Beginning early in 2020, the COVID-19 pandemic caused unprecedented loss of life around the world and, over the past two years, at least 900,000 deaths in the US alone. In December 2021, the Centers for Disease Control and Prevention (CDC) released the Mortality Multiple Cause files for 2020 as part of the National Vital Statistics System. These data include individual records for deaths within the US, including detailed cause of death information for both underlying and contributing factors. The CDC also added the data to the CDC Wonder tool allowing researchers to calculate the mortality rates for specific geographies or groups that would otherwise be suppressed as identifiable or limited in the individual-level data. According to this data, 350,831 death certificates listed COVID-19 as the underlying cause of death in 2020. This corresponds to a COVID-19 mortality rate of 106 people per 100,000 making COVID-19 the third leading cause of death in the country behind only heart disease and cancer.

The profound and unexpected death toll of COVID-19 has highlighted the importance of high quality mortality data to understand not only the ways in which New Yorkers and Americans die, but how different conditions interact with each other to contribute to the disparities in life expectancy and mortality rates across race, ethnicity, geography, gender, and age. Many of these disparities exist in the severity and treatment of illnesses in addition to their mortality.

To shed light on the role mortality data plays in understanding health and health disparities, in this analysis we highlight the total mortality, crude mortality rate per 100,000, and age-adjusted mortality rate per 100,000 for all of the underlying conditions that the CDC listed as one of the “15 leading causes of death” in the past 20 years, excluding COVID-19. These 15 leading causes include both individual conditions like Alzheimer’s disease which correspond to a single cause of death (ICD-10) code and groupings of similar conditions like “Diseases of the Heart” which correspond to multiple ICD-10 codes.

Using the data from the CDC, we created an interactive map to visualize trends in mortality for the US. We were struck by the following five facts below.

The Leading Causes of Death Are Remarkably Stable Over Time and Geography

We identified the 15 leading causes of death for each year from 2000 to 2020 which resulted in a total of just 17 distinct causes of death. Heart disease and cancer are overwhelmingly the two most common causes of death in all years and in all states. In 2019, cancer and heart disease accounted for nearly half (49.5 percent) of all deaths in New York. Across all states and years, cancer and heart disease combined account for between 38 and 60 percent of deaths, respectively.

Although causes of death are predominantly stable there have been some substantial changes over the years. The mortality rate of assault (homicide) has gone down dramatically over the past two decades and homicide is no longer one of the 15 leading causes of death in the US. In New York, the homicide mortality rate declined between 1999 and 2019 with one notable exception which was a substantial uptick in 2001 tied to the September 11th attacks, as the CDC considers death due to terrorism in the same category as homicide. The homicide mortality rate increased in 2020 for the first time since 2001 which follows the national increase in homicide deaths. However, New York’s homicide mortality rate remains far below the national average. The increased rate of homicides did not affect all New Yorkers equally and had a disproportionate effect on people of color, particularly Black and Hispanic New Yorkers who already faced higher homicide rates.

Homicides per 100,000 by Race and Ethnicity in New York State

Mortality Data Highlights Racial and Ethnic Health Disparities

Black and Hispanic New Yorkers have a higher all-cause age-adjusted mortality rate than white New Yorkers which corresponds to lower life expectancy. The three causes of death with the largest disparities are assault, hypertensive disorders, and diabetes. Black New Yorkers are 12 times more likely to die from assault, and more than twice as likely to die from hypertensive disorders and diabetes compared to white New Yorkers when controlling for age. The patterns for Hispanic New Yorkers are similar as they are nearly four times as likely to die from assault, 1.4 times more likely to die from hypertension, and 1.3 times more likely to die from diabetes than white New Yorkers.

Influenza and Pneumonia is the only communicable disease category that was a leading cause of death before the COVID-19 pandemic began in 2020 and, notably, both Black and Hispanic New Yorkers were more likely to die from the flu than white New Yorkers (when controlling for age, Black New Yorkers were 30 percent more likely to die and Hispanic New Yorkers were 26 percent more likely). COVID-19 would result in similar disparities between New Yorkers of color and white New Yorkers.

Difference in Age-Adjusted Mortality Rate for Black and White New Yorkers

The Way Causes of Deaths Are Aggregated Can Obscure Policy-Important Patterns

We followed the CDC’s cause of death groupings for identifying the leading causes of death which resulted in the following categories:

Yearly Leading Causes of Death 1999-2