Life Expectancy Monitor

World Map

Difference in Life Expectancy at various ages

Cause of Death Distribution

Cause of Death / Age Decomposition of the Change in Life Expectancy at Birth

How we estimate demographic indices

Life expectancy

From observed age-specific death rates at a given time, a life table is constructed with life expectancy as one of its elements. Life expectancy, or the average number of years lived by a population, is an easily interpretable metric, clinically meaningful, and simple to understand. Life expectancy can be calculated at birth and at other ages, in all cases representing an age-aggregated measure of mortality in a given time. Life expectancies are particularly useful for comparisons across time and populations, since they are not affected by population size or age-structure. (Preston et al. 2001)

Cause reduced

Associated single decrement life tables are used for the hypothetical scenario where mortality from one cause of death is deleted. A similar procedure can be applied for life tables in which one or various causes of death are reduced partially at the same time. Those new life tables will have higher life expectancies than the original, which allows to quantify the impact of reducing specific causes of death on increasing life expectancy (Chiang et al. 1968).

Decomposing the change in life expectancy over time

To better understand differences in life expectancy of the original population vs its cause-reduced version, we decomposed those differences into age- and cause-specific contributions. We used the stepwise decomposition to disentangle the age and cause of death contribution to comparisons of life expectancy across time or populations. This methodology assumes that causes of death are exhaustive, not allowing mortality to be duplicated, and independent. This independence refers to the situation where the removal of one cause leaves the risk of dying from all other causes unchanged (Andreev et al 2002).


  • Andreev EM, Shkolnikov VM, and Begun AZ. Algorithm for decomposition of differences between aggregate demographic measures and its application to life expectancies, healthy life expectancies, parity-progression ratios and total fertility rates. Demographic Research 7: 499-522. 2002.
  • Chiang CL. Introduction to Stochastic Processes in Biostatistics. Wiley Publishers, New York, 1968.
  • Preston SH, Heuveline P, and Guillot M. Demography: measuring and modeling population processes. Blackwell Publishers, Oxford, 2001.

Data sources

The Life Expectancy Monitoring Tool uses data from public databases available at the Institute for Health Metrics and Evaluation (IHME 2021) and the United Nations World Population Prospects (United Nations 2019).

The age- and cause-specific number of deaths as well as life tables were obtained from the IHME, which estimates the death counts attributable to each cause of death (this is our base source for all the calculations on cause reduce mortality). Similarly, the United Nations World Population Prospects involves statistical estimations using survey and census to account for the paucity of registration data in those areas (aggregate values of population counts, total fertility rate, population growth and life expectancies were obtained from this source and which appear in the maps).

Our results based on IHME and UN data are estimates before Covid-19, so survival is very likely to be overestimated.

Despite the IHME and UN data limitations, these source provide consistent global comparability over a long time to uncover the past and current trends in many low-income countries.


This website and its contents herein, including all data, mapping, charts, and analysis, are provided strictly for educational reasons and research purposes.

Cause of death changes and their effect on life expectancy

The life expectancy monitoring tool allows the user to selected mortality changes over the entire lifespan or at specific ages, as well as for overall mortality or for specific causes of death. For example, how would life expectancy look if cardiovascular mortality were to be reduced by 50%? Or how would life expectancy look if infant mortality was eliminated? The tool facilitates assessing changes and comparisons in life expectancy under those selected scenarios of mortality change. Furthermore, the tool lets the user compare cause-of-death profiles and life expectancies across time, countries and sexes.

Life expectancy changes in achieving SDG3

The sustainable development goals (SDG) were set in 2015 with many specific targets to be achieved by 2030. The third SDG refers to “Ensure healthy lives and promote well-being for all at all ages” []. The targets of this SDG goal refer to several actions of either eliminating or reducing mortality from certain diseases that are amenable to health interventions. Hence, the possible effects on life expectancy of achieving those mortality reductions and eliminations are highly relevant. The life expectancy monitor tool allows analysts to evaluate countries’ progress in achieving the SDGs as well as assessing possible specific scenarios or targets that a population wishes to achieve.

Updates and news

Monitor Version: 0.11.0

Last Update: “2022-03-17 18:16:52 CET”

News in previous versions:

  • 2022-03-17 - v0.11.0 - Add new functionality (boomarks, reset) and provide a more consistet arrangement of the cod with the ICD classification;
  • 2022-02-23 - v0.10.0 - Create datatab corresponding to dashboard figures (lifetables, cod distributions and decomposition values);
  • 2022-02-10 - v0.9.0 - Update database by adding macro-regions;
  • 2021-12-09 - v0.8.0 - Change the name of the R library from {MortalityCauses} to {lemur} and add dashboard documentation.
  • 2021-06-03 - v0.4.0 - Important advance in app functionality and figure coordination.
  • 2021-05-12 - v0.3.0 - Start dashboard development.
  • 2021-04-08 - v0.2.0 - Add the life expectancy decomposition method.
  • 2021-03-29 - v0.1.0 - Add the modified life table method.
  • 2021-03-01 - v0.0.1 - Start of the project implementation.

The source code and the development repository can be found on Github @mpascariu/lemur under the GNU GPLv3 license.


Development team and maintainers

Marius D. PASCARIU PhD Biometric Risks Modeling Chapter, SCOR Global Life

Prof. Jose Manuel ABURTO University of Oxford

Doug Leasure PhD University of Oxford

Prof. Vladimir CANUDAS-ROMO Australian National University (ANU)