Global Burden of Disease (GBD)
The Global Burden of Disease (GBD) study provides a comprehensive picture of mortality and disability across countries, time, age, and sex. It quantifies health loss from hundreds of diseases, injuries, and risk factors, so that health systems can be improved and disparities eliminated.
What is the GBD and why is it important?
The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) is the single largest and most detailed scientific effort ever conducted to quantify levels and trends in health. Led by the Institute for Health Metrics and Evaluation (IHME) at the University of Washington, it is truly a global effort, with over 9,000 researchers from more than 160 countries and territories participating in the most recent update.
GBD creates a unique platform to compare the magnitude of diseases, injuries, and risk factors across age groups, sexes, countries, regions, and time. For decision-makers, health sector leaders, researchers, and informed citizens, the GBD approach provides an opportunity to compare their countries’ health progress to that of other countries, and to understand the leading causes of health loss that could potentially be avoided, like high blood pressure, smoking, and household air pollution.
Who works on the GBD study?
The GBD study is one of the world’s largest scientific collaborations and involves thousands of researchers around the world. It includes a Scientific Council, a Management Team, a Core Analytic Team, and a robust network of global Collaborators working together to produce the most accurate, up‐to‐date, and comparable estimates of burden worldwide.
How is the GBD study conducted?
The GBD produces regular estimates of all‐cause mortality, deaths by cause, years of life lost due to premature mortality (YLLs), years lived with disability (YLDs), and disability‐adjusted life years (DALYs).
The critical milestones for ongoing estimation include regular updates to the GBD estimates, referred to as the “GBD cycle.” For each cycle, the entire time series back to 1990 is re‐estimated using all available data to ensure the most complete and highly comparable set of estimates possible. Previous results will be archived every time new results are released.
The GBD protocol covers the key principles and assumptions, products, roles and responsibilities, processes, and architecture of the GBD study.
What methods and data inputs were used?
We start by gathering health data from hospitals, governments, surveys, and other databases around the world. Our research teams then clean and sort the data and use modeling tools to generate estimates for locations and years where data are not available. Those results are released as part of each GBD cycle and made available through our results tools, data visuals, and publications.
Transparency is one of our core principles, and we are working to help expand access to the raw data and methodology that we use in our evaluations.
More details on our methods and specific processes for each area of study can be found within our peer-reviewed publications.
Where can I get help using GBD tools?
Our data visuals may need significant time and internet bandwidth to load. If you are having trouble loading a visualization, we suggest that you refresh the page. We also suggest that you access the visualizations with the most up-to-date version of your internet browser (Chrome, Firefox, or Safari).
We also offer an online training, which provides a general overview of the GBD to help learners understand the conceptual framework, the key metrics, and the analytical strategies used in the study.
How can I cite GBD?
To cite GBD 2019 – the most recent round of Global Burden of Disease results – use the following citation:
Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2019 (GBD 2019). Seattle, United States: Institute for Health Metrics and Evaluation (IHME), 2020.
I still have questions, who do I contact?
To contact the GBD Management Team, please email [email protected].
For other general questions about our GBD work, please email [email protected].
GBD concepts and terms defined
The more than 350 causes of health loss studied within the GBD are arranged in hierarchical nested categories – referred to as “levels” – within the GBD “cause lists.” At the highest level, causes are split into very large categories: communicable, maternal, neonatal, and nutritional causes; non-communicable diseases; and injuries. Within each of those categories, causes of health loss are broken down with increasing specificity at each level. For example, consider typhoid fever, which is a Level 4 cause in the GBD cause list:
Level 1: Communicable, maternal, neonatal, and nutritional diseases
Level 2: Enteric infections
Level 3: Typhoid and paratyphoid
Level 4: Typhoid fever
The cause list is mutually exclusive and collectively exhaustive at every level of aggregation; causes not individually specified are captured in residual categories, such as “other intestinal infectious diseases.”
For a full list of causes (diseases and injuries) from the GBD study, visit the GBD data and tools guide.
The cause list is agreed upon annually by the Scientific Council.
