 0.75 million [4]. Despite a significant change in neonatal mortality over the years, progress has been inadequate towards achieving Millennium Development Goal 4 (MDG-4) [5]. In 2015, the Sustainable Development Goals (SDGs) have been introduced, seeking to achieve all the goals by 2030. Goal 3 of the Sustainable Development (SDG 3) is focused on promoting MDG-4 to reduce the under-five mortality by two thirds, between 1990 and 2015 and will continue beyond 2015, until neonatal mortality is at least as low as 12 per 1000 live births and under-5 mortality to at least as low as 25 per 1000 live births [6].

Various factors can effectively reduce neonatal mortality to greater levels, early initiation of breastfeeding is one of them and it has benefits for survival and beyond. The World Health Organization (WHO) has recommended that all neonates to be breastfed within 1 h of birth. The deleterious effects of infections related infant deaths can be prevented by early initiation of breastfeeding (or human milk feeding) and exclusive breastfeeding which is the easiest, cost-effective and life-saving intervention for the health of a newborn [7].

Early initiation of breastfeeding and exclusive breastfeeding for the first 6 months of life prevents around 20% newborn deaths and 13% under-five deaths [8]. It can also reduce mortality due to neonatal infections (sepsis, pneumonia, tetanus, and diarrhea) [9] which contribute 36% in neonatal deaths from all causes, and preterm birth an additional 27% [10]. In spite of the strong evidence supporting immediate and long term health benefits, early initiation of breastfeeding in South Asia remains low with varying rate with 36.4% in India, 24% in Bangladesh and 8.5% in Pakistan [11‐13]. In India, only 65% of children are exclusively breastfed for the first 6 months and 45% of children receive breastfeeding within 1 h of birth, though breastfeeding is one of the most important interventions of child survival [14]. In the year of 2016, the Government of India launched the National Breastfeeding Promotion Programme MAA (mothers’ absolute affection) to ensure adequate awareness is generated among the masses, especially mothers, on the benefits of breastfeeding. The Programme will be implemented at three levels: Macro level through mass media; meso level in health facilities and micro level at communities [15].

Prior studies have shown that early initiation of breastfeeding is associated with a lower risk of neonatal mortality [16‐19]. It is found that globally, only half of newborn babies are breastfed during their first hour of birth, despite strong evidence of nutritional and immunological benefits of early initiation in reducing neonatal mortality and morbidity [18, 20, 21]. The first milk (colostrum) contains bioactive immune factors which protect a neonate against a variety of infections and allergic diseases [22]. A recent systematic review of the literature based on 25 studies from seven countries in South Asia, revealed that early initiation of breastfeeding is predominately associated with socioeconomic, health related and individual factors, [23] and that insufficient attention is afforded to intrapartum and neonatal characteristics.

A few studies in India have examined associations between timing initiation of breastfeeding and neonatal mortality in different communities [24, 25]. There is a need for a national level study as it is important to examine the association between timing of initiation of breastfeeding and neonatal mortality nationally. The present study investigates the role of timing of initiation of breastfeeding in improving neonatal survival using a nationally representative large scale population-based dataset in India. In order to achieve the Sustainable Developmental Goals within stipulated time, we need to emphasize on the evidence-based conclusion regarding the linkages between timing of breastfeeding and neonatal mortality reduction as there is an urgent need to improve breastfeeding practices in India.

The India Human Development Survey-II (IHDS-II), 2011–12, has been used for the present analysis, which is a nationally representative and multi-topic survey of 42,152 households in 1503 villages and 971 urban neighborhoods across India. Most of these households had been interviewed for IHDS-I (2005). Both surveys cover all states and union territories of India with the exception of Andaman/Nicobar and Lakshadweep. Sample Registration System (SRS) dataset is used to see the rate of change in NMR from the year 2011 to 2015. District Level Household & Facility Survey (DLHS-4), 2012–2013 and AHS (Annual Health Survey), 2012–13 state data have been used to show the district wise distribution of women who have breastfed their child within 1 h of birth.

Figures 1 and 2 show the State Wise Neonatal Mortality rates (NMR) in the year of 2011 and 2015. In 2011, six of the 22 states had an NMR of more than 30 per 1000 live births, which is the threshold level of Millennium Development Goal 4 for Neonatal Mortality Rate. By 2015, only three states had an NMR more than 30.

From the Fig. 3, we have seen the district wise  prevalence of breastfeeding within 1 h of birth. The spatial variation showed less than 40% of women were practicing breastfeeding their babies within 1 h of birth, in some districts of Rajasthan, Punjab, Haryana, Bihar, Uttar Pradesh and North-East. One of the notable feature that emerges from the map is that many districts where the prevalence of practicing breastfeeding is low, lies mostly in Bihar and distributed uniformly across the state.

Table 1 gives the total number and percentage distribution of live births in accordance with the background characteristics. There were only 21% live births that have breastfed within the 1 h of birth. A majority of babies lived in rural areas, whereas more than three-fourths were Hindu dominant households and belong to Other Backward classes and Schedule tribes/caste. Maternal characteristics showed most of the babies were born to mothers in the age groups between 25 and 34 years. Only 11% of mothers had completed higher education. Less than half of the mothers were illiterate. Of the total number of mothers, only 3 % were obese and more than half belonged to normal BMI (56.83%). By examining the birth size of the babies, Table 1 depicts that almost 76% of babies were of average size at birth. The data shows an equivalent distribution of boys and girl across the nation.

