Fast track — ArticlesWorldwide mortality in men and women aged 15–59 years from 1970 to 2010: a systematic analysis
Introduction
Public health efforts in the 1980s and 1990s had a substantial focus on improving mortality and morbidity in children.1, 2, 3 Both the numbers and rates of death in children under 5 years of age have been falling for several decades, although many countries are not on track to achieve Millennium Development Goal (MDG) 4, which calls for a two-thirds reduction in the mortality rate in children younger than 5 years between 1990 and 2015.1, 3, 4, 5, 6 Concomitant with decreases in under-5 mortality, global fertility has declined from a total fertility rate (TFR) of 5·0 in 1950 to 2·5 in 2009, leading to a substantial increase in the mean and median age of most populations.7 Declining under-5 mortality rates and ageing populations also mean that a larger proportion of deaths occur in adults.7, 8 Despite the increase in adult population and the related change in population health issues that follow this demographic shift, there has been much less global health focus on the health and survival of adults.
In 1992, Feachem and colleagues9 drew attention to deaths in adults aged 15–59 years. Deaths in individuals aged younger than 60 years can be considered premature by any standard.10, 11, 12 Deaths in the most economically and socially active groups can also have major effects on society.13, 14, 15 For these reasons, the World Bank and subsequently WHO have reported adult mortality risk, also referred to as 45q15.8, 16 For a given year, 45q15 represents the probability that an individual who has just turned 15 years will die before reaching the age of 60 years, on the assumption that the age-specific mortality conditions of the year are constant throughout this individual's life.
Interest in adult mortality has been intensified through the Millennium Declaration. MDG 5 on maternal health focuses on one of the important causes of death in women aged 15–49 years. Adult female mortality rates are an essential component of the measurement of the maternal mortality ratio.17 Two of the three diseases covered by MDG 6, tuberculosis and HIV, are largely killers of adults—95% and 85% of deaths from these diseases, respectively, occur in people older than 15 years.8 Although maternal mortality, HIV, and tuberculosis have received substantial policy attention and development assistance for health,18 the rising burden of non-communicable diseases in developing countries has received much less policy attention.19, 20, 21
Despite the attention on specific diseases that affect adults, policy traction towards improving overall adult health outcomes continues to be low.9 This disinterest has led to widespread neglect for building and maintaining data systems for measuring adult mortality.22, 23 Tracking change in this basic outcome of adult health is important for assessing progress, improving interventions, and driving further investment.24 However, challenges in measurement of adult mortality have been noted historically and continue to persist.2, 22, 23, 25, 26
Assessments every 2 years by the United Nations Population Division (UNPD) are undertaken to produce population estimates and projections for most countries for 5-year periods. These efforts also produce estimates of life expectancy, age-specific deaths, and 45q15, but for most developing countries, the estimates are based on models that assume a close correlation between adult and under-5 mortality.7 WHO also periodically produces estimates of 45q15; these often differ substantially from UNPD estimates. WHO does not produce a complete time series (only selected years) and, similar to UNPD, major limitations of WHO's approach are that the process is neither transparent nor replicable.27 Large-scale systematic assessments that cover all publicly available data sources over several decades have been completed for children but not for adults.1, 28, 29 In countries without complete vital registration systems, there has been disagreement about the best methods to analyse and interpret partial vital registration data, survey data for sibling histories, and survey or census recall of deaths.2 For some regions, important comparative analyses of available data have been undertaken but these efforts did not cover the entire developing and developed world.30 The absence of a systematic assessment of the evidence on trends in adult mortality, especially in developing countries, has resulted in a “scandal of ignorance”.2, 30, 31 Several developments in data availability and analytical methods now make a systematic assessment of trends in 45q15 feasible. Improved methods for analysis of incomplete vital registration or sample registration data are now available.32 Issues of survivor bias in the analysis of sibling histories widely available in sub-Saharan Africa have been addressed and practical implementation of this approach worked out.33, 34 More demographic surveillance system and sample registration data are available.35 Finally, techniques used to synthesise multiple data sources for children can be adapted for studying adult mortality (Rajaratnam J K, et al, unpublished data). In this study, we systematically analysed data for 187 countries from 1970 to 2010 to estimate 45q15 for men and women.
