Global, regional, and national age-sex-specific mortality and life expectancy, 1950-2017: a systematic analysis for the Global Burden of Disease Study 2017

Abstract

Background: Assessments of age-specific mortality and life expectancy have been done by the UN Population Division, Department of Economics and Social Affairs (UNPOP), the United States Census Bureau, WHO, and as part of previous iterations of the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD). Previous iterations of the GBD used population estimates from UNPOP, which were not derived in a way that was internally consistent with the estimates of the numbers of deaths in the GBD. The present iteration of the GBD, GBD 2017, improves on previous assessments and provides timely estimates of the mortality experience of populations globally.

Methods: The GBD uses all available data to produce estimates of mortality rates between 1950 and 2017 for 23 age groups, both sexes, and 918 locations, including 195 countries and territories and subnational locations for 16 countries. Data used include vital registration systems, sample registration systems, household surveys (complete birth histories, summary birth histories, sibling histories), censuses (summary birth histories, household deaths), and Demographic Surveillance Sites. In total, this analysis used 8259 data sources. Estimates of the probability of death between birth and the age of 5 years and between ages 15 and 60 years are generated and then input into a model life table system to produce complete life tables for all locations and years. Fatal discontinuities and mortality due to HIV/AIDS are analysed separately and then incorporated into the estimation. We analyse the relationship between age-specific mortality and development status using the Socio-demographic Index, a composite measure based on fertility under the age of 25 years, education, and income. There are four main methodological improvements in GBD 2017 compared with GBD 2016: 622 additional data sources have been incorporated; new estimates of population, generated by the GBD study, are used; statistical methods used in different components of the analysis have been further standardised and improved; and the analysis has been extended backwards in time by two decades to start in 1950.

Findings: Globally, 18·7% (95% uncertainty interval 18·4-19·0) of deaths were registered in 1950 and that proportion has been steadily increasing since, with 58·8% (58·2-59·3) of all deaths being registered in 2015. At the global level, between 1950 and 2017, life expectancy increased from 48·1 years (46·5-49·6) to 70·5 years (70·1-70·8) for men and from 52·9 years (51·7-54·0) to 75·6 years (75·3-75·9) for women. Despite this overall progress, there remains substantial variation in life expectancy at birth in 2017, which ranges from 49·1 years (46·5-51·7) for men in the Central African Republic to 87·6 years (86·9-88·1) among women in Singapore. The greatest progress across age groups was for children younger than 5 years; under-5 mortality dropped from 216·0 deaths (196·3-238·1) per 1000 livebirths in 1950 to 38·9 deaths (35·6-42·83) per 1000 livebirths in 2017, with huge reductions across countries. Nevertheless, there were still 5·4 million (5·2-5·6) deaths among children younger than 5 years in the world in 2017. Progress has been less pronounced and more variable for adults, especially for adult males, who had stagnant or increasing mortality rates in several countries. The gap between male and female life expectancy between 1950 and 2017, while relatively stable at the global level, shows distinctive patterns across super-regions and has consistently been the largest in central Europe, eastern Europe, and central Asia, and smallest in south Asia. Performance was also variable across countries and time in observed mortality rates compared with those expected on the basis of development.

Interpretation: This analysis of age-sex-specific mortality shows that there are remarkably complex patterns in population mortality across countries. The findings of this study highlight global successes, such as the large decline in under-5 mortality, which reflects significant local, national, and global commitment and investment over several decades. However, they also bring attention to mortality patterns that are a cause for concern, particularly among adult men and, to a lesser extent, women, whose mortality rates have stagnated in many countries over the time period of this study, and in some cases are increasing.

Funding: Bill & Melinda Gates Foundation.

Figure 1

Estimated proportion of deaths that are registered and reported globally and by GBD super-region, for both sexes combined, 1950–2016 Each line represents the proportion of deaths that are registered and reported for a given GBD super-region or globally from 1950 to 2016. The reason for the dips in the most recent years is that lags in reporting mean that estimated deaths are higher than what is reported, resulting in a huge drop in completeness from 2015, where the reported deaths are more complete. GBD=Global Burden of Diseases, Injuries, and Risk Factors Study.

Figure 2

Total number of deaths by age, globally, for both sexes combined, 1950–2017 Each stacked bar represents the total number of deaths in the given year attributable to each age group, from 1950 to 2017, for both sexes combined. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days.

Figure 3

Natural logarithm of age-specific mortality rates, globally, for both sexes combined, 1950–2017 Each line represents the natural logarithm of the global death rate for a single year by age group, from 1950 to 2017, for both sexes combined. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days.

