GBD 2023: New findings on global mortality and life expectancy

Despite population growth and aging, the 2023 global age-standardized mortality rate declined 67% since 1950, and all countries and territories marked declines. Global life expectancy returned to pre-pandemic levels at 76.3 years for females and 71.5 years for males, which is more than 20 years higher compared to 1950.

The Global Burden of Disease 2023 study was officially released on October 12, 2025 alongside three capstones published in The Lancet: new findings on global demographics, causes of death, and the burden of diseases, injuries, and risk factors.

IHME Director Dr. Chris Murray covers the demographics capstone—a major analysis using new data sources and methods to understand global mortality and life expectancy from 1990 to 2023.

Video transcript

This transcript has been lightly edited for clarity

In the Global Burden of Disease 2023 analysis of demographic data, we have introduced new data sources, as we do each cycle of the GBD, but we’ve also profoundly advanced the methods that we use to analyze demographic data, particularly for age-specific mortality. And this matters particularly in low-resource settings where we have historically, and still currently, depended heavily on survey sources, complete birth histories for child mortality, and then sibling histories that are deployed in household surveys, particularly in sub-Saharan Africa, but elsewhere as well. 

And what we’ve done now, with new methods developed by the math sciences team at IHME, is analyzed those sources, particularly the sibling histories, age group by age group, each five-year age group separately, to provide us with some new insights into what’s happening in those parts of the world.

And then, of course, as the vital statistics have come in after the COVID pandemic, we’ve started to get insights into what occurred during the pandemic, and then in the recovery phase. So that’s the changes in data and methods, and what have we observed? Well, if we zoom out, look at the long-term view, life expectancy has been steadily increasing, with some notable exceptions: the HIV epidemic, the rise of mortality in Eastern Europe in the late 1980s, 1990s and, of course, the COVID pandemic.

But we shouldn’t forget that we’ve seen a 70–75-year period of extraordinary progress in human health. If we zoom in more closely into the last five years, we had the big drop in life expectancy in most regions, but not Southeast Asia or East Asia, in 2020, and then even larger drops in 2021 in life expectancy in some parts of the world as new variants came along, particularly the Delta variant of COVID, that led to big increases in mortality. And then as the Omicron variant and subsequent families, or variants in that family, we’ve seen COVID mortality drop very noticeably. Net effect is we have the dip in life expectancy in 2020, 2021, and most places have recovered since then – although there are many countries in the world that are not yet back on the trajectory that life expectancy was on before the COVID pandemic. So, recovery but not quite back to where we should have been if COVID had not come along.

A second big observation in our demographic analysis is around changes in some parts of the world, particularly some high-income countries, in the rise of mortality in children and adolescents related to suicide and some other causes.

And then the big rise in drug-related deaths in the last 10 or 15 years in the United States, in a number of countries in Europe, Mexico, Canada, and others. And so we’ve seen, in those younger adult age groups, a divergence from the trends in some parts of the world. So it’s not simply one of steady progress, but there are these sort of targeted groups where mortality is rising. And of course, given these are often in younger adults, these are of great concern in the countries where that’s occurring. 

Our numbers – not the trend – but our numbers in sub-Saharan Africa, because of the improved methods, the age-specific assessment in younger women and younger men, are higher than they were in the past and, as a consequence, our numbers at older ages are somewhat lower because, previously, we were analyzing sibling history data from the whole age group from 15 to 60 as a block.

And the more detailed age-specific focus means that we are starting to understand some age groups have higher mortality and others lower than we previously thought. Now, the net effect of this will be seen in our analysis of cause-specific mortality, but, in general, I’d say that demographic assessments suggest we’re returning back to the trajectories that we wished we were on, but not quite there. And we saw very marked variation of the impact of COVID. And we need to tackle some of the new drivers of younger mortality in select regions.