WHO Mortality Database
Interactive platform visualizing mortality data
The WHO mortality database is a collection death registration data including cause-of-death information from member states. Where they are collected, death registration data are the best source of information on key health indicators, such as life expectancy, and death registration data with cause-of-death information are the best source of information on mortality by cause, such as maternal mortality and suicide mortality.
WHO requests from all countries annual data by age, sex, and complete ICD code (e.g., 4-digit code if the 10th revision of ICD was used). Countries have reported deaths by cause of death, year, sex, and age for inclusion in the WHO Mortality Database since 1950. Data are included only for countries reporting data properly coded according to the International Classification of Diseases (ICD).
Today the database is maintained by the WHO Division of Data, Analytics and Delivery for Impact (DDI) and contains data from over 120 countries and areas. The raw datafiles that comprise the WHO Mortality Database are available here.
Considerations for using data from the WHO mortality database
Users should keep in mind that not all countries report data to the WHO. National death registration statistics with cause of death are generated by a complex, multistep process involving multiple government agencies and hundreds of daily activities1.
In many countries, particularly in sub-Saharan Africa and Asia, these systems have not been institutionalized. Further, medical certification of cause of death requires trained personnel, of whom there is a shortage in many of these same countries.
Even when data are reported to the WHO Mortality Database, the data quality varies by country and over time. There are two main dimensions to quality of mortality data, completeness and quality of cause-of-death information. Completeness refers to the
percentage of deaths that are registered with cause-of-death information.
Even when deaths are registered, information on cause-of-death may be missing, incomplete (ill-defined) or incorrect. The percentage of deaths certified to one of a short list of leading garbage codes – that is, a cause which is not a valid
underlying cause of death or is ill-defined – is an indicator of quality of cause-of-death information.
Together, completeness and percentage of deaths assigned to ill-defined causes of death can be used to determine the quality of the data. An indicator called usability, which combines both dimensions of quality, gives the percentage of deaths that
are registered with a meaningful information on cause of death.
WHO’s Global Health Estimates (GHE) of mortality by cause make adjustments for completeness and quality of cause-of-death information. These corrections allow for direct
comparisons among countries with different quality of death registration data, but methods used for correcting data are uncertain and difficult to validate.
Data in this interactive portal
Data reported by member states and selected areas are displayed in this portal’s interactive visualizations if the data are reported to the WHO mortality database in the requested format and at least 65% of deaths were recorded in each country and
year.
In this interactive portal, deaths are grouped into 189 causes of death for display. The ICD codes assigned to each cause category are available here. For some countries and years, certain causes of death are not available. For example, neither chlamydia nor hepatitis C were included in the 9th revision of the ICD, and therefore no data are available for these causes if underlying data were coded using the 9th revision of the ICD, which was the predominant revision used in the 1980s and 1990s. Further, some member states do not report deaths by ICD code, resulting in discrepancies in the ICD codes included in certain cause categories. These are noted in footnotes that appear when the affected data are visualized.
In most visualizations, it is possible to select number of deaths, percentage of cause-specific deaths out of total deaths, death rate, or age-standardized death rate. Total deaths and death rates (deaths divided by population) may be used to understand the burden of injuries and diseases in each country, for example, for understanding health system needs. Countries that have more young people have lower crude death rates simply because of their young population – sometimes even in cases where people have a higher risk of death at every age. Age-standardized death rates combine age-specific death rates using fixed weights for each age group. These allow for comparisons over time and across countries that adjust for changes in the percentage of younger and older people in a population.
Brief history of the WHO Mortality Database
A brief history of international work standardizing and collecting mortality data by cause of death is provided below. Because uniform classification of causes of death was needed before comparable statistics on cause of death could be collected, key developments in international standards for reporting diseases and health conditions are also noted.
1893
The first International List of Causes of Death was adopted by the International Statistical Institute. In the following decades, the List was revised regularly to reflect advances in health and medical science by the International Statistical Institute, and eventually became the International Classification of Diseases (ICD).