Figure 4

Global log (death rate) age-pattern for both sexes combined, by year, 1950–2017 Each line represents the logarithm of global age-specific mortality rates for a given year between 1950 and 2017 and for both sexes combined. The inset shows a closer view of age groups from 20 to 39 years.

Figure 5

Life expectancy at birth and by GBD super-region for both sexes combined, 1950–2017 GBD=Global Burden of Diseases, Injuries, and Risk Factors Study.

Figure 6

Life expectancy at birth for the countries with population greater than 100 million, 2017 Each line represents life expectancy at birth from 1950 to 2017 for females (A) and males (B).

Figure 7

Life expectancy at birth, by location, for females (A) and males (B), 2017, and difference in life expectancy at birth, by location, for females (C) and males (D) between 2017 and 1950 ATG=Antigua and Barbuda. FSM=Federated States of Micronesia. LCA=Saint Lucia. TLS=Timor-Leste. TTO=Trinidad and Tobago. VCT=Saint Vincent and the Grenadines.

Figure 7

Life expectancy at birth, by location, for females (A) and males (B), 2017, and difference in life expectancy at birth, by location, for females (C) and males (D) between 2017 and 1950 ATG=Antigua and Barbuda. FSM=Federated States of Micronesia. LCA=Saint Lucia. TLS=Timor-Leste. TTO=Trinidad and Tobago. VCT=Saint Vincent and the Grenadines.

Figure 8

Difference in life expectancy at birth between females and males for SDI (A) and GBD super-region (B), 1950–2017 Each line represents the difference in life expectancy between females and males (female life expectancy minus male life expectancy) for a given SDI level in 2017 (A) and GBD super-region (B), for each year between 1950 and 2017. SDI=Socio-demographic Index. GBD=Global Burden of Diseases, Injuries, and Risk Factors Study.

Figure 9

Median observed to expected ratio of age-specific mortality rate, globally, for both sexes combined, in the 1950s, 1960s, 1970s, 1980s, 1990s, 2000s, 2010s, and 2017 Each line represents the median ratio across locations for each decade of the observed age-specific mortality rate to that expected on the basis of SDI, for each of the 23 GBD age groups. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. GBD=Global Burden of Diseases, Injuries, and Risk Factors Study.

Figure 10

Age-specific, all-cause mortality rate with the ratio of observed to expected by SDI, 2017 Each square represents the ratio of the observed all-cause mortality rate to that expected on the basis of SDI for females (A) and males (B) for that given location, sex, and age group. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. *Round brackets indicate excluded endpoints whereas square brackets indicate included endpoints.

Figure 10

Age-specific, all-cause mortality rate with the ratio of observed to expected by SDI, 2017 Each square represents the ratio of the observed all-cause mortality rate to that expected on the basis of SDI for females (A) and males (B) for that given location, sex, and age group. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. *Round brackets indicate excluded endpoints whereas square brackets indicate included endpoints.

Figure 10

Age-specific, all-cause mortality rate with the ratio of observed to expected by SDI, 2017 Each square represents the ratio of the observed all-cause mortality rate to that expected on the basis of SDI for females (A) and males (B) for that given location, sex, and age group. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. *Round brackets indicate excluded endpoints whereas square brackets indicate included endpoints.

Figure 10

Age-specific, all-cause mortality rate with the ratio of observed to expected by SDI, 2017 Each square represents the ratio of the observed all-cause mortality rate to that expected on the basis of SDI for females (A) and males (B) for that given location, sex, and age group. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. *Round brackets indicate excluded endpoints whereas square brackets indicate included endpoints.

Figure 10

Age-specific, all-cause mortality rate with the ratio of observed to expected by SDI, 2017 Each square represents the ratio of the observed all-cause mortality rate to that expected on the basis of SDI for females (A) and males (B) for that given location, sex, and age group. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. *Round brackets indicate excluded endpoints whereas square brackets indicate included endpoints.

Figure 10

Age-specific, all-cause mortality rate with the ratio of observed to expected by SDI, 2017 Each square represents the ratio of the observed all-cause mortality rate to that expected on the basis of SDI for females (A) and males (B) for that given location, sex, and age group. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. *Round brackets indicate excluded endpoints whereas square brackets indicate included endpoints.

Figure 10

Age-specific, all-cause mortality rate with the ratio of observed to expected by SDI, 2017 Each square represents the ratio of the observed all-cause mortality rate to that expected on the basis of SDI for females (A) and males (B) for that given location, sex, and age group. The early neonatal age group is 0–6 days, late neonatal is 7–27 days, and postneonatal 28–364 days. SDI=Socio-demographic Index. *Round brackets indicate excluded endpoints whereas square brackets indicate included endpoints.