Skip to main content
American Journal of Public Health logoLink to American Journal of Public Health
editorial
. 2021 Dec;111(12):2096–2099. doi: 10.2105/AJPH.2021.306545

Why Was COVID-19 Not the First Leading Cause of Death in the United States in 2020? Rethinking the Ranking List

Shu-Yu Tai 1, Tsung-Hsueh Lu 1,
PMCID: PMC8667839  PMID: 34878866

In the United States, COVID-19 was the third leading cause of death, after heart disease and cancer, during 2020, according to the provisional mortality data report released by the National Center for Health Statistics (NCHS) on March 31, 2021.1,2 By contrast, COVID-19 was the leading cause of death, followed by dementia and Alzheimer’s disease, ischemic heart disease, stroke, and lung cancer, in England and Wales in February 2021, according to the provisional mortality analysis released by the Office for National Statistics on March 18, 2021.3 One month later, COVID-19 became the third leading cause of death.4 Some people might wonder why the names of the categories used for rankings in the United States (e.g., cancer or heart disease) differed from those used in England and Wales (e.g., lung cancer or ischemic heart disease). That is, the countries did not use the same list in ranking the leading causes of death.

BROAD VERSUS SPECIFIC RANKING CATEGORIES

Rankings of the leading causes of death are the most commonly cited health statistics by mass media and health-related advocacy organizations in many countries because of their easy comprehensibility. However, different countries use different lists for their rankings. Some countries use International Classification of Diseases, Tenth Revision (ICD-10; Geneva, Switzerland: World Health Organization; 1992) codes (e.g., neoplasms, diseases of the circulatory system, diseases of the respiratory system) as a foundation for their rankings. By contrast, some countries use more specific categories (cancer of specific site, ischemic heart disease, diabetes mellitus, land transport accidents) as a foundation. As expected, the use of different lists results in different ranking orders, hindering valid international comparisons.

The rankings that the Office for National Statistics uses are different from those the NCHS uses. The Office for National Statistics includes 65 ranked categories in its list,5 whereas the NCHS includes 51.6 The NCHS list incorporates some broad categories such as cancer, heart disease, and unintentional injuries. These categories have 89, 31, and more than 100 three-character ICD-10 codes, respectively. However, diabetes mellitus has five subcategories (ICD-10 codes E10–E14), and COVID-19 has only one subcategory (ICD-10 code U07.1).6 These discrepancies lead to unfair comparisons between cancer and COVID-19.

MORE SPECIFIC AND ACTIONABLE INFORMATION

Furthermore, health policy decision makers require more specific and actionable information to determine the relative importance of different health problems and subsequently allocate resources appropriately. From the perspective of disease control and prevention, cancers of different sites are heterogeneous in terms of etiology, progression, screening capability, and treatment modalities. Grouping cancers of various levels of preventability and treatability into one category for mortality rankings cannot provide specific or actionable information.

Only 16 cancers are included in the avoidable mortality indicators used by the Organisation for Economic Co-operation and Development and Eurostat to assess the performance of health systems across member countries.7,8 Nine cancers (oral, esophageal, stomach, liver, lung, mesothelioma, skin, bladder, and cervical) are classified as preventable through implementation of effective public health and primary prevention interventions, and eight cancers (cervical, colorectal, female breast, uterus, testicular, thyroid, Hodgkin’s disease, and lymphoid leukemia) are classified as treatable (or amenable) through timely and effective health care interventions.8

Similarly, grouping unintentional injuries (accidents) into a single ranked category masks specific and actionable information. Prevention strategies and countermeasures are quite different for causes such as land transport accidents, falls, drownings, poisonings, and discharging of a firearm. The list should present external causes separately for ranking the leading causes of death, especially for younger people, for whom external causes are the primary causes of death.

PROPOSED STANDARD RANKING LIST

To solve the aforementioned comparability and specificity problems, a standard list for ranking the leading causes of death in the World Health Organization (WHO) member countries was proposed at the World Health Organization Family of International Classifications network meeting in 2004. The principles for constructing the standardized list include the following:

  • 1.

    the grouping of causes should have an epidemiological basis for disease prevention and control;

  • 2.

    residual categories (e.g., other, not specified) should be avoided, and only one such category should exist for all remaining causes;

  • 3.

    a balance between aggregating and disaggregating causes should be sought;

  • 4.

    broad and heterogeneous categories (such as cancer and unintentional injuries) should be avoided;

  • 5.

    the list should consist of mutually exclusive diseases or groups of diseases;

  • 6.

    the list should be based on the current ICD version; and

  • 7.

    a list that meets these criteria will likely contain a range of 40 to 80 candidate categories.9

Despite the disapproval of the US delegates,10 the WHO ranking list was approved by the organization and has been used by many countries and researchers.11–13

COVID-19 IN THE REVISED WHO RANKING LIST

If we apply the revised WHO ranking list5 to 2020 provisional mortality data to determine the rankings of leading causes of death in the United States, COVID-19 becomes the leading cause of death, and several new specific causes such as ischemic heart disease, lung cancer, heart failure, and hypertension are among the top 10 leading causes (Table 1).

