Background

A staggering 2.5 million newborns died, globally, of preventable or treatable causes within the first 28 days of life in 2018 [1]. This represented approximately 63% of all infant deaths and 47% of deaths in children five years and under [1]. Sub-Saharan Africa (41%) and Southern Asia (37%) accounted for close to 80% of the 2018 global estimates for neonatal mortality [1], where interventions to prevent these deaths are minimal [2–4]. Expansion of effective interventions to all who need it can improve survival rates of neonates, however, launching strategies necessary to meet this aim remain a challenge in these regions [5]

Prior research showed socioeconomic inequity in neonatal deaths tended to occur in disadvantaged subpopulations [5]. The majority of deaths took place in Low and Middle Income Countries (LMIC) [6], disfavouring the poor, illiterate and rural communities [5,7]. Interestingly, over the last 20 years, the inequality in NMR has been decreasing in most of the LMICs [5]. As the global community continues to work toward an equity-oriented international agenda, to meet the United Nations’ Sustainable Development Goals (SDG), it is imperative that evidence-based interventions reach the disadvantaged subpopulations to alleviate NMR disparities.

In 2018, nearly 100,000 babies died within the first 28 days following birth in Ethiopia [1], contributing to over 50% of the under-five mortality rate in the country. McKinnon et al. (2014) flagged Ethiopia as one of two countries where “wealth related inequality increased by more than 1.5 neonatal deaths per 1000 live births per year between 2000 and 2011” (5, p. 165). This underscores the importance of further research to investigate NMR relative to various dimensions of inequality (wealth, education, residence and sex) and how the disparity has changed over the past nearly two decades in Ethiopia.

Few research endeavors have attempted to assess NMR inequality in Ethiopia [5,7,8]. Our study enriched available evidence in a number of areas. First, it investigated NMR inequality through the rigorous method recommended by the World Health Organization (WHO) Handbook on health inequality monitoring. This method considers different dimensions of inequality measures in the analysis to gain a better understanding of the NMR disparity [9]. Commonly used absolute (difference) and relative (ratio) simple measures of inequality were calculated to reflect the magnitude of the difference in NMR between studied subgroups and how inequality changed over time [9]. In addition to the simple measures, which allow comparison between two subgroups only, the study also estimated the inequality using complex measures that account for all subgroups of an equity stratifier [9]. Second, it used the common equity stratifiers recommended by the WHO [9] which is in contrast to most prior studies that limited NMR inequality analysis to socioeconomic dimensions (i.e. wealth and education) [5,7]. Third, previous studies used traditional regression methods [8,10,11] that do not follow the WHO recommendation for inequality studies [9] or were limited to certain specific areas in the country without including the national context [11,12]. Finally, it assessed NMR inequality for the four Ethiopia Demographic and Health Surveys (EDHS 2000, 2005, 2011, 2016) and was not limited to the first and the last EDHS rounds, unlike prior studies [5,13]. This contributed to understanding the dynamics of NMR differentials in Ethiopia over sixteen years.

The study was designed to answer two questions: 1) what is the status of NMR inequality in Ethiopia across different equity stratifiers?; and 2) how have levels of NMR inequality in different equity stratifiers changed over time between 2000 and 2016?

Methods

Results

Table 1 presents NMR disaggregated by the four dimensions of inequality for each of the four EDHS. In 2000, the highest NMR was recorded among the poor, middle and rich subgroups, whereas the lowest was observed among the poorest and richest. This uneven distribution of NMR resulted in an inverted U shape of wealth-related inequality. In 2016, NMR was disproportionately high amongst the rich, with no statistically significant differences between the other four sub-categories. Individuals who completed secondary education experienced the lowest NMR in 2000 and 2005. However, in 2016, NMR was more evenly distributed across all the three subgroups of education. Across all surveyed years, a significant gender related disparity persisted to the disadvantage of male neonates.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Neonatal mortality rate (deaths per 1000 live births) in Ethiopia disaggregated by economic status, education, residence and sex across four time periods (2000, 2005, 2011, 2016).

