Child mortality
Child mortality refers to the deaths of infants and children under five years of age, with the under-five mortality rate (U5MR)—defined as the probability of dying between birth and age five, expressed per 1,000 live births—serving as the standard metric for assessment.[1] In 2023, the global U5MR reached 37 deaths per 1,000 live births, reflecting a 59% decline from 93 in 1990 and averting an estimated tens of millions of deaths through widespread improvements in living standards, sanitation, nutrition, and medical interventions such as vaccines and antibiotics.[1][2] This progress has been most pronounced in regions with economic development and effective public health policies, though rates remain elevated in low-income areas burdened by poverty, conflict, and inadequate infrastructure, exceeding 70 per 1,000 in sub-Saharan Africa.[3] Leading causes include neonatal conditions like preterm birth complications and birth asphyxia (accounting for about half of under-five deaths), followed by infectious diseases such as pneumonia, diarrhea, and malaria, often exacerbated by malnutrition and limited access to clean water and treatment.[2][3] Despite these advances, approximately 4.8 million children died before age five in 2023, underscoring persistent challenges in scaling proven interventions amid demographic pressures and governance failures in high-burden regions.[3]Definitions and Scope
Key Metrics and Classifications
The under-five mortality rate (U5MR) is the primary metric for assessing child mortality, defined as the probability that a newborn will die before reaching exactly five years of age, expressed per 1,000 live births.[1][4] This rate integrates risks across infancy and early childhood, capturing vulnerabilities from birth through age four, and serves as a key indicator for child health and development progress under Sustainable Development Goal 3.2.1.[5] The infant mortality rate (IMR) measures deaths occurring before one year of age per 1,000 live births, encompassing both neonatal and post-neonatal periods.[6] Neonatal mortality specifically refers to deaths within the first 28 days of life (often approximated as 0-30 days in surveys), while post-neonatal mortality covers deaths from 1 to 11 months. These sub-metrics highlight distinct etiological phases, with neonatal deaths often linked to birth complications and post-neonatal to infectious diseases or malnutrition. Child mortality for ages 1-4 years is the probability of death in that interval per 1,000 surviving children, completing the under-five framework by isolating toddler and preschool risks such as diarrhea, pneumonia, and injuries. Classifications extend to perinatal mortality, which includes stillbirths and early neonatal deaths (first week), though it is sometimes distinguished from live-birth-focused metrics.[3] The United Nations Inter-agency Group for Child Mortality Estimation (UN IGME) standardizes these through Bayesian models applied to vital registration, household surveys, and censuses, ensuring comparability across low-data contexts.[5]Measurement Challenges and Biases
In many low- and middle-income countries, vital registration systems are incomplete or absent, leading to substantial underreporting of child births and deaths, particularly neonatal ones, which complicates accurate measurement of under-5 mortality rates (U5MR).[7] For instance, in developing nations, fewer than 10% of deaths may be officially recorded in some regions, forcing reliance on household surveys like Demographic and Health Surveys (DHS), which use retrospective birth histories from mothers.[8] These surveys often underestimate mortality due to recall bias, where respondents omit deaths of children born further in the past or underreport neonatal events, with studies showing discrepancies of up to 20-30% compared to prospective surveillance data.[9] [10] Methodological differences exacerbate biases; for example, summary birth histories in surveys tend to produce lower U5MR estimates than full histories or direct vital records because of omission of early or high-mortality events.[11] Age misreporting and selective non-response further distort data, as families with deceased children may be harder to contact or less willing to participate, introducing selection bias tied to mortality risk.[12] In conflict or rural areas, logistical challenges amplify these issues, with underreporting of stillbirths and early neonatal deaths reaching rates where true magnitudes are estimated to be 50% higher than reported.[13] Estimation agencies like the United Nations Inter-agency Group for Child Mortality Estimation (UN IGME) address these gaps through Bayesian B-splines and cohort component models that adjust for biases such as HIV prevalence or survey omissions, but these introduce parametric uncertainties and reliance on assumptions about trends and covariates.