To increase accessibility and transparency of GBD 2019 estimates, we have prepared summaries for each cause and nine impairments. A selection of these appear in the main text of the research article, and all are available online. GBD classifies causes in a hierarchy of four levels. Causes at all levels are presented in the summaries.
Level 1 causes are aggregates of non-communicable diseases; injuries; and a category combining infectious diseases, maternal and neonatal disorders, and nutritional deficiencies.
At Level 2, there are 22 disease and injury aggregate groupings such as respiratory infections and tuberculosis, cardiovascular diseases, and transport injuries.
Level 3 includes specific causes such as tuberculosis, stroke, and road injuries. In some cases, these Level 3 causes are the most detailed classification, while for others a more detailed category is specified at Level 4. Examples of Level 4 causes include latent tuberculosis infection, ischemic stroke, and pedestrian road injuries.
GBD also makes estimates for impairments. These are conditions that occur as a consequence of a range of causes in the GBD hierarchy, and for which there tends to be more data on the total occurrence of the impairment rather than the cause-specific estimates. We then constrain the cause-by-cause estimates to sum to the total of the impairment.
Four custom aggregate causes produced for GBD 2019 are also included as summaries: total cancers, total burden related to non-alcoholic fatty liver disease (NAFLD), total burden related to hepatitis B, and total burden related to hepatitis C. Total cancers is an aggregate of Level 3 causes, while the other custom estimates are aggregates of Level 4 causes. Total cancers is presented in place of the standard “neoplasms” aggregate cause that incorporates both malignant and benign neoplasms.
Each two-page summary begins with a summary results statement; a definition of the disease, injury, or impairment; and a brief listing of modeling updates in GBD 2019 compared to previous GBD rounds.
Table 1 displays the total number of sources used in the GBD 2019 estimate for the disease, injury, or impairment in five categories: incidence, prevalence, remission, causes of death, and other.
Table 2 provides global numbers of prevalence, incidence, deaths, years of life lost (YLLs, estimated as death counts multiplied by a remaining life expectancy from a standard life table), years lived with disability (YLDs, estimated as prevalent cases multiplied by a disability weight reflecting the severity of each disease outcome), and disability-adjusted life years (DALYs, the sum of YLLs and YLDs), for the cause or impairment for both sexes combined, males, and females in 2019. Age-standardized rates (age-specific rates adjusted to the GBD global standard population) are also provided, along with percentage change for each metric from 2010 to 2019. Dashes represent metrics for which estimates were not produced.
Table 3 provides rankings of deaths, YLLs, YLDs, and DALYs for the disease, injury, or impairment for the years 1990, 2010, and 2019. The total cancers custom aggregate was ranked compared to Level 2 causes, while the other custom aggregates were ranked compared to Level 3 causes.
The pie chart (Figure 1) shows the composition of each cause. For aggregate causes (Levels 1, 2, and aggregate Level 3 causes) the chart shows global DALYs for the constituent causes. For most detailed causes, the distribution of YLLs and YLDs is shown, and YLDs are listed by sequelae as appropriate. Wedges contributing less than 2% of the total were combined into “other causes” or “other YLDs” for visibility.
For causes that are attributable to at least one risk quantified in GBD, a bar chart (Figure 2, if applicable) details the proportion of global DALYs attributable to those risks. For causes with more than 10 risks, the 10 risks contributing the highest proportions are shown.
The scatterplot (Figure 2 or 3) compares age-standardized DALY rates and Socio-demographic Index (SDI), a composite measure of income per capita, fertility, and education calculated for each GBD location and year. The scale of SDI ranges from 0 to 100, with low values representing countries with low sociodemographic status.
The stacked bar chart (Figure 3 or 4) disaggregates global DALYs into YLLs and YLDs by age group and by sex.
The map (Figure 4 or 5) illustrates the distribution of age-standardized DALY rates by country or territory, along with selected subnational locations.
In place of DALYs, prevalence is shown in the figures for causes that do not result in DALYs (latent tuberculosis, for example).