Key points

Introduction

Each year, millions of children die, the vast majority in poor countries. Tragically, most of these deaths are preventable with technologies that are currently available and recommended for universal implementation. Progress is being made: 5.9 million children younger than 5 years died in 2015, down from 12.4 million in 1990. This reduction, although substantial, was insufficient to meet Millennium Development Goal (MDG) 4: reduce child mortality by two-thirds between 1990 and 2015. In 2000, 189 countries endorsed the MDGs, which consisted of 8 specific goals to reduce poverty and improve health and development. In addition to MDG4, the other 2 health-related goals, MDG3 (reduce maternal mortality) and MDG6 (reduce infectious diseases), were not met. The “post MDG era” has ushered in the Sustainable Development Goals (SDGs), a much broader array of 17 ambitious goals with 169 targets. One of the 17 SDGs specifically relates to health: SDG3, and its 9 targets cover a much wider scope of problems (eg, injuries, mental health, and chronic noncommunicable diseases) than was tackled by the MDGs. SDG3.2, the target related to child health, calls for ending “preventable deaths of newborns and children under 5 years of age, with all countries aiming to reduce neonatal mortality to 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” by 2030.

This article provides an overview of the scope and causes of childhood deaths, interventions currently recommended to combat these killers, trends in child mortality, and potential reasons for these trends through the lens of the global potential to meet SDG3.

Introduction

Each year, millions of children die, the vast majority in poor countries. Tragically, most of these deaths are preventable with technologies that are currently available and recommended for universal implementation. Progress is being made: 5.9 million children younger than 5 years died in 2015, down from 12.4 million in 1990. This reduction, although substantial, was insufficient to meet Millennium Development Goal (MDG) 4: reduce child mortality by two-thirds between 1990 and 2015. In 2000, 189 countries endorsed the MDGs, which consisted of 8 specific goals to reduce poverty and improve health and development. In addition to MDG4, the other 2 health-related goals, MDG3 (reduce maternal mortality) and MDG6 (reduce infectious diseases), were not met. The “post MDG era” has ushered in the Sustainable Development Goals (SDGs), a much broader array of 17 ambitious goals with 169 targets. One of the 17 SDGs specifically relates to health: SDG3, and its 9 targets cover a much wider scope of problems (eg, injuries, mental health, and chronic noncommunicable diseases) than was tackled by the MDGs. SDG3.2, the target related to child health, calls for ending “preventable deaths of newborns and children under 5 years of age, with all countries aiming to reduce neonatal mortality to 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” by 2030.

This article provides an overview of the scope and causes of childhood deaths, interventions currently recommended to combat these killers, trends in child mortality, and potential reasons for these trends through the lens of the global potential to meet SDG3.

Scope of the problem

Health is defined by the World Health Organization as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity.” Health is clearly more than just survival; however, as a starting point, it is hard to avoid the fact that more than 16,000 children are estimated to die each day. “Child,” as used in this article and as typically defined in the global health field, includes persons younger than 5 years because of their particular biologic and social vulnerability. The under-5 mortality rate (U5MR), defined as the number of deaths among children younger than 5 years per 1000 live births, is often used as an indicator of the health of a population more broadly. If conditions favor the health and welfare of this vulnerable group, the situation generally can be considered favorable for the overall society.

Age-specific mortality rates decline appreciably beyond 5 years. Ninety-nine percent of child deaths occur in low-income and middle-income countries (LMICs). A child born in sub-Saharan Africa (SSA) faces a 1 in 12 chance of dying before his or her fifth birthday compared with 1 in 140 for a child born in the United States and 1 in 167 for high-income countries (HICs) on average. The global child health community has started to focus on the broader picture of health, including morbidity, developmental disability, and long-term impacts on adult chronic disease and economic capacity. Unfortunately, child mortality remains as a tenacious problem requiring confrontation.

Major causes of child mortality

Forty-five percent of child mortality occurs in the neonatal period (first 28 days of life) ( Fig. 1 ). These deaths are largely preventable and their causes are discussed in detail in the article by Zulfiqar A. Bhutta and colleagues, “ Neonatal and perinatal infections ,” in this issue. Four problems are responsible for approximately 60% of postneonatal deaths: pneumonia, diarrhea, injuries, and malaria. Mortality from these causes is also mostly preventable with sustainable implementation of available interventions as described later in this article. Undernutrition, including lack of sufficient macronutrients (eg, protein, calories), micronutrient-deficient diets (eg, vitamin A, zinc, iron), and suboptimal breastfeeding practices, contributes to 45% of all child deaths. Undernutrition increases susceptibility to infectious diseases, reduces recovery from disease and injury, and is associated with longer-term sequelae, such as cognitive impairment, poor school performance, and noncommunicable diseases in adulthood.

Global causes of under-5 deaths in 2015.

