Skip to main content

01.12.2018 | Research article | Ausgabe 1/2018 Open Access

BMC Public Health 1/2018

The slowing pace of life expectancy gains since 1950

BMC Public Health > Ausgabe 1/2018
Carolina Cardona, David Bishai
Wichtige Hinweise

Electronic supplementary material

The online version of this article (https://​doi.​org/​10.​1186/​s12889-018-5058-9) contains supplementary material, which is available to authorized users.



New technological breakthroughs in biomedicine should have made it easier for countries to improve life expectancy at birth (LEB). This paper measures the pace of improvement in the decadal gains of LEB, for the last 60-years adjusting for each country’s starting point of LEB.


LEB increases over the next 10-years for 139 countries between 1950 and 2009 were regressed on LEB, GDP, total fertility rate, population density, CO2 emissions, and HIV prevalence using country-specific fixed effects and time-dummies. Analysis grouped countries into one-of-four strata: LEB < 51, 51 ≤ LEB < 61, 61 ≤ LEB < 71, and LEB ≥ 71.


The rate of increase of LEB has fallen consistently since 1950 across all strata. Results hold in unadjusted analysis and in the regression-adjusted analysis. LEB decadal gains fell from 4.80 (IQR: 2.98–6.20) years in the 1950s to 2.39 (IQR:1.80–2.80) years in the 2000s for the healthiest countries (LEB ≥ 71). For countries with the lowest LEB (LEB < 51), decadal gains fell from 7.38 (IQR:4.83–9.25) years in the 1950s to negative 6.82 (IQR: -12.95--1.05) years in the 2000s. Multivariate analysis controlling for HIV prevalence, GDP, and other covariates shows a negative effect of time on LEB decadal gains among all strata.


Contrary to the expectation that advances in health technology and spending would hasten improvements in LEB, we found that the pace-of-growth of LEB has slowed around the world.
Additional file 1 Table A-1-1. LEB sources, by country and year. Table A-2-1. List of countries included in the analysis, when HIV is not included as a control variable. Table A-2-2. List of countries included in the analysis, when HIV is included as a control variable. Table A-2-3. List of countries by strata using a representative year for each decade, during the analysis countries were re-stratified to each LEB stratum every single year. Table A-2-4. Summary statistics, Lowest Stratum (LEB < 51). Table A-2-5. Summary statistics, Stratum II (51 ≤ LEB < 61). Table A-2-6. Summary statistics, Stratum III (61 ≤ LEB < 71). Table A-2-7. Summary statistics, Highest Stratum (LEB ≥ 71). Table A-2-8. F-tests of equality between decadal dummies parameters, 1960–69 against 2000–09 and 1980–89 against 2000–09. Table A-2-9. The effects of LEB, income per capita, fertility, population density, CO2 emissions, and time on LEB decadal gains, 1950–2009, Fixed effects model. Table A-2-10. LEB decade gains, by region, decade and LEB strata. Table A-4-1. Countries with HIV imputed values, FE regression (imputation A) and constant prevalence (imputation B). Figure A-3-1. LEB decadal gains by decade, Stratum II (51 ≤ LEB < 61) and Stratum III (61 ≤ LEB < 71). Figure A-3-2. LEB decadal gains distribution by strata, comparison between decades (PDF 1890 kb)
Über diesen Artikel

Weitere Artikel der Ausgabe 1/2018

BMC Public Health 1/2018 Zur Ausgabe