1923
The League of Nations Health Organisation published its first Annual Epidemiological Report to support health policies related to combat emerging epidemics such as smallpox, plague, typhus, relapsing fever and cholera. The aim was “to supply prompt information to the authorities of each country on the health situation in neighboring countries”. These reports were the start of coordinated efforts to gather information from multiple countries to support epidemiological research and better understanding of geographical and seasonal distribution of diseases.
1948
After the demise of the League of Nations, the Health Organisation became the World Health Organization (WHO), founded on 7 April 1948 and based in Geneva. According to Article 64 of its constitution, WHO is mandated to request each Member State to provide statistics on mortality. In support of this, the first World Health Assembly endorsed the sixth revision of the International List of Causes of Death, now called the International Statistical Classification of Diseases, Injuries and Causes of death (ICD). For the first time, this revision included classification of nonfatal diseases, rules for selection of underlying cause of death, and a recommendation to tabulate deaths by underlying cause of death. The landmark development of the ICD has enabled data on causes of death to be generated in a comparable way across the globe and contributes to a large extent to the understanding of impact and evolution of diseases and conditions over time.
1951
WHO published the first Annual Epidemiological and Vital Statistics. This volume, which replaced the Annual Epidemiological Report, covered the period 1939-1946, and was divided into Part I, Vital Statistics and Causes of Death and Part II, Causes of Death from Notifiable Diseases. Countries submitted their mortality statistics on standard questionnaire in paper format which were then verified and compiled to produce the annual report. Although the questionnaire was sent to every country in the world, the volume contained mortality by cause of death from only 33 countries2.
1967
The WHO Nomenclature Regulations of 1967 affirms the importance of compiling and publishing statistics of mortality and morbidity in comparable form. Member States are obliged to provide WHO with the statistics in accordance with the Regulations.
1971
WHO published the World Health Statistics Annual 1968, the first in the series. This volume presented cause-of-death data for 53 countries. The authors noted that the countries covered in the volume account for approximately 24% of the world’s population, and that coverage was particularly poor in Africa (less than 1% of population covered) and Asia (11% of population covered)3. The last print edition of this series, released in 1998, presented mostly 1993-1995 data for nearly 90 countries.
In the late nineteen seventies countries started to submit data to WHO in magnetic tapes which facilitated data processing. The annual reports however published only a summary of the data. In response to the need for detailed information, WHO made available for public use standardized transcripts on electronic storage media such as magnetic tape or diskette.
1999
With the advancement of digital technology, the WHO Mortality database was published online. For the first time, scientists around the world could freely download an international series of cause-of-death data from 1950 on to support their research. This database underpinned core research work and contributed to the development of health policies and improvements in population health. Some examples are described below.
Number of countries and areas with data in the WHO Mortality Database by ICD code and year, as of April 2022.
Because of the typically observed lag of 18–24 months before countries report finalized latest data, it should not be inferred from these charts that reporting for the most recent years has decreased.
2022
The mortality database currently contains data from over 120 countries. In response to user’s need for visualizing and efficiently accessing data in the WHO Mortality Database, this interactive portal was launched.
How the Mortality Database is used
The WHO mortality database is the premier data source for international comparative epidemiological studies of mortality by cause. It provides key empirical data underpinning WHO’s reporting on sustainable development goal indicators on noncommunicable diseases, suicides, maternal mortality and road traffic injuries, among others. Some examples of seminal uses of the mortality database are given below.
Mortality from smoking
A challenge in early efforts to estimate the population health effects of tobacco smoking was accurately measuring exposure to tobacco: the cumulative effects from smoking depend on factors such as the age of initiation, intensity of smoking, duration and type of tobacco product (low tar, high tar, filtered, unfiltered and nicotine content). However, these factors were not reflected in smoking prevalence data. In 1992, Peto et al published4 a novel metric, the smoking impact ratio (SIR) to estimate tobacco-attributable deaths in developed countries while accounting for differences in tobacco use patterns. The authors used country-specific lung cancer mortality rates from the WHO Mortality Database to convert the smokers in each country into equivalents of smokers in an epidemiological study (the American Cancer Society’s second Cancer Prevention Study, CPSII) for whom relative risks of smoking for other causes are available. In other words, they used country-level data on lung cancer mortality rates as a proxy for smoking prevalence. The authors combined the SIR values with relative risks from the CPSII epidemiological study and mortality rates by age, sex, cause and country from the WHO Mortality Database to estimate total mortality attributed to tobacco smoking. The authors found that the average loss in life expectancy for those killed by tobacco in middle age (35 – 69 years) was about 23 years, and estimated that just over 20% of those living in developed countries would eventually be killed by tobacco. These findings provided major impetus for further tobacco control actions.