(Reproduced with permission from Liu L, Oza S, Hogan D, et. al. Global, regional, and national causes of under-5 mortality in 2000-15: an updated systematic analysis with implications for the Sustainable Development Goals. Lancet 2016; 0140-6736.)

Undernutrition and the common direct causes of child deaths are not exotic tropical diseases, but rather diseases of poverty, brought under control a century ago in the United States and other HICs before the advent of vaccinations and antimicrobials, through social changes addressing crowding, sanitation, nutrition, and basic living conditions. Medical breakthroughs, such as antibiotics, immunizations, and insecticides (to prevent malaria, which was endemic in parts of the United States, for example) further accelerated gains in controlling disease and preventing childhood mortality.

Although less proximate, the social determinants of health, are perhaps even more important underlying causes of mortality. Children in rural areas have a 1.7-fold increased risk of dying compared with their urban counterparts. With increasing urbanization, however, urban slum dwellers face survival challenges. Services may not be available in rural areas, but urban slum dwellers may lack access due to cost, discrimination, and other factors and suffer overcrowded and unsanitary living conditions. Race, ethnicity, and gender are other important determinants.

Maternal education is arguably the most important determinant of child survival. Children whose mothers have no formal education are, on average, 2.8 times as likely to die before their fifth birthday compared with those with mothers with secondary or higher education. This is not categorical: for each additional year of maternal schooling, deaths drop by 9%. Although literacy and health literacy are important outcomes of women’s education, they do not fully explain the effect on child health; agency and decision-making power likely play important roles.

Approximately 10% of the world lives below the international poverty line: $1.90 per day. Absolute poverty is an extreme hardship, limiting access to the basic necessities to support survival (eg, essential medicines and health care, water, sanitation, adequate nutrition, education). Relative poverty is defined as large gaps between rich and poor within a society and is associated with worse societal health outcomes, especially among the poorer segments. This phenomenon has been demonstrated in both rich and poor countries. Overall, children living in the poorest fifth of households compared with the wealthiest quintile within any given country, face a 1.9-fold increased risk of mortality. Unfortunately, as described later in this article, income inequality has dramatically increased over the past few decades, along with increases in health disparities in many countries. For example, among 36 countries experiencing national declines in U5MR, half had an increased gap in child mortality between the wealthiest and poorest quintiles.

Interventions to reduce child mortality

The following section reviews currently recommended prevention and treatment interventions for tackling the leading childhood killers beyond the neonatal period and provides an overview of intervention coverage rates, defined as the proportion of individuals needing an intervention who receive it. For example, if the condition is diarrheal disease and the intervention is oral rehydration solution (ORS), the denominator would be children younger than 5 years with diarrhea, and the numerator all children with diarrhea who received ORS.

Pneumonia

Pneumonia is the leading cause of death after the neonatal period, killing more than 900,000 children annually. Pneumococcus and Haemophilus influenzae type b (Hib) are prevalent causes of childhood pneumonia, the most important etiologies of severe pneumonia in young childhood, and the cause of about half of childhood pneumonia deaths globally. Other common microbes include viruses, especially respiratory syncytial virus; other bacteria, particularly Staphylococcus aureus and Klebsiella pneumoniae ; and Mycobacterium tuberculosis , especially among individuals infected with the human immunodeficiency virus (HIV). Where HIV prevalence is high, Pneumocystis jiroveci is an important cause of childhood pneumonia deaths, despite recommendations for cotrimoxazole among HIV-infected individuals as an inexpensive and effective prophylaxis against P jiroveci pneumonia.

Prevention measures are important in reducing pneumonia incidence and case fatality. Undernutrition leads to reduced immunity and increased difficulty in clearing secretions due to weakened respiratory muscles. Optimal breastfeeding practices and adequate complementary feeding, including adequate micronutrients (especially those involved in immune protection, such as zinc), are important interventions to prevent the incidence of and case fatality from respiratory infections. For example, infants have a 15-fold and 2-fold greater risk of death from pneumonia if not breastfed or partially breastfed, respectively, in first 6 months of life compared with exclusively breastfed. Currently, only 43% of young infants globally are exclusively breastfed, and only 74% and 46% continue to breastfeed through the first and second years of life, respectively. Furthermore, only 19% of those 6 to 23 months old receive a minimally acceptable diet of complementary foods.

Household air pollution is associated with a 1.8-fold risk of contracting pneumonia and is largely caused by burning of solid fuels (eg, wood, charcoal, dung, crop waste) in dwellings for heat and cooking. These polluting fuels are used by 40% of the world’s population and primarily by poor households. Chimney stoves have been shown to reduce household air pollution by half and severe pneumonia by approximately 30%. Hand hygiene (washing with soap) is important for preventing the spread of respiratory infection and requires water security: access to sufficient quantities of water. This is a challenge for the many women and children who walk long distances to collect water for their households.