Section snippets
Data sources
Empirical measurements of adult mortality were drawn from four types of sources (table 1): (1) vital registration data, (2) sample registration systems (when available), and nationally representative survey or census data that enable direct estimation of age-specific adult mortality rates from questions about either (3) deaths in the household or (4) the survival of siblings of a respondent. In a few cases, an estimate of 45q15 from survey reports was used when other sources of microdata or
Results
The empirical measurements of 45q15 and the results of our data synthesis, including uncertainty intervals, are available for each country in webappendix pp 18–391; five examples of these data plots are shown in figure 1. Data for adult male mortality in Chile provide an example of a complete time series with complete vital registration data. In this case, our Gaussian process regression model produced estimates that were very close to the observed data and the uncertainty intervals represent
Discussion
Our analysis of all available empirical data for adult mortality from 1970 to 2010 for 187 countries shows that the rates of change for adult mortality are substantially more varied than are those for mortality in children under 5 years of age (Rajaratnam J K, et al, unpublished data). Risk of mortality is generally higher for men than for women; however in 2·5% of country-years, risk is higher for women than for men. There is also a widening gap between risks of male and female mortality.
References (92)
- et al.
Can we achieve Millennium Development Goal 4? New analysis of country trends and forecasts of under-5 mortality to 2015
Lancet
(2007) - et al.
Levels and trends in under 5 mortality, 1990–2008
Lancet
(2010) - et al.
Measuring the impacts of working-age adult mortality on small-scale farm households in Kenya
World Dev
(2004) - et al.
Estimates of maternal mortality worldwide between 1990 and 2005: an assessment of available data
Lancet
(2007) - et al.
Financing of global health: tracking development assistance for health from 1990 to 2007
Lancet
(2009) Comprehensive policy analysis for health system reform
Health Policy
(1995)- et al.
Interim measures for meeting needs for health sector data: births, deaths, and causes of death
Lancet
(2007) Preventing HIV/AIDS through poverty reduction: the only sustainable solution?
Lancet
(2004)- et al.
Adult mortality: time for a reappraisal
Int J Epidemiol
(2004) Countdown to 2015 for maternal, newborn, and child survival: the 2008 report on tracking coverage of interventions
Lancet
(2008)
The decline in child mortality: a reappraisal
Bull World Health Organ
Child mortality in the developing world
Popul Dev Rev
World population prospects: the 2008 revision population database
The global burden of disease 2004 update
The health of adults in the developing world
Years of potential life lost (YPLL)—what does it measure?
Epidemiology
The fall and rise of US inequities in premature mortality: 1960–2002
PLoS Med
Potential years of life lost between ages 1 and 70: an indicator of premature mortality for health planning
Int J Epidemiol
The economic effects of the AIDS epidemic in sub-Saharan Africa: a general equilibrium analysis
Rev Econ Dev
Labor effects of adult mortality in Tanzanian households
Econ Dev Cult Change
Developing countries face double burden of disease
Bull World Health Organ
Emerging non-communicable disease epidemic in Africa: preventive measures from the WHO Regional Office for Africa
Ethn Dis
Counting the dead and what they died from: an assessment of the global status of cause of death data
Bull World Health Organ
Drivers of inequality in Millennium Development Goal progress: a statistical analysis
PLoS Med
Measurement of adult mortality in less developed countries: a comparative review
Popul Index
Adult mortality: levels, patterns and causes
World health statistics 2009
The state of the world's children 2008: child survival
State of the world's children 2009
Adult health: a legitimate concern for developing countries
Am J Public Health
Should data from demographic surveillance systems be made more widely available to researchers?
PLoS Med
What can we conclude from death registration? Improved methods for evaluating completeness
PLoS Med
Death by survey: estimating adult mortality without selection bias from sibling survival data
Demography
Measuring adult mortality using sibling survival: a new analytical method and new results for 44 countries, 1974–2006
PLoS Med
INDEPTH model life tables for sub-Saharan Africa
Mortality and population databases
Demographic yearbook
Mongolian statistical yearbooks
The mortality in Antananarivo from 1984 to 1995
Deaths: preliminary data for 2007
Nat Vital Stat Rep
Sample registration system statistical report 1983–2006
Report on sample vital registration in Bangladesh 1981–1992
Report on sample vital registration system, 2005–06
Report on sample vital registration system, 2007
Cited by (140)
Improving job strain might reduce inequalities in cardiovascular disease mortality in european men
2020, Social Science and MedicineGlobal, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980–2015: a systematic analysis for the Global Burden of Disease Study 2015
2016, The LancetCitation Excerpt :GBD employs various analytical tools and a diverse set of data sources to generate comparable estimates of deaths and mortality rates broken down by age, sex, cause, year, and geography. Multiple publications show more detail on the various aspects of the methods.7,8,12,19 Part 1 of the methods appendix (pp 4–51) is a structured and succinct explanation of each step.
Death and capital
2021, Macroeconomic DynamicsDoes Democracy Matter for Lifespan Inequalities? Regime Type and Premature Mortality by Sex
2023, Studies in Comparative International Development