https://doi.org/10.1371/journal.pone.0234483.t001

Inequality of NMR was analyzed with relative and absolute measures, using simple and complex techniques, as presented in Table 2. Wealth (i.e. economic status) related inequality differed widely by the type of inequality measures. The simple measure of Difference indicated no wealth-driven NMR disparity in all the surveys except in 2011, where, on average, an excess of 13 neonatal deaths per 1000 live births were recorded for the poorest subgroups. The SII showed that wealth-based inequality remained in 2011 and 2016. The inequality shifted from the richer subgroups in 2011 (SII = -18; -28, -9) to the poorer ones in 2016 (SII = 11; 2.3, 20). The PAR showed a wealth-related disparity in favour of the wealthier households during the first three surveys as the poorest households experienced higher NMR. In 2000, on average, the national NMR would have been reduced by 14 neonatal deaths per 1000 live births (PAR = -14; -15, -12) had all the wealth quintiles been on the par with the richest fifth quintile. Interestingly, the inequality significantly declined over time until it disappeared in 2016.The RCI measure also confirmed NMR inequality in 2011, with the poor enduring most of the neonatal deaths (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Inequality in neonatal mortality rate measured by the different inequality measures across the four dimensions of inequality, between 2000 and 2016 EDHS.

https://doi.org/10.1371/journal.pone.0234483.t002

Both the relative and absolute measures indicated that NMR appeared to be largely concentrated among infants of women with no formal education. In 2000, the Ratio showed that NMR among the uneducated women was about 2.5 times higher than that for women who had completed secondary education. PAR and RCI also showed clear disparity in NMR, again disfavouring the illiterate subgroup. By contrast, estimates of the SII revealed that NMR was largely prevalent among women with higher educational attainment in both 2000 and 2011. While educational inequality in NMR had been falling over time based on PAR and D measures, other summary measures showed fluctuations in the results (Table 2).

While no urban-rural differentials of NMR were documented using the simple measures of Ratio and Difference, PAR showed that NMR was higher in rural settings in 2000, 2005, and 2011 but disappeared in 2016. Significant sex disparities were evident as NMR was prevalent among male neonates in all the surveyed time periods by all the measures(Table 2).

Discussion

To our knowledge, this is the first attempt to comprehensively evaluate the nature of NMR disparity in Ethiopia over a sixteen year period. The findings showed inequalities in NMR favouring socioeconomically advantaged subpopulations (i.e. wealthier and more educated), female neonates and those living in urban settings.

Patterns of inequality varied widely by the type of inequality measure analyzed over the time period studied. Estimates of PAR showed that wealth-related inequalities in NMR fell steadily throughout the study period until it disappeared in 2016. Between 2000 and 2011, on average, NMR decreased by 9 neonatal deaths per 1000 live births. However, based on estimates of SII, the study found wealth-related inequality in 2011 and 2016, but not in 2000 and 2005. The inequality significantly increased in 2011 compared with levels in 2000 and 2005; it subsequently decreased between 2011 and 2016. The RCI similarly demonstrated the pro-rich scenario of neonatal mortality in 2011.

Even though findings differed by type of measures, the study confirmed that wealth-based inequality generally narrowed over time and even disappeared in 2016 for most measures. The slowly improved coverage of maternal and child health care services over time [19], even if the gap in use of these services between the poor and rich persisted or widened [21], could suggest a possible contribution to the noticeable drop in the wealth-based inequality. Moreover, the region’s economic growth over the last decade and the significant fall in the proportion of people living in extreme poverty [14] could be other drivers for decreasing the poor-rich NMR gap over time. Our finding that NMR disproportionately impacts the poor was aligned with some prior work [29] but not with others [13].

NMR was concentrated among neonates born to women with no education, coinciding with findings from available literature [5]. Educational inequality in NMR either fell steadily or remained unchanged depending on the type of summary measures used. The 2016 survey year appeared to be the most equitable, where NMR was equally prevalent among the three subgroups of education by most measures of inequality. A notable exception was the resulting SII educational inequality in 2000 and 2011, favouring the uneducated subpopulation. Conflicting with the findings obtained through other measures and from prior studies [5], this may be anomalous, requiring further study.

Large urban-rural disparity in NMR appeared during the first three EDHS time points. However, it had been consistently decreasing throughout the studied years until it disappeared in 2016. Between 2000 and 2011, it was possible to reduce the country’s NMR between 2 to 12 deaths per 1000 live births (on average) had the rural setting been on par with the urban areas in terms of neonatal mortality reducing interventions. The disproportionately higher clustering of neonatal mortality reducing interventions (such as access to skilled birth attendants) in urban areas can explain the pro-urban profile of NMR in 2000, 2005 and 2011 [24–26].The disappearance of inequality by 2016 may be due to improved coverage of these interventions in rural settings over time [19].This finding was consistent with prior studies that confirmed the pro-urban nature of NMR [30] but not with others [31].

