A comprehensive global analysis of adult mortality in men and women aged 15--59 years shows that Australia (both sexes) and South Korea (women) haves some of the highest annual rates of decline. The lowest risk of death in this age group was recorded in Iceland (men) and Cyprus (women). But due to the HIV epidemic (Africa) and social dysfunction after major upheaval (former Soviet state) and other factors, mortality rates are higher in many countries in 2010 than they were 40 years ago. The findings are reported in an Article published Online First and in an upcoming Lancet, written by Dr Christopher Murray, Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA, USA, and colleagues.
Although much attention is given to mortality in children under-5, adult mortality is a crucial priority for global health, but receives much less attention. The authors say: "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'." In this study, the authors created estimates for the probability that an individual who has just turned 15 will die before they reach age 60 (termed 45q15). They used four sources to calculate mortality rates across 187 countries: vital registration data, sample registration systems and survey/census data, deaths in household, and sibling survival data.
Over the 40 years, overall global 45q15 fell by 34% in women and 19% in men. Risk of mortality was almost always higher in men than women (up to 2.2 times higher). Yearly rates of decline from 1970 to 1990 were 0•7% for men and 1•4% for women. From 1990, worldwide male 45q15 increased until 1995 and then began to decline. From 1990, female 45q15 declined at a much faster rate than did male 45q15, apart from during the period between 1994 and 2003, when both were declining roughly at roughly the same pace. Worldwide, the gap between adult male and female 45q15 was 63 per 1000 in 1970 and increased to 80 per 1000 in 2010. Worldwide, the 1990s reversal in the trend in adult mortality is probably a result of the HIV pandemic and the sharp rise in adult mortality in countries of the former Soviet Union. This put the high rates in sub-Saharan Africa as a whole (578 per 1000 men, 446 per 1000 women) into context: such rates have not been seen in a high-income country such as Sweden since the year 1751.
Every region has its story to tell. One of the most striking is the rapid decline in adult mortality for women in South Asia; in 1970 this was the region with the highest levels of female mortality and, by 2010, it had declined by 56%. Male mortality declined more across Australasia than other regions, while in North America the decline was slower than elsewhere. Across southeast Asia, rates of decline since the mid-1990s have stabilised for both sexes. While Africa has high rates for both sexes, the whole of sub-Saharan Africa has seen mortality decline since 2005, probably due to reduced HIV prevalence and increased access to antiretroviral treatment.
The 'top-ten' countries (with lowest mortality) for both men and women had some surprises for both 1970 and 2010, although there was more consistency in the women's league table. Australia moved from 44th place for male mortality in 1970 to 6th in 2010; while Paraguay, surprisingly at 5th in 1970, had dropped to 70th 40 years later. Other surprises were Cuba, at 3rd in 1970, and Costa Rica, at 10th--but both had also slid well out of the top ten to 36th and 30th respectively, by 2010. Greece had the lowest male mortality worldwide in 1970, yet by 2010 it was only 22nd in the league. Only Sweden, the Netherlands and Norway kept their top ten places in the male league over the 40 years. The top five countries today are, in order, Iceland, Sweden, Malta, the Netherlands, and Switzerland. Italy, Qatar, Australia, and Israel also moved into the top ten for 2010.
For women, again Australia was a star performer, rising from 36th to 8th, but it was eclipsed by South Korea, which rose from 123rd in 1970 to 2nd in 2010. Greece, Sweden, Switzerland, Spain, Italy, Iceland, and Cyprus all kept their top ten places over the 40 years. Today, the top five countries are, in order, Cyprus, South Korea, Japan, Greece, and Italy.
The story for the United Kingdom is mixed. Mortality for both sexes has almost halved between 1970 and 2010. However, today British adult women have a similar risk of death to those in Slovenia and Albania. In Western Europe, only Danish and Belgian women had higher risk of death than British women. British men fare much better than those from all countries of Eastern Europe and are also mid-ranking in Western Europe. The highest risk of death for adult males in Western Europe was recorded in Finland, followed closely by Portugal, France, and Denmark.
Among high-income countries, the USA is notable for having a rate of decline of less than 1•5% per year for women, as are Norway (women) and Spain (men). Only a handful of countries have managed to decrease both male and female mortality by more than 2•0% per year from 1970 to 2010. These include countries across the scale of income: high (Australia, Italy, South Korea); middle (Chile) and low (Tunisia, Algeria).
The authors say that five factors might explain the changes in adult mortality: the diseases of affluence; socioeconomic development; improved health technologies; social dysfunction in the aftermath of the collapse of the Soviet Union; and the HIV epidemic. But they say: "The decline in adult morality is manifesting itself differently in each region and could be attributed to one or a combination of the following factors: increase in income, improvement in education, adoption of modern health technology, and access to health care."
They add: "Well documented periods of rapid decline, such as in South Korea between 1990 and 2010, when adult mortality declined by over 50%, or between 1995 and 2010 in Estonia, when there was greater than 35% decline, show that rapid progress is possible. These bursts of accelerated decline in mortality suggest that policies adopted in one setting could be transferred to other settings."
These latest adult mortality estimates are substantially different from those reported by the UNDP because IHME examined a larger dataset and applied new statistical methods to the data. In addition, the UNDP has often based its adult mortality estimates on child mortality estimates using model life tables that rely on data mainly gathered before 1970.
The authors say that the strongest step a country can take to improving the surveillance of adult female and male mortality is to strengthen national vital registration systems. They say: "No amount of post-data collection analysis will be as effective as implementation of a complete system."
They conclude that around three times as many adults aged 15-60 die per year (around 24 million) as do children under-5. They say: "The prevention of premature adult death is just as important for global health policy as is the improvement of child survival. The global health community needs to broaden its focus and to learn from measures applied in countries such as Australia and South Korea to ensure that those who survive to adulthood will also survive until old age. This refocus will require much greater efforts to equip all countries with reliable and timely mortality surveillance systems, preferably complete vital registration systems, to guide prevention policies and programmes."
In an accompanying Comment, Dr Ai Koyanagi and Dr Kenji Shibuya, Department of Global Health Policy, Graduate School of Medicine, University of Tokyo, Japan, back the study authors' call for more complete registration systems. They add: "Because much of the variation in adult mortality cannot be explained by the combination of economic development, the HIV epidemic, and child mortality, the new analysis challenges the common theories behind health transition, which will stimulate debates on alternative theories and the roles of social determinants, health systems, and medical technologies."
For Dr Christopher Murray, please contact either William Heisel or Jill Oviatt, IHME Communications. William Heisel T) +1 (206) 897-2886 / + 1 (206) 612-0739 E) firstname.lastname@example.org
Jill Oviatt T) +1 (206) 897-2862 /+1 (206) 861-6684 E) email@example.com
Dr Kenji Shibuya, Department of Global Health Policy, Graduate School of Medicine, University of Tokyo, Japan. T) +81 80 2055 0440 E) firstname.lastname@example.org
For full Article and Comment, see: http://press.
See also attached spreadsheet for league tables for all the countries for men and women in 1970, 1990, and 2010.