Trends in age-standardized lung cancer mortality rate, 1950-2018, selected countries.
WHO’s Global Health Estimates
In 1993, the first Global Burden of Disease (GBD) study was featured in the World Bank’s World Development Report 1993: Investing in Health. At the time, it was the most comprehensive effort to estimate mortality by an exhaustive set of over 120 causes for the world’s population. It used data from the WHO Mortality Database where available. Today, the World Health Organization regularly publishes Global Health Estimates of mortality by age, sex, and cause, globally and by country. For 67 member states and Puerto Rico and for most causes of death – where data meet quality criteria – estimates of deaths by cause are computed from the data submitted to the WHO Mortality Database. These estimates include limited adjustments for incomplete reporting of cause of death (if needed) and for deaths assigned to ill-defined causes. For the remaining countries, estimates are derived from statistical models prepared by the GBD project, currently coordinated by the Institute for Health Metrics and Evaluation in Seattle. The GBD project uses data from the WHO Mortality Database together with other sources of data such as verbal autopsy data to estimate mortality by age, sex, cause, and country.
Source: WHO Global Health Estimates 2019
Avoidable mortality
In addition to identifying the diseases and injuries that contribute most to mortality, health planners want to know how much of this can be avoided and at what cost. This information can also be used to evaluate the performance of health systems. In 1976, Rutstein et al. proposed a list of sentinel conditions from which "disease, disability or death" should not occur if appropriate care was provided – the "airplane crashes of the health care system.” 5 This concept of avoidable deaths aimed to assess the quality of health services. A series of studies, some of which included multi-country analyses using the WHO Mortality Database, have operationalized and refined this concept. Debate has focused on whether avoidable mortality should include deaths that could be prevented by public health activities, for example by anti-tobacco policies, or if it should narrowly assess the quality of health care received, that is, should include only causes of death that are amenable to health care. In 2019, Eurostat and the OECD published a joint list of causes for regular reporting of avoidable mortality, comprising both preventable and treatable (amenable) mortality. They have defined preventable and treatable mortality as follows:
- Preventable mortality: Causes of death that can be mainly avoided through effective public health and primary prevention interventions (i.e. before the onset of diseases/injuries, to reduce incidence).
Treatable (or amenable) mortality: Causes of death that can be mainly avoided through timely and effective health care interventions, including secondary prevention and treatment (i.e. after the onset of diseases, to reduce case-fatality).
The OECD computes avoidable mortality using the WHO Mortality Database for publication in its Health at a Glance series.
References
- 1 Muñoz DC, Abouzahr C, de Savigny D. The ‘Ten CRVS Milestones’ framework for understanding civil registration and vital statistics systems. BMJ global health. 2018 Mar 1;3(2).
- 2 WHO. Annual Epidemiological and Vital Statistics 1939-1946. Part 1: Vital Statistics and Causes of Death. World Health Organization: Geneva, 1951.
- 3 WHO. World Health Statistics Annual 1968. World Health Organization: Geneva, 1971.
- 4 Peto R, Boreham J, Lopez AD, Thun M, Heath C. Mortality from tobacco in developed countries: indirect estimation from national vital statistics. The Lancet. 1992 May 23;339(8804):1268-78.
- 5 Rutstein DD, Berenberg W, Chalmers TC, Child CG, III, Fishman AP, Perrin EB. Measuring the quality of medical care. A clinical method. N Engl J Med 1976 March 11;294(11):582-8.
For questions or more information, please contact [email protected].