The relative and absolute sex inequalities were notable. NMR was prevalent among male neonates in all the surveyed periods. In 2016, when sex-based disparity was most pronounced, male neonates experienced, on average, an excess of 22 deaths per 1000 live births and endured about twice as many deaths when compared with female neonates. Unlike other equity stratifiers based inequality, sex-based disparity did not improve with time. Other studies have found similar NMR inequality results to the advantage of female as compared to male neonates [31]. Why, then, are male neonates at higher risk of dying than female neonates? Yet another aspect to explore in future studies.

The nature of NMR disparities in our study can be analyzed using the “inverse equity hypothesis” proposed by Victora CG et al (2000) [32]. The hypothesis provides an explanation for the trends in health inequity, over time, between the poor and rich, and postulates that inequities become worse before getting better. It suggests that during initial scale up of public health interventions, there is higher uptake and earlier adoption among the rich subgroups, before the poorer subgroups even start utilizing the services. With time, the morbidities and mortalities decline among the rich while the poorer subgroups have not yet fully begun to use the services. Subsequently, the morbidities and mortalities of poorer subgroups are not seen to decrease at the level experienced by the rich. This difference in uptake of interventions between the two groups during the early period leads to the wider inequities in health indicators. Once the interventions reach the poor, the rich have already achieved the “minimum acceptable level” of the unfavourable indicator of interest, like NMR, thus inequities between the two groups narrow in terms of those indicators.

Congruent with the proposed hypothesis, utilization of maternal health care services, such as antenatal care or skilled delivery, has been disproportionately higher for subgroups who are socioeconomically better-off and live in urban areas in Ethiopia [28]. Between 2000 and 2011, these services were used mainly by the rich, more educated and urban populations while the poorer, uneducated individuals living in rural settings did not use them. Consequently, the NMR was higher among the disadvantaged groups during this period. In 2016, with the scale up of maternal health services in Ethiopia, uptake increased among the disadvantaged subgroups leading to decreasing NMR. The wealthier subgroups had reached their minimum acceptable level of NMR at this point; thus, the pace of NMR reduction was less, allowing for poorer subgroups lagging behind to close the NMR inequality gap. Available evidence also suggests that scale-up of complementary interventions such as facility delivery can lead to the decrease of neonatal deaths [33]. Given the demonstrated impact of such interventions, the government of Ethiopia may consider expanding some or all of these interventions in order to meaningfully reduce the burden of neonatal death in the country.

Our study offers policy-relevant information regarding the issue of neonatal mortality from the perspective of the socioeconomic status of the people. Equitable economic policies need to be an integral element of health policies in the country in order to benefit the poor and middle-class sub populations and increase their use of essential maternal health care services. This, in turn, helps narrow the poor-rich gap in health outcome indicators, such as NMR, and ultimately eliminate this income driven disparity. Another important policy implication of our work points to the demonstrated influence of education on health.

Education has to be a fundamental element of public health interventions in the country. It is a powerful tool in preventing health care disparities by interrupting the ongoing impact of poverty [34]. The government of Ethiopia can focus on accelerating coverage of at least secondary education to alleviate disparities in NMR between the subpopulations, which, in turn, can translate to the national NMR reduction. By incorporating equitable income and education policies into health policy, the country has the potential to succeed in attaining not just the SDG for NMR, but in reaching other key indicators as well.

The strength of this study is two-fold. First, inequality in NMR was assessed through different summary measures; adopting a number of inequality measures in the study helped showcase various possible patterns of NMR inequality. Second, the findings used an established high quality WHO monitor database prepared by experts, contributing to the quality of evidence analyzed and reported in this paper. However, the study also had some limitations. It did not identify factors that contribute to the measured NMR inequality. Future studies should consider using a decomposition approach to study factors that contribute to the NMR inequality to see whether their contribution changes with time. Furthermore, the EDHS dataset-based findings cannot be applied to areas smaller than the subnational regions and city administrations. Inequality in NMR should be considered in smaller areas such as villages, towns, districts and zones.

Conclusions

The study found inequalities in NMR across the different equity stratifiers. Encouragingly, however, NMR inequality in Ethiopia has been on the decline according to sixteen years of studied data. The improvement in inequality of NMR over time may be associated with many factors such as Ethiopia’s dedication to meet the NMR SDG of 12 deaths per 1000 newborns by 2030, improved economic growth, and improved coverage of maternal and child health services. Inequality studies such as this are useful to inform equity-oriented interventions aimed at eliminating inequality; they signal affected subgroups to strategically target programming and alleviate the problem. Policy makers should prioritize the subpopulations experiencing the highest NMR without ignoring the whole population. Despite gains made so far, much work is still needed to minimize, if possible, eliminate, NMR disparity across different equity stratifiers in Ethiopia.

Acknowledgments

The authors acknowledge WHO for making this equity analysis software freely accessible.