Technical Notes: Trends in Mortality and Life Expectancy

VARIABLE DEFINITIONS AND METHODOLOGY:
Infant Mortality Rate is the probability of a child dying between birth and one year of age expressed per 
1,000 live births. The indicator is used as a measure of children's well-being and the level of effort being 
made to maintain child health--over three-quarters of child deaths in the developing world are caused by 
diseases that can prevented or cured by low-cost interventions such as immunization, oral rehydration 
therapy (ORT), and antibiotics.

Under-Five Mortality Rate is the probability of a child dying between birth and age five expressed per 
1,000 live births. The data is reported by countries from their national civil registration systems. Some are 
estimates prepared by national and international statistical services from population surveys. The 
completeness and accuracy of the data which these systems produce vary from one country or area to 
another.

Life Expectancy at Birth is the average number of years that a newborn baby is expected to live if the 
age-specific mortality rates effective at the year of birth apply throughout his or her lifetime.  The 
Population Division of the UN Secretariat prepares estimates and projections based on data from national 
statistical sources. When needed, other sources, mainly population censuses and demographic surveys, 
are consulted. In countries highly affected by the HIV/AIDS epidemic, estimates of the impact of the 
disease are made explicitly by projecting the yearly incidence of HIV infection.

Information on the methodologies used to collect these data can be found in the technical notes for each 
variable in the EarthTrends searchable database, at http://www.earthtrends.org. 


FREQUENCY OF UPDATE BY DATA PROVIDERS: 
Mortality:  Data for infant mortality are available every 10 years from 1960 to 2000, plus 1995.  Data for 
under-5 mortality exist for 1960, 1990 and 2000.  

Life Expectancy:  This data set contains estimates for every five years for all countries from 1950 to 
2050.  The UN Population Division updates this information every two years.


DATA RELIABILITY AND CAUTIONARY NOTES: 
Mortality:  The data on mortality of children after infancy is typically obtained from civil registration 
records on deaths of young children and population census information on the size of the population of 
children under one year of age.  In many developing countries, however, the registration of deaths is less 
complete than the registration of births.  Sample surveys collect birth histories to fill gaps in knowledge 
and avoid underestimating mortality.

These data are the result of a intensive effort on the part of UNICEF and its partners to develop a 
methodology producing the most consistent estimates possible for infant and under-5 mortality.  The data 
are based on a careful process of fitting curves to existing data.  Nevertheless, comparisons between 
countries should still be made with caution, because the amount, timeliness, and quality of available 
information vary widely by country.  Some countries have numerous sources of data covering the same 
period, allowing for cross-references while other estimates are derived from many fewer sources of poor 
quality.  For information on the underlying data for each country's regressions, refer to the country 
estimates and new country data available at http://www.childinfo.org/cmr/kh98meth.html.  


Life Expectancy:  The United Nations applies age- and sex-specific mortality rates to the base-year 
population to determine the number of survivors at the end of each 5-year period. Original demographic 
models are based on surveys and censuses with well-understood qualities, which make these data fairly 
reliable, although accuracy varies.  Data are adjusted for overenumeration and underenumeration of 
certain age and sex groups (e.g., infants, female children, and young males), misreporting of age and sex 
distributions, and changes in definitions, when necessary. These adjustments incorporate data from civil 
registrations, population surveys, earlier censuses, and, when necessary, population models based on 
information from socioeconomically similar countries.  Historical data are used when deemed accurate, 
also with adjustments and scaling. However, accurate historical data do not exist for many developing 
countries. In such cases, the UN Population Division uses available information and demographic models 
to estimate the main demographic parameters.

SOURCES:  
Mortality:  United Nations Childrens' Fund (UNICEF). 2003.  Progress Since the World Summit for 
Children: The ChildInfo Statistical Database.  New York: UNICEF.  Available on-line at: 
http://www.childinfo.org/cmr/revis/db1.htm.  

Life Expectancy: Population Division of the Department of Economic and Social Affairs of the United 
Nations Secretariat. 2002. World Population Prospects: The 2000 Revision. Dataset on CD-ROM. New 
York: United Nations.