Vaccinations play a vital prevention role. Secondary bacterial pneumonia is a common sequelae of measles and pertussis infections. Measles and diphtheria-pertussis-tetanus (DPT) vaccines are inexpensive and effective; they were rolled out in 1976 as part of the original Expanded Program on Immunizations. The largest pneumococcus and Hib disease burden has been and continues to be in LMICs. Immunizations against these pathogens were incorporated into routine immunization schedules in HICs toward the end of the twentieth century. Cost and immunization system (eg, cold chain) expansion have impeded rapid scale-up in poor countries. Furthermore, serotype coverage in pneumococcal vaccines has favored markets in HICs. Hib vaccine was first introduced in an LMIC in 1997; 191 countries now incorporate Hib vaccine into national immunization schedules and global coverage is 64% ( Fig. 2 ). Pneumococcal vaccine rollout commenced less than a decade ago in LMICs; it is part of routine schedules in 129 countries and global coverage is 37%. Although coverage rates for measles and DPT (data not shown) vaccines are 85% and 86%, respectively, and have increased substantially over decades, their coverage rates have virtually stagnated since 2008 and are below the World Health Organization (WHO) and United Nations International Children’s Emergency Fund (UNICEF) goal of achieving 90% coverage by 2010 for routine vaccines. Furthermore, global and national coverage estimates mask variations and disparities in coverage within countries. A recent analysis demonstrated pro-rich and pro-urban inequities in immunization coverage in most LMICs and pro-male inequities in Southeast Asian LMICs. Some countries substantially reduced these inequities; however, they have increased in other countries. Differences between the wealthiest and poorest quintiles for immunization and other intervention coverage rates as global averages are depicted in Fig. 3 .

Data represent percent coverage globally except where indicated. ^a The 15% and 9% figures in the first blue bars represent data from 2011; wide-scale adoption into immunization schedules in LMICs commenced more recently, hence 2000 data are not available or negligible. ^b Excludes data from China. ^c Excludes data from China and India. ^d Based on data from SSA. The 4% figure is data from 2005. ^e Based on data from SSA.

( Data from Refs. )

Coverage of select interventions comparing highest to lowest wealth quintiles. Q1 is the poorest wealth quintile. Q5 is the richest wealth quintile.

(Data represent percent coverage and are global estimates. Immunization and ITN data are from Barros AJ, Ronsmans C, Axelson H, et al. Equity in maternal, newborn, and child health interventions in Countdown to 2015: a retrospective review of survey data from 54 countries. Lancet 2012;379(9822):1225–33; and The remainder are from UNICEF Data: Monitoring the Situation of Children and Women. Available at: data.unicef.org . Accessed November 25, 2016.)

Although prevention plays an important role in pneumonia control, appropriate antibiotic treatment is critical for reducing case fatality. Despite increasing resistance to readily accessible and inexpensive first-line amoxicillin, in vivo efficacy continues to be good, at least at this time. Prompt and appropriate case management hinges on the following: (1) parent/caretaker recognition of symptoms, especially tachypnea and retractions, for which prompt health care should be sought, (2) access to appropriate care without delay (ie, services are available, geographically proximate, affordable, good quality, and nondiscriminatory), (3) accurate diagnosis and treatment by health workers, and (4) availability and affordability of treatment and completion of a full therapeutic course. However, parental recognition that fast breathing and retractions require urgent medical attention is inadequate and fewer than 65% of children with pneumonia symptoms are taken for appropriate health care. Fewer still actually receive antibiotics. Tragically, there has been little improvement in these coverage statistics over the past decade (see Fig. 3 ), especially in SSA.

Use of lay health workers (LHWs) to diagnose and treat pneumonia in the communities where children live, especially in rural areas with lack of access to health facilities, is increasingly showing promise. Studies have consistently shown that LHWs trained in pneumonia case management can accurately identify and effectively treat pneumonia. However, in 2010 fewer than one-third of SSA countries had policies in place that allow LHWs to treat children with pneumonia, fewer than 20% had programs in place to implement community case management, and fewer than 10% had scaled up such programs to a national level. Policies and programs conducive to improving access to and quality of care in communities and in health facilities and efforts to educate families about danger signs for which to seek care, in addition to improved coverage of prevention interventions, are critical to reducing pneumonia mortality.

Diarrheal disease

The number of children dying from diarrhea has declined dramatically over the past couple of decades; however, diarrhea remains the second leading cause of child mortality beyond the neonatal period, causing more than 500,000 child deaths annually. These deaths are almost exclusively in LMICs. Indeed, a death from diarrhea in the United States would justifiably raise alarm; it should be no less acceptable elsewhere.