TABLE 1—

Ten Leading Causes of Death According to Two Ranking Lists: United States, 2020

Cause of Death (ICD-10 Code) No. of Deaths
National Center for Health Statistics lista
 Heart disease (I00–I09, I11, I13, I20–I51) 690 882
 Cancer (C00–C97) 598 932
 COVID-19 (U07.1) 435 323
 Unintentional injuries (V01–X59, Y85–Y86) 192 176
 Chronic lower respiratory diseases (J40–J47) 159 050
 Stroke (I60–I69) 151 637
 Alzheimer’s disease (G30) 133 382
 Diabetes (E10–E14) 101 106
 Influenza and pneumonia (J09–J18) 53 495
 Kidney disease (N00–N07, N17–N19, N25–N27) 52 260
 Suicide (U03, X60–X84, Y870) 44 834
World Health Organization listb
 COVID-19 (U07.1) 435 323
 Ischemic heart diseases (I20–I25) 369 579
 Dementia and Alzheimer’s disease (F01, F03, G30) 242 273
 Chronic lower respiratory diseases (J40–J47) 159 050
 Stroke (I60–I69) 151 637
 Lung cancer (C33-C34) 143 041
 Heart failure (I50–I51) 108 866
 Hypertensive diseases (I10–I15) 104 527
 Diabetes (E10–E14) 101 106
 Diseases of the urinary system (N00–N39) 73 037
 Accidental poisoning (X40–X49) 67 355

Note. ICD-10 = International Classification of Diseases, Tenth Revision (Geneva, Switzerland: World Health Organization; 1992).

aThe numbers of deaths for 2020 were obtained from Ahmad and Anderson.2

bThe numbers of deaths for specific causes were estimated according to specific cause (such as lung cancer)/broad cause (such as cancer) in 2019 obtained from CDC WONDER.

Using the same WHO list, we were able to compare the rankings of the United States in 2020 with those in England in February 2021, and the categories were as follows: COVID-19, dementia and Alzheimer’s disease, ischemic heart disease, stroke, lung cancer, chronic lower respiratory tract diseases, symptoms and ill-defined conditions, influenza and pneumonia, and hematopoietic cancer. Notably, heart failure, hypertension, diabetes, and diseases of the urinary system were among the 10 leading causes of death in the United States but not in England.

We were also able to compare the rankings of the United States in 2019 with those of high-income countries in 2019 released by WHO; the categories were ischemic heart disease, dementia and Alzheimer’s disease, stroke, lung cancer, chronic obstructive pulmonary disease, lower respiratory tract infections, colorectal cancer, kidney disease, hypertensive heart disease, and diabetes.14 Accidental poisoning was the 10th leading cause in the United States but not in high-income countries overall.

COMPARABILITY WITHIN AND BEWEEN COUNTRIES

The main reason for preserving the broad categories in NCHS ranking list is to retain comparability across times for users at the national and local levels in the United States. The NCHS list was established at the 1951 Public Health Conference of Records and Statistics,6 and sophisticated diagnostic tools were not widely available at that time (the first computerized tomography scanners were installed in the United States in 1973). One pragmatic consideration was the use of the broad category “cancer” instead of “cancer of specific site.” This consideration also existed for the WHO ranking list, which used “stroke” instead of “cerebral infarction or cerebral hemorrhage.” Furthermore, many countries other than United States still use broad categories in their rankings, probably with the similar consideration of comparability across time.

With regard to comparability between countries on specific causes of death, the NCHS has also published detailed cause-specific data on numbers of deaths and death rates.15 If users are not able to find specific causes in that report, they can access CDC WONDER to determine numbers of deaths for four-character ICD codes.16 The numbers of deaths for specific causes in Table 1 were retrieved from CDC WONDER.

In conclusion, ranking the leading causes of death is still a popular means of determining the relative importance of health problems and is a starting point for further comprehensive analyses of cause-specific mortality rates. Using more specific categories for rankings can provide more specific and actionable information for health policymakers. The United States should regularly provide the rankings of leading “specific” causes of death in addition to “broad” causes of death.

CONFLICTS OF INTEREST

The authors declare that they have no conflicts of interest.

REFERENCES


Articles from American Journal of Public Health are provided here courtesy of American Public Health Association

RESOURCES