Fast track — ArticlesWHO estimates of the causes of death in children
Introduction
Child survival efforts can be effective only if they are based on reasonably accurate information about the causes of deaths. Cause-of-death information is needed to prioritise interventions and plan for their delivery, to ascertain the effectiveness of disease-specific interventions, and to assess trends in disease burden in relation to national and international aims.
Until 2004, global and regional estimates of child deaths attributable to specific causes were available from two major sources. First, technical groups, often collaborating with the United Nations (UN), published estimates of the number of deaths in children younger than age 5 years directly attributable to single diseases such as diarrhoea,1 measles,2 and HIV/AIDS.3 Second, WHO coordinated a yearly process that brought together single-cause estimates from WHO technical units, and reconciled them into a single proportional cause-of-death distribution that represented 100% of deaths in children younger than age 5 years worldwide.4 These cause-specific proportions were then applied to the total number of deaths in children younger than age 5 years, which we will refer to as the envelope, derived from country-specific mortality rates—ie, probability of dying before age 5 years. WHO,5 UNICEF,6 World Bank,7 and the UN Population Division8 estimated the mortality envelope in children aged younger than 5 years independently until 2004, producing different figures. WHO and UNICEF have now agreed on and applied standard procedures to produce a consistent envelope of deaths in this age-group for 2000–03.9
Increased attention to evidence-based decision making in public health and the global commitment to the millennium development goals10 has generated renewed interest in strengthening child-health epidemiology as a foundation for improved efforts to reduce mortality in children younger than age 5 years. Beginning in 1998, WHO sought outside expertise to improve the quality of this area of work. The initial focus was on deaths from pneumonia, but quickly expanded to include other major causes of child mortality as well as morbidity, the role of undernutrition as an underlying cause of death,11, 12 and crosscutting methodological issues.
The Child Health Epidemiology Reference Group (CHERG) was established in 2001 by WHO. Their mandate was to review and improve data collection, methods, and assumptions underlying estimates of the cause-specific proportional distribution of child deaths for the year 2000, beginning with the major causes of child mortality worldwide: pneumonia; diarrhoea; malaria; causes of death in the first 28 days of life, referred to here as neonatal causes; and measles. The CHERG investigators have worked in small disease-specific groups since that time to apply systematic methods and develop new estimates for deaths due to these causes.
In 2004, WHO initiated a consultative process to incorporate CHERG results into the broader WHO mortality estimates for children younger than age 5 years. The final proportional distribution of deaths by cause was then applied to the mortality envelope for this age-group. Our aim is to describe the methods and results of these efforts: new estimates for mortality by cause in children younger than age 5 years, worldwide and in the six WHO regions, as the average of yearly estimates for the period from 2000–03.
Section snippets
Methods
We focus on the WHO process for development of improved estimates of cause-specific mortality in children younger than age 5 years. More detailed information on the methods used by the CHERG working groups for every cause or methodological issue is available elsewhere,13, 14, 15, 16, 17 as are reports on the review and revision of the mortality envelope for children younger than age 5 years,9 the development of cause-specific estimates of morbidity in this age group due to acute respiratory
Overview of methods used by CHERG working groups
All of the CHERG working groups applied the same general set of procedures to produce their estimates, although these procedures evolved over time and the cause-specific epidemiological models also differed based on the characteristics of the diseases being studied. First, every group of investigators reviewed the methods, assumptions, and results from previous estimates and developed models specifying the key epidemiological variables for every cause or group of causes and their association to
Incorporation of CHERG results into WHO estimates
The CHERG results were incorporated into broader WHO estimates of deaths in children younger than age 5 years by cause with a standard set of procedures (figure 1). Full details are available elsewhere,9 but the process is described briefly below.
Estimation of envelope
WHO has updated life tables for its member states every year since 2000, by using mortality rates for children younger than age 5 years jointly produced by WHO and UNICEF.9 The number of total deaths in this age group was estimated by applying the mortality rates from the WHO abridged life tables to the population estimates obtained from the UN Population Division.8 The envelope for neonatal deaths was estimated with mortality data from surveys and vital registration systems in countries where
Proportional distribution of deaths by cause
WHO uses information on the distribution of child deaths by cause from vital registration systems in the 72 countries where those systems are judged to be sound (based on standard application of coding that follows the rules of the International Statistical Classification of Diseases and Related Health Problems (ICD) for death certificates and reliable diagnostic procedures) and have coverage rates of 85% or above. These data are all from high-income and middle-income countries.
Cause-of-death
Estimation of deaths by cause, 2000–03
The final estimates of the proportional distribution of deaths by cause were applied to the average yearly mortality envelope for all countries without adequate vital registration data for 2000–03. These national estimates were summed to arrive at regional and global deaths by cause in children aged younger than 5 years. A single average yearly estimate was prepared for the period 2000–03 because 2000 was the reference year for all CHERG estimates and only estimates of HIV/AIDS have been
Role of the funding source
The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Results
The new estimates (figure 2) show that worldwide, 73% of deaths in children younger than age 5 years are attributable to six causes: pneumonia (19%), diarrhoea (18%, which includes 17% of deaths in children 1–59 months and 3% of neonatal deaths), malaria (8%), neonatal sepsis or pneumonia (10%), preterm delivery (10%), and asphyxia at birth (8%). The four communicable disease categories account for more than half (54%) of all child deaths. Sepsis or pneumonia in neonates and pneumonia in older
Discussion
The results presented here provide important information, at new (albeit imperfect) degrees of accuracy, about the causes of child mortality worldwide and in the six WHO regions. What are their implications for policies, public-health priorities, planning, and practice?
The new estimates show that worldwide more than seven in ten of the 10·6 million annual deaths in children younger than age 5 years are attributable to six causes, and that four communicable disease categories account for more
References (36)
- et al.
Meeting international goals in child survival and HIV/AIDS
Lancet
(2002) - et al.
Undernutrition as an underlying cause of child deaths associated with diarrhoea, pneumonia, malaria, and measles
Am J Clin Nutr
(2004) - et al.
Estimates of world-wide distribution of child deaths from acute respiratory infections
Lancet Infect Dis
(2002) - et al.
4 million neonatal deaths: When? Where? Why?
Lancet
(2005) - et al.
Where and why are 10 million children dying every year?
Lancet
(2003) - et al.
How many child deaths can we prevent this year?
Lancet
(2003) - et al.
The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000
Bull World Health Organ
(2003) - et al.
The global burden of measles in the year 2000: a model that uses country-specific indicators
J Infect Dis
(2003) - et al.
Global burden of disease in 2002: data sources, methods and results
(2003) World health report 2004: changing history
(2004)
The state of the world's children 2004
World development indicators 2004
World population prospects: the 2002 revision
General assembly, 56th session: road map towards the implementation of the United Nations millennium declaration—report of the Secretary-General
Undernutrition as an underlying cause of malaria morbidity and mortality in children less than five years old
Am J Trop Med Hyg
CHERG diarrhoea mortality report
CHERG malaria mortality report
Cited by (1616)
Multi- and extensively drug resistant Escherichia coli isolated from South African children under 5 years old with diarrhoea
2024, Diagnostic Microbiology and Infectious DiseaseMathematic skills in low birth weight or preterm birth: A meta-analysis
2024, Educational Research ReviewMalnutrition among under-five children in Nepal: A focus on socioeconomic status and maternal BMI
2024, Clinical Epidemiology and Global HealthA constant proportional caputo operator for modeling childhood disease epidemics
2024, Decision Analytics Journal