The decline in diarrhea mortality is unlikely due to prevention measures, as incidence rates have remained stable over the past couple of decades and account for nearly 1.7 billion child episodes annually. The important impacts of childhood enteric diseases on morbidity are increasingly being recognized. Specifically, environmental enteric dysfunction (EED) is thought to be caused by enteric infections (clinically apparent or asymptomatic) and/or exposure to a preponderance of nonpathogenic intestinal microbes due to exposure to fecally contaminated environments, which is common in settings without adequate sanitation facilities. EED contributes to undernutrition, itself a risk factor for infectious disease acquisition and mortality; thus the vicious cycle of malnutrition, infections, and EED ( Fig. 4 ).

The vicious cycle of intestinal infection, EED, and undernutrition.

( Adapted from Denno D. Global child health. Pediatr Rev 2011;32(2):e25–38.)

There is much overlap between the interventions to combat pneumonia and diarrhea, such as hand hygiene, adequate nutrition, and improved care seeking and case management. Rotavirus, Shigella , Cryptosporidium , and enterotoxigenic Escherichia coli are important microbiologic causes of moderate-to-severe diarrhea, and the latter 2 organisms plus typical enteropathogenic E coli are predominant causes of mortality. These microbes are spread by the fecal-oral route via hands, food, utensils, flies, and water. Improvements in handwashing with soap, sanitation (ie, safe feces disposal), and water quality could reduce diarrhea risk by 48%, 36%, and 17%, respectively, and prevent other water-related diseases. “Moreover, water, hygiene and sanitation have other important benefits, including the emancipation of women from drudgery and the enhancement of human dignity.”

At 91%, the 2015 global coverage rate for improved water sources (those protected from fecal contamination, such as piped water, boreholes, protected wells, and rainwater) met the MDG7 water target. However, more than 663 million people still use unimproved sources; nearly half in SSA and one-fifth in South Asia. Moreover, this indicator does not describe water security or quality. For example, at least 1.8 billion people worldwide are estimated to drink water that is fecally contaminated. The situation is more dire for sanitation, an intervention of even greater import for preventing enteric infections. Only 68% of the world’s population has access to improved sanitation facilities (those that hygienically separate human waste from human contact, such as units that flush or pour-flush into sewer or septic systems or pit latrines); the MDG7 sanitation target was not achieved. The situation is worse for rural dwellers, and South Asian and SSA individuals where coverage is 51%, 47%, and 30%, respectively. SDG6 contains 8 targets and covers the water cycle, water quality, and waste management more broadly. However, both MDG7 and SDG6 lack any within-country equity measures, by socioeconomic status, for example. Furthermore, neither include hygiene targets, which, as described previously, are arguably the most important “WaSH” (water, sanitation, and hygiene) interventions for preventing diarrheal and other infections.

Vaccines hold promise for preventing diarrhea due to specific agents. In 2006, Mexico was one of the first LMICs to rollout rotavirus vaccine; coverage rates there are currently 81% compared with 23% globally (see Fig. 2 ). Eighty-four countries have incorporated rotavirus vaccine into routine schedules. Vaccines against cholera and enterotoxigenic E coli are under development.

Oral rehydration therapy (ORT) is a key treatment intervention: the sugar-salt composition prevents and corrects dehydration and electrolyte losses. ORT is estimated to save the lives of 1 million children annually. ORT uses prepackaged ORS mixed with water or the use of appropriate other fluids, such as homemade solutions. The recommendation to continue feeding through illness has replaced the old medical notion of “rest the gut.” Continued feeding promotes gut recovery and mitigates the impact of infection on growth. International guidelines also recommend an oral zinc course for all children who contract diarrhea and live in areas with a presumed high prevalence of zinc deficiency (ie, most low-income countries). Treatment with zinc reduces illness severity and duration and decreases the likelihood of diarrheal episodes in subsequent months. Antibiotics play a limited role in current diarrhea treatment guidelines, with the rationale that most illnesses are viral, that outcomes of some bacterial enteric infections can be worsened by antibiotics (eg, prolonged carrier state for some Salmonella infections), and due to concerns about promotion of antibiotic resistance. Antibiotics are currently recommended only for dysentery, previously a sensitive marker of Shigella infections, which does warrant antimicrobial treatment. However, with shifting Shigella serotype patterns and waning Shigella dysenteriae prevalence, dysentery is no longer a reliable indicator of shigellosis. Reconsideration of indications for antibiotics may be warranted.

ORS has been the cornerstone of diarrhea treatment for decades. Despite this, coverage rates are low, especially among the poorest quintile, and improvements have been sluggish (see Figs. 2 and 3 ). Zinc treatment was first recommended in 2004, but only 3 countries have coverage rates exceeding 20%: Bangladesh (44%), Nepal (31%), and Malawi (28%). The unacceptable death toll from diarrhea is a tragic reminder of the work that remains in delivering known, effective prevention and treatment to children most in need.