[14] [15] UN IGME estimates, while comprehensive, diverge from alternatives like the Institute for Health Metrics and Evaluation (IHME), with global U5MR differences of about 5% in recent years (e.g., 56.7 vs. 53.9 per 1,000 live births in 2010), reflecting varying model specifications and data weights.[16] [17] Such discrepancies highlight systemic challenges in data quality, where modeled outputs from international bodies, though peer-reviewed, can mask local inaccuracies if primary sources are politically incentivized to underreport progress toward targets like the Sustainable Development Goals.[18] Overall, these challenges result in U5MR uncertainty intervals that can span 10-20% in high-burden countries, underscoring the need for improved civil registration and vital statistics (CRVS) systems to reduce dependence on biased surveys and models.[19] While UN IGME methods enhance comparability, their reliance on adjusted survey data from potentially under-resourced national systems introduces a layer of estimation error that users must interpret cautiously, especially given incentives for optimistic reporting in aid-dependent contexts.[20]Historical Context
Pre-Industrial and Early Modern Eras
In pre-industrial societies, prior to the mid-18th century, child mortality rates—defined as deaths under age five—typically ranged from 300 to 500 per 1,000 live births, meaning 30-50% of children did not survive early childhood.[21] These estimates derive from historical demography, including family reconstitution studies and skeletal analyses, which reveal consistent patterns across agrarian economies where high fertility rates compensated for substantial losses.[22] Infant mortality, concentrated in the first year, accounted for roughly half of these deaths, driven by perinatal complications, diarrheal diseases, and respiratory infections amid absent sanitation infrastructure and medical knowledge.[23] European parish registers from the early modern period (c. 1500-1800) provide granular evidence, particularly from England and the Low Countries. In England, infant mortality stabilized at approximately 140-250 per 1,000 live births between 1538 and 1837, with negligible declines until smallpox inoculation emerged late in the era; post-neonatal deaths (1-12 months) often exceeded neonatal ones due to weaning onto contaminated solid foods.[24] [25] Similar rates prevailed in urban centers like London, where overcrowding amplified epidemics, yielding under-five mortality near 400 per 1,000 in the 17th century.[26] Rural areas fared marginally better, but endemic threats like plague recurrences and tuberculosis persisted, underscoring the era's vulnerability to seasonal and microbial pressures without vaccines or antibiotics.[21] Socioeconomic disparities influenced outcomes, with lower-class children facing 20-50% higher risks than elites, attributable to nutritional deficits and exposure in shared living spaces; for instance, framework knitters in English parishes exhibited elevated early childhood mortality from 1600 onward.[27] In non-European contexts, such as Ottoman or Chinese records where fragmentary, analogous rates of 200-300 infant deaths per 1,000 suggest universality in pre-industrial settings, tied to subsistence agriculture and limited public health measures.[22] Overall, these periods reflect a demographic regime where child survival hinged on innate immunity and maternal breastfeeding duration, with little technological mitigation until the Enlightenment's nascent hygiene reforms.[23]Industrialization and 20th-Century Declines
The industrialization era, beginning in the late 18th century in Britain and spreading across Europe and North America, initially correlated with rising child mortality in urban centers due to rapid population growth, overcrowding, and inadequate sanitation infrastructure, which facilitated the spread of waterborne diseases like cholera and typhoid.[28] For example, in British cities around 1800-1850, under-5 mortality rates often exceeded 25-30% amid these conditions, though rural areas maintained lower rates closer to pre-industrial levels of approximately 40-50%.[21][29] Public health interventions in the mid-to-late 19th century reversed these trends through engineering solutions such as sewerage systems, chlorinated water supplies, and waste management, which reduced diarrheal and enteric infections—the primary killers of children.[30] In England and Wales, infant mortality (under age 1) fell from about 150 per 1,000 live births in 1840 to around 100 by 1900, driven by these sanitation reforms and supplementary measures like milk purification to combat tuberculosis and pasteurization to prevent bacterial contamination.[22] Similar patterns emerged in Sweden, where under-5 mortality declined from roughly 30% in 1800 to 10% by 1900, reflecting broader improvements in hygiene and living standards.[22] In the 20th century, child mortality in industrialized nations plummeted further, with under-5 rates in the United States dropping from approximately 200 per 1,000 in 1900 to under 30 by 1960.