2.3.5 Changes in Overall, Infant, Child, and Maternal Mortality

Facts/messages: In most countries of the world, there has been a major decrease in mortality, which represents significant progress in regard to human wellbeing and an impact on population growth. There are still important challenges with respect to infant and child mortality, as indicated in the global and regional MDG reports. The recent estimation exercises by both the Health 4 (WHO, UNICEF, UNFPA, World Bank) and the Institute for Health Metrics and Evaluation (IHME) show that the global maternal mortality ratio has fallen by about a third since 1990, although this is not enough to reach MDG 5.A, the reduction of the maternal mortality ratio by three quarters between 1990 and 2015.

Methodology: Use indicators of national life tables or based on population projections as well as indicators of infant and child mortality in the censuses, and estimates of maternal mortality according to different sources (surveys, administrative registers, indirect estimates). Avoid the use of crude death rates, unless it is to demonstrate the impact of the reduction of mortality on demographic growth. It is also useful to make comparisons with other countries in the region. The demographic and social factors influencing infant, child and maternal mortality should be highlighted. In some countries it may be worthwhile to look more deeply into mortality associated with violence and accidents, for example in South Africa and Papua New Guinea. There may also be interest in addressing femicide as an indicator of violence against women. Use indicators of morbidity (if available) and death rates due to accidents and violence. When using the maternal mortality estimates prepared by the UN or IHME, it is important to be aware of the methodology underlying them. For instance, while the levels of maternal mortality estimated with the UN methodology in countries that do not have reliable vital registration depend directly on the number and the level of national estimates available, the trends depend primarily on the evolution of the explanatory variables and may well contradict trends based on national data.

The SPECTRUM software package, developed by the Future Institute, contains a maternal mortality module, although the focus of this module is not on the social and economic consequences of maternal mortality, but rather on prevention strategies. It estimates the impact of various scores from the Maternal-Neonatal Program Index (MNPI), which is an index of 81 indicators for national efforts to improve maternal and neonatal health services, on a country’s maternal mortality ratio. The model helps managers to gain a better understanding of the impacts of policies, budgets, and service delivery improvements on maternal health outcomes.

Primary Sources:

  • Vital statistics (where these are complete or satisfactorily corrected);
  • Administrative registers;
  • DHS and MICS surveys;
  • National Population Projections. For corrected and projected mortality data, one may want to refer to the life tables underlying national population projections.

Secondary Sources:

  • The UN Statistical Yearbook provides most of these country data, with an assessment of their reliability;
  • The Human Mortality Data Base (http://www.mortality.org) is a collaborative project of the University of California at Berkeley and the Max Planck Institute in Rostock, Germany, with detailed mortality data for 37 developed countries;
  • UN Population Division and World Health Organization (WHO). Life tables for all countries of the world, to support their population projections;
  • For the maternal mortality methodology, refer to the IHME article in The Lancet (April, 2010), the official UN publication Trends in Maternal Mortality: 1990 to 2008 (Sept., 2010) and the technical note entitled Questions and answers on the estimation of maternal mortality: an updated technical note. The international estimates of both maternal and child mortality are controversial, so it is recommended to take a close look at all sides of the argument before adopting any of them;
  • WHO: Regional databases on health status and coverage indicators.


  • African Development Bank and UNFPA (2005). Training Module on Integration of Population Issues into African Development Bank programmes and projects. Module 2, Session 1 on Levels and trends in Population Size, Fertility, Mortality and Migration;
  • SPECTRUM: http://www.futuresinstitute.org/Pages/Spectrum.aspx;
  • Lamlenn B. Samson (2008). Guidance note for the in-depth analysis of data from a Population and Housing Census. Dakar, CST: section on Analysis of Mortality Levels and Differentials.

43  UNFPA (2003). Needs Assessment Report. Chapter on Methodology. Page 9.
44  UNFPA (2005): Meeting Report. Chapter III.c. Measurement: Development of Indicators.
45  http://www.measuredhs.com/aboutsurveys/search/listmodules_main.cfm.