DALY is an abbreviation for disability-adjusted life year. It is a universal metric that allows researchers and policymakers to compare very different populations and health conditions across time. DALYs equal the sum of years of life lost (YLLs) and years lived with disability (YLDs). One DALY equals one lost year of healthy life. DALYs allow us to estimate the total number of years lost due to specific causes and risk factors at the country, regional, and global levels.
Disability weights represent the severity of a disease and range from 0 (perfect health) to 1 (equivalent to death).
We assigned disability weights to each health condition in the GBD based on surveys that used pair-wise comparison methods in which respondents were asked to indicate which of two health states briefly described to them they considered to be “healthier.” A total of four surveys were conducted in low- and middle-income countries, four internet surveys conducted in multiple European countries, one telephone survey in the USA, and one open-access online survey conducted globally to provide pair-wise comparisons. DWs were then derived from these comparisons and given numerical values.
GBD regions are based on two criteria: epidemiological similarity and geographic closeness. To see a full list of countries and their corresponding regions, visit the GBD data and tools guide.
For some types of analysis in the GBD, seven super-regions have been established, which group regions on the basis of cause of death patterns. The image below shows which regions belong to each super-region.
HALE, or healthy life expectancy, is the number of years that a person at a given age can expect to live in good health, taking into account mortality and disability. While life expectancy summarizes a population’s mortality experience, HALE uses the same concept, but adjusts years lived at each age by the probability of health loss. The average amount of health loss a person experiences rises with age. For people at older ages, remaining years are therefore more affected by disability, so HALE adjusts downwards those remaining years of life more than for younger people.
Why are age-specific neonatal mortality rates often greater than 1.0?
Mortality rates can be greater than 1.0 for early neonatal (ENN or 0-6 days), late neonatal (LNN or 7-27 days), and post neonatal (PNN or 28-364 days) age groups because there is the opportunity for more deaths to occur within an age group than the population within them.
For example, in the ENN age group, each child spends 7 days out of the full 365 days of the year there. The population of the ENN age group is the person-years spent in the age group, so roughly 365 / 7 = 52 people would need to progress smoothly through the ENN group to contribute one person-year towards the ENN population. Even more people would be required if the death rate in the ENN age group was particularly high, and many children only contributed one or two days to the ENN age group before dying.
In summary, at least 52 people must pass through the ENN age group in order to generate one person-year of time for the ENN population estimate. Following this, if there were more than one death out of the 52 who passed into the ENN age group, then more than one death per person-year of population would be recorded, leading to a mortality rate greater than 1.0.
The share of deaths or DALYs that can be attributed to – i.e., estimated to occur due to – exposure to a particular risk factor (e.g., alcohol-attributable deaths, or deaths attributable to air pollution, etc.)
Sequelae are consequences of diseases and injuries. Health states are groupings of sequelae that reflect key differences in symptoms and functioning.
SDI is an abbreviation for Socio-demographic Index, a summary measure that identifies where countries or other geographic areas sit on the spectrum of development.
It is expressed on a scale of 0 to 100, with 0 being the lowest SDI value and 100 being the highest. SDI was constructed based on three measures: i) lag-distributed income per capita; ii) average years of schooling in ages 15 and older; and iii) total fertility rate (TFR) for females under age 25.
For example, a low SDI value will be assigned to a country with lower income per capita, fewer average years of schooling, and higher TFR relative to other countries; conversely, a higher SDI value will be assigned to a country with higher income per capita, greater average years of schooling, and lower TFR relative to other countries.
TFR is an abbreviation for total fertility rate, a summary measure representing the average number of children a woman would deliver over her lifetime.
Uncertainty intervals are a range of values that are likely to include the correct estimate of health loss for a given cause. Limited data create substantial uncertainty.
Years of life lost (YLLs) are years lost due to premature mortality. Invented by researcher Mary Dempsey, YLLs are calculated by subtracting the age at death from the longest possible life expectancy for a person at that age. For example, if the longest life expectancy for men in a given country is 75, but a man dies of cancer at 65, this would be 10 years of life lost due to cancer.
YLD is an abbreviation for years lived with disability, which can also be described as years lived in less than ideal health. This includes conditions such as influenza, which may last for only a few days, or epilepsy, which can last a lifetime. It is measured by taking the prevalence of the condition multiplied by the disability weight for that condition. Disability weights reflect the severity of different conditions and are developed through surveys of the general public.