[31] This acceleration stemmed from multiple factors: enhanced nutrition reducing susceptibility to infections, compulsory education limiting exposure to communicable diseases, and institutional monitoring of vital statistics enabling targeted policies.[30] Notably, nearly 90% of the decline in U.S. infectious disease mortality among children occurred before 1940, prior to the mass deployment of antibiotics or most vaccines, underscoring the foundational role of non-pharmaceutical interventions like sanitation and poverty reduction.[32] Post-1940s medical breakthroughs amplified these gains: antibiotics such as penicillin curtailed bacterial pneumonias and meningitides, while vaccines against diphtheria (1920s), pertussis (1930s-1940s), and measles (1960s) eliminated epidemic peaks that had persisted despite prior hygiene advances.[33][34] By 2000, under-5 mortality in high-income countries had reached below 1%, a level unimaginable a century earlier, though disparities lingered between socioeconomic groups due to uneven access to these cumulative protections.[31] These declines exemplify how causal chains—from economic growth enabling infrastructure to scientific validation of germ theory—drove empirical reductions, rather than isolated interventions.[35]Post-1990 Global Trends
The global under-5 mortality rate declined from 93 deaths per 1,000 live births in 1990 to 37 in 2023, representing a 59% reduction.[2] This progress averted an estimated 150 million child deaths compared to 1990 levels, driven by expanded immunization programs, improved sanitation, and better access to basic healthcare.[1] The absolute number of under-5 deaths fell from approximately 12.6 million in 1990 to 4.8 million in 2023, though population growth partially offset per capita gains.[3] Millennium Development Goal 4 (MDG 4), which aimed to reduce under-5 mortality by two-thirds from 1990 to 2015, spurred international efforts including Gavi vaccine alliances and integrated management of childhood illnesses.[36] By 2015, the rate had dropped to around 43 per 1,000 live births, achieving over half the target but falling short of the full two-thirds reduction.[37] Success varied regionally: East Asia and the Pacific saw declines exceeding 70%, while sub-Saharan Africa achieved about 50%, remaining at 76 per 1,000 in 2015 due to persistent challenges like HIV prevalence and conflict.[38] Post-2015, under Sustainable Development Goal 3.2, which targets under-5 mortality below 25 per 1,000 by 2030, progress slowed, with the annual reduction rate dropping from 3.7% in 2000–2015 to 2.2% in 2015–2023—a 42% deceleration.[39] Neonatal deaths, comprising nearly half of under-5 mortality, declined more slowly than post-neonatal rates, reflecting gaps in maternal and newborn care.[3] The COVID-19 pandemic exacerbated this trend, disrupting services and contributing to an estimated 8.7 million additional child deaths between 2020 and 2021 beyond pre-pandemic projections.[40] Regional disparities persist, with sub-Saharan Africa accounting for 57% of global under-5 deaths in 2023 despite comprising 29% of births, at a rate of 71 per 1,000.[39] In contrast, Latin America and Eastern Europe achieved rates below 20 per 1,000 by 2023, correlating with higher GDP per capita and urbanization.[38] These trends underscore that while scalable interventions like oral rehydration and vaccines yielded broad gains, inequities in resource allocation and governance limit convergence.[31]Current Epidemiology
Global and Regional Rates
In 2023, the global under-five mortality rate (U5MR)—defined as the probability of dying between birth and age five, expressed per 1,000 live births—reached 37 (uncertainty interval 35–41), reflecting a 52% reduction from 77 in 2000 and a 59% drop from 93 in 1990. This equates to approximately 4.8 million under-five deaths worldwide, with 2.3 million occurring in the neonatal period (first 28 days of life). Progress has decelerated in recent years, with annual reductions averaging just 1.7% since 2015, compared to 3.7% from 2000 to 2015, amid challenges like conflict, climate impacts, and uneven access to interventions.[39][2]00501-4/abstract) Regional variations underscore stark inequities, driven by differences in healthcare infrastructure, sanitation, nutrition, and economic development. Sub-Saharan Africa accounts for over half of global under-five deaths despite comprising about 30% of live births, with a U5MR of 64 per 1,000 live births—nearly 20 times higher than in high-income regions. South Asia follows as the second-highest burden, though its rates have fallen faster due to scaled vaccinations and maternal health improvements. In contrast, Eastern Asia and Latin America exhibit rates closer to those in developed areas, reflecting stronger public health systems.[41][3][42]| Region | U5MR (per 1,000 live births, 2023) |
|---|---|
| Sub-Saharan Africa | 64 |
| South Asia | 29 |
| Latin America & Caribbean | 13 |
| East Asia & Pacific | 10 |
| High-income countries | ~5 |