In Table 2, odds ratios were given with 95% of confidence interval after adjusting the covariates. If a woman did not breastfeed their newborn within 1 h after his/her birth then the odds of neonatal mortality is increased four (OR 3.54; 95% CI 2.34, 5.38) times compared to those neonates who have breastfed within 1 h of birth. Once controlling for all the community and household level variables in Model I, the risk was still significant. There was a significant difference in neonatal mortality between interstate regions. The odds of neonatal mortality are decreased by 40% (OR 0.60; 95% CI 0.42, 0.86) for the West region as compared to the Northern region. Variation can also be seen in the Northern and Southern region; neonates were 52% (OR 0.48; 95% CI 0.30, 0.75) less likely to die in the South than the North region. In the case of wealth quintile, neonates who belonged to the richest family has 48% (OR 0.52; 95% CI 0.35, 0.77) lower risk than those from the poorest family. When we add maternal and child level factors to Model II, Model III result showed that as the age of mother increases, neonatal deaths decreased. If a woman belongs to age 25–34, the odds of neonatal mortality is decreased by 46% (OR 0.54; 95% CI 0.39, 0.76) than those women who belong to age 15–24, further odds of neonatal mortality decreased to 49% (OR 0.51; 95% CI 0.33, 0.97) for the women belonging 35 years and above. Education levels of the mother also have a significant effect on neonatal mortality. Higher educated mothers have experienced less neonatal deaths than illiterate mothers. The expected odds of neonatal deaths is three times (OR 3.27; 95% CI 1.69, 6.34) higher for very small size babies than large size babies at birth. Neonatal death is decreased by 56% (OR 0.46; 95% CI 0.32, 0.67) for higher birth order children, and by 26% (OR 0.74; 95% CI 0.59, 0.94) for a girl child compared to a boy child.

Table 3 provide the post estimated value of logit model for Population Attributable Fraction (PAF) for different situations of breastfeeding. In the Table 3, the Population Attributable Risk was estimated for the situation when all babies are exposed to breastfeeding. Scenario 0 represents the baseline scenario which is an overall predicted probability of neonatal deaths at adjusted level while, Scenario 1 (fantasy scenario) represents the predicted probability of neonatal deaths when all babies are exposed to breastfeeding and other factors in the model are adjusted. In this study, we found PAF to be − 0.15 which indicates that when all babies are exposed to early breastfeeding, the risk of the neonatal deaths could be reduced to a maximum of 15% from the present level. Similarly, in Table 4 we have chosen a scenario (Scenario 1) when all the babies were not breastfed within the first hour of birth We found that in this extreme worst situation, the risk of neonatal deaths will increase to the level of 60%.

The present study supports the protective effect of early breastfeeding initiation on death within the first 28 days, including all-cause mortality. Wide interstate and intrastate variations exist in neonatal mortality across the country [26]. The present study also depicts that there is a north-south variation seen in neonatal deaths. It is clear from the maps that northern part of India is more vulnerable to neonatal mortality than southern part of India. Neonatal deaths can be prevented or reduced by early initiation of breastfeeding and exclusive breastfeeding [7]. There is a stark variation seen in children who have breastfed within 1 h of birth across the districts. The districts of Kerala, Tamil Nadu, Maharashtra, Odisha, Chhattisgarh, Madhya Pradesh, Himachal Pradesh, Uttarakhand and some districts of the North-East babies are more exposed to early initiation of breastfeeding, while babies from the districts of Bihar, Uttar Pradesh, Haryana, and Punjab are more exposed to delayed initiation of breastfeeding, where the rate of neonatal deaths are also high.

Newborn care immediately after the birth is vital since 40% of all neonatal deaths occur on the first day of life and 56% during the first 3 days [27]. The timing of breastfeeding initiation is found to be the most significant risk factor which affects neonatal deaths [7]. Early initiation of exclusive breastfeeding serves as the starting point for a continuum of care for mother and newborn that can have long-lasting effects on health and development [28]. The findings of the present study resemble its predecessor. There was an odd of 3-fold increase in the risk of neonatal deaths if there was a delay in initiation of breastfeeding.

In India, there are wide disparities in mortality by caste, region, place of residence, and economic status, among other characteristics [29]. These variations are related to differences in wealth, nutrition, education, availability of health services, culture and gender equality status [30]. The literature suggests that infant mortality rates have been inversely related to socioeconomic status, and child mortality is higher among the low-income families than non-poor families [31], which is consistent with our study.

Maternal and child level determinants as the weight of neonate, gestational age, and age of mother play a major role in the neonate’s survival [32]. Also, consistent with the earlier findings, maternal factors like mother’s age and education were prominent factors in neonatal mortality. Along with the maternal characteristics, child level factors have a notable effect on neonatal mortality. Very small babies were more likely to die than larger babies, babies and males were more vulnerable to neonatal death than females.

It is also found that two-thirds of the incidence of neonatal mortality can be attributed to delayed breastfeeding i.e., out of the total population, 60% of the babies died within 1 month of their birth because of they were not breastfed within 1 h. It has strong policy implications as we could reduce the risk of neonatal deaths to a great extent by giving them timely breastfeeding.

The limitations of the present study are that IHDS-II data were self-reported, as information on each outcome and determinants were collected retrospectively, there could be recall bias, that may have impact on results.

Though India has witnessed momentous changes in the infant health scenario over the years, the changes have not been uniform. Neonatal deaths are influenced by host of variables like community, household, maternal and infant level factors. The early initiation of breastfeeding can reduce the risk of neonatal deaths by odds of three times. So, these findings support the recommendations of early initiation of breastfeeding as an intervention to reduce neonatal mortality. The implementation of policies and pro-breastfeeding routines are the major recommended interventions to achieve SDGs to reduce neonatal mortality.