IHME Data

Completed as part of the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2021, this dataset provides estimates of anemia prevalence and years lived with disability by 37 underlying causes, three severity levels, age, and sex for 204 countries and territories and selected subnational geographies in five year increments from 1990 to 2021. Please refer to the related publication for information on modeling methods and analysis.

Get Data Files

Stomach cancer mortality rate estimates were produced at the county level in the United States, by racial/ethnic group, for each year between 2000-2019. These estimates were generated using population and deaths data from the National Center for Health Statistics.

This dataset includes the following:

• CSV files of county-, state-, and national-level estimates of stomach cancer mortality rates for each age group, sex, year, and racial-ethnic group (non-Hispanic White [White], non-Hispanic Black [Black], non-Hispanic Asian or Pacific Islander [Asian], non-Hispanic American Indian Alaska Native [AIAN], and Hispanic or Latino [Latino]). Blank cells are for masked estimates
• Code used to generate the estimates

Established in 2015 by the United Nations, the Sustainable Development Goals (SDGs) specify 17 universal goals for achieving "peace and prosperity" by reducing inequality, improving health and education, and more. Each goal contains a number of specific targets and indicators for measurement and is intended to be achieved by 2030. This dataset provides estimates on progress for indicator 5.2.1, the proportion of age-standardized prevalence of ever-partnered women ages 15 years and older who experienced physical or sexual violence by a current or former intimate partner in the last 12 months. Progress on this indicator is reported as index values (scaled 0 to 100) which cover 204 countries and territories from 1990 to 2021. The indicator is a component of SDG 5 (Achieve gender equality and empower all women and girls), target 5.2 (Eliminate all forms of violence against all women and girls in the public and private spheres, including trafficking and sexual and other types of exploitation).

This dataset includes estimates generated by IHME to assess trends in maternal mortality across five racial and ethnic groups in the U.S. The dataset includes MMR (maternal mortality ratio) estimates for Hispanic and any race; non-Hispanic American Indian and Alaska Native; non-Hispanic Asian, Native Hawaiian, or Other Pacific Islander; non-Hispanic Black; and non-Hispanic White females ages 10-54 for each year from 1999 through 2019. The dataset includes national estimates, estimates for each Census region, estimates for each racial and ethnic group and Census region, and estimates for each racial and ethnic group and state. 

Subnational estimates were produced of the overlap of prevalence of target-age children who have never received a dose of diphtheria-tetanus-pertussis-containing vaccine (No-DTP) with that of five related health indicators: 1) children with stunting, 2) mortality among children under 5, 3) children who had diarrhea who did not receive oral rehydration therapy, 4) prevalence of lymphatic filariasis (LF), and 5) individuals who did not sleep under insecticide-treated bednets. Data are presented at the second administrative level for five countries: Angola, Democratic Republic of the Congo, Ethiopia, Indonesia, and Nigeria. Data are presented for the years 2000 and the most recent year of data available for the respective health indicators. Data include designations into population-weighted quartiles according to both prevalence and counts, and at both country-specific and multinational levels. Values for percent overlap and area under the curve (AUC) are also included at the national level.

Researchers at IHME systematically reviewed, identified, and extracted data from scientific literature studies that estimated the reduction in risk of COVID-19 among individuals with a past SARS-CoV-2 infection in comparison to those without a previous infection. The outcomes assessed were reinfection, symptomatic reinfection, and severe reinfection (hospitalization or death). Extracted SARS-CoV-2 lineages were ancestral, mixed (two different specified variants – e.g., ancestral and Alpha), Alpha (B.1.1.7), Beta (B.1.351), Delta (B.1.617.2), and Omicron (BA.1) and its sub-lineages (BA.2, BA.4/BA.5). A total of 65 studies from 19 different countries were identified. The researchers also produced a meta-analysis of the effectiveness of past infection by outcome (infection, symptomatic disease, and severe disease), variant, and time since infection.

Research by the Global Burden of Disease Health Financing Collaborator Network produced projected health spending estimates for 2020-2050 for 204 countries and territories. The estimates cover total health spending, health spending disaggregated by source into three domestic financing source categories (government, out-of-pocket, and prepaid private), and development assistance for health (DAH). Retrospective health spending estimates for 1995-2019 and key covariates (including GDP per capita, total government spending, total fertility rate, and fraction of the population older than 65 years) were used to forecast GDP and health spending through 2050. Estimates are reported in constant 2021 US dollars, constant 20201purchasing-power parity-adjusted (PPP) dollars, and as a percent of gross domestic product.

This version of the Development Assistance for Health (DAH) Database includes estimates for 1990-2021, which are based on project databases, financial statements, annual reports, IRS 990s, and correspondence with agencies. The DAH Database enables comprehensive analysis of trends in international disbursements of grants and loans for health projects in low- and middle-income countries from key agencies. The data are disaggregated by source of funds, channel of funding, country and geographic region, health focus areas, and program areas.
To better understand the data and how to use it, please refer to the IHME DAH Database 2021 User Guide.

This dataset provides estimates of vaccination coverage for meningococcal serogroup A conjugate vaccine (MenAfriVac®) for 24 countries in the meningitis belt of sub-Saharan Africa between 2010 and 2021. Indicators include mean and 95% uncertainty intervals for the estimated coverage for children aged 1 to 5 and children and young adults aged 1 to 29. These estimates include coverage from both mass vaccination campaigns and routine immunization delivery. The estimation process primarily utilized survey report data and country-reported administrative vaccine coverage data.

Estimates from the Global Burden of Disease Study 2019 (GBD 2019) were used to create the Healthcare Access and Quality (HAQ) Index overall and for select age groups in 204 countries and territories from 1990 to 2019. For GBD 2019, HAQ Index construction methods were updated to use the arithmetic mean of scaled mortality-to-incidence ratios (MIRs) and risk-standardised death rates (RSDRs) for 32 causes of death that should not occur in the presence of timely, quality health care.

This dataset contains estimates for the overall and age-group specific indices along with each unscaled HAQ Index indicator by location in 1990 and 2019. Code used to produce the estimates is also available for download.

Results were published in The Lancet Global Health in December 2022 in “Assessing performance of the Healthcare Access and Quality Index, overall and by select age groups, for 204 countries and territories, 1990–2019: a systematic analysis from the Global Burden of Disease Study 2019.”

This dataset contains estimates of deaths, vaccine coverage, hospital resource use, daily infections and testing, mask use, and social distancing due to COVID-19 for a number of countries and subnational areas. The projections for cumulative and daily deaths, infections, and social distancing each include three scenarios:

• The reference scenario is the forecast IHME deems most likely (for more on this scenario, please see the expanded summary in the Data Release Information Sheet under the "Files" tab of this record)
• The 80% mask use scenario assumes all locations reach 80% mask use within seven days
• The antiviral access scenario assumes globally distributed antivirals

The model that produced these estimates incorporated data on observed COVID-19 deaths, hospitalizations, and cases, protective measures, mobility, and other factors. 

Researchers at IHME and the University of Oxford produced estimates of deaths and years of life lost (YLLs) associated with bacterial infections caused by 33 pathogens across 204 locations in 2019. This study extends the results of the 2019 Global Burden of AMR study and uses its overall methodological approach to provide more granular estimates. A variety of data were gathered to inform these estimates, including multiple cause of death data, hospital discharges, minimally invasive tissue sampling, systematic literature reviews, and microbiology lab results from hospitals and national and multi-national surveillance systems, with a total of 343 million individual records or isolates and 11,361 study-location-years collected. These data informed 6 modelling components which were then combined with results from GBD 2019 to estimate the burden of AMR.

Researchers at IHME and the University of Oxford produced estimates of deaths, disability-adjusted life years (DALYs), years lived with disability (YLDs), and years of life lost (YLLs) associated with and attributable to bacterial antimicrobial resistance (AMR) in 88 pathogen-drug combinations for 53 countries within the WHO European Region in 2019. A variety of data were gathered to inform these estimates, including multiple cause of death data, hospital discharges, minimally invasive tissue sampling, systematic literature reviews, and microbiology lab results from hospitals and national and multi-national surveillance systems, with a total of 471 million individual records or isolates and 7,585 study-location-years collected. These data informed 8 modelling components which were then combined with results from GBD 2019 to estimate the burden of AMR. Estimates were produced for two counterfactual scenarios: no infection and drug-susceptible infection.

Global estimates were produced on the impacts of malnutrition in children younger than 5 years with orofacial clefts. Patient data collected by Smile Train clinical partners were merged with the estimates and insights on the direct health consequences of malnutrition from the Global Burden of Diseases, Injuries, and Risk Factors Study. The resulting estimates encompass relative rates of underweight in those with clefts, total children with underweight status and cleft occurrence, excess malnutrition cases in those with clefts, and the associated malnutrition-related consequences including deaths and disease burden. These results are detailed by country, age, sex, and year.

This dataset contains relative risks, excess mortality, and death count estimates computed from probabilistically linked hospital and mortality records for 72,021,918 patients in the Brazilian Public Health System (SUS) between January 1, 2000, and April 21, 2015. Follow-up duration was measured from the date of the patients’ first hospitalization until their death, or until April 21, 2015. Severe mental illness was defined as schizophrenia, bipolar disorder, or depressive disorder by ICD-10 codes used for the admission. Relative risks were calculated with 95% CIs, comparing mortality among patients with severe mental illness with those with other diagnoses for patients aged 15 years and older. Ill-defined causes of death were redistributed according the Global Burden of Disease study (GBD) methodology when present as the underlying cause.

This dataset contains health expenditure by state and the District of Columbia for the United States for the years 2003-2019, reported in spending per person in 2020 United States Dollars (USD). These data include total health expenditure, health spending by payer, and health spending by type of service, sourced from the Centers for Medicare and Medicaid Services (CMS) State Health Expenditure Accounts from 1991-2014 and estimated by IHME for 2015-2019. In addition to true spending estimates, these data contain standardized estimates of total health spending and spending by payer, where estimates were controlled for state variation in age, regional prices, income, population density, and health risk factors. Standardized estimates were produced by IHME for the years 2003-2019.

Estimates were produced for mortality rates, life expectancy, and population at the state level in the United States, and by racial/ethnic group, for each year between 1990-2019. These estimates were produced using population and deaths data from the National Center for Health Statistics.

This dataset includes the following:

• CSV files of state-, and national-level estimates of mortality rates and life expectancy for each age group, sex, year, and racial-ethnic group (non-Hispanic White, non-Hispanic Black, Hispanic, and non-Hispanic, Other). Blank cells are for masked estimates
• Code used to generate the estimates

Estimates were produced for mortality rates and life expectancy at the county level in the United States, and by racial/ethnic group, for each year between 2000-2019. These estimates were produced using population and deaths data from the National Center for Health Statistics.

This dataset includes the following:

• CSV files of county-, state-, and national-level estimates of mortality rates and life expectancy for each age group, sex, year, and racial-ethnic group (all of which are non-Latino, except for the Latino group): White, Black, American Indian and Alaska Native (AIAN), Asian and Pacific Islander (API), and Latino. Blank cells are for masked estimates
• Code used to generate the estimates

These development assistance for human resources for health (DAHRH) estimates are generated using data from IHME’s Development Assistance for health Database DAH, COVID development assistance database and the Organization for Economic Cooperation and Development’s Creditor Reporting System (CRS) online database. The IHME databases enables comprehensive analysis of donor funding aimed towards activities that support the health workforce in low- and middle-income countries from key agencies. The DAHRH estimates are disaggregated by source of funds, channel (disbursing agency) of funding, geographic region (Global Burden of Disease Study (GBD) super region), and the type of human resources for health activities supported (program areas).

This dataset contains estimates of daily and cumulative infections with SARS-CoV-2 from the beginning of the pandemic through November 14, 2021, as well as estimates of cumulative COVID-19 deaths (adjusted for under-reporting) and estimates of the infection-detection ratio, infection-hospitalization ratio, and infection-fatality ratio. These are made available by day for 190 countries and territories – and subnational locations in 10 of those countries – aggregated into 21 regions and 7 super-regions and globally. Methods and limitations for the underlying models can be found in detail in the publication.

Estimates were produced for target-age routine immunization (RI) coverage of 11 vaccines in 107 low- and middle-income countries (LMICs) between 2000–2020. 5x5 km level-coverage was estimated for six vaccines, and second administrative level-coverage was estimated for five vaccines. Coverage estimates were also produced for age-specific RI measles-containing vaccine first dose (MCV1) at the 5x5 km level and combined RI and supplemental immunization (RI plus SIA) MCV1 and second-dose measles (MCV2) at the second administrative level. Estimates were produced using data from household-based surveys.

This dataset includes the following:

• GeoTIFF raster files for pixel-level coverage estimates for vaccines estimated at the 5x5km level
• CSV files of aggregated coverage estimates at the first and second administrative levels for all vaccines

Get Data Files

This dataset contains estimates of the COVID-19 infection-fatality ratio (IFR) for global populations during the period April 15, 2020 to January 1, 2021. IFR is defined as the probability of an individual dying from COVID-19 once infected with the SARS-CoV-2 virus. In these files, IFR is expressed as a percent: deaths divided by infections multiplied by 100. Location-specific estimates include 190 countries and territories, as well as subnational locations in 11 countries and territories. Specific time points for each location include April 15, 2020; July 15, 2020; October 15, 2020; and January 1, 2021. The age-specific IFR estimates are time-invariant and pool data from all locations with age-stratified data; only data from prior to vaccine introduction in each location was used for the age-specific estimates. The analytic process that produced these estimates accounted for several known biases, such as under-reporting of COVID-19 deaths and the waning sensitivity of SARS-CoV-2 antibody tests.

This dataset contains national-level estimates of unadjusted and adjusted COVID-19 infections per capita and infection fatality ratio (IFR) between January 1, 2020 and September 30, 2021 for 177 countries. Cumulative COVID-19 infections per 1,000 population include both unadjusted estimates and estimates standardized for environmental seasonality (measured as the relative risk of pneumonia), population density, gross domestic product (GDP) per capita, proportion of the population living below 100 m, and a proxy for previous exposure to other betacoronaviruses. IFR per 1,000 infections include both unadjusted estimated and estimates standardized for the age distribution of the population, mean body-mass index (BMI), exposure to air pollution, smoking rates, the proxy for previous exposure to other betacoronaviruses, population density, age-standardized prevalence of chronic obstructive pulmonary disease and cancer, and GDP per capita.

(For U.S. Resiliency Data.)

The dataset contains predictions of the incremental cost-effectiveness ratios (ICERs) for the HPV vaccine in 195 countries. Predictions are based on meta-regression estimates of ICERs on three sets of independent variables (true variation measured by country characteristics, intervention characteristics, and bias variables). Data used in the meta-regression analysis are 613 ICERs from 75 studies in the Tufts University’s cost-effectiveness analysis registries, and additional extractions of ICERs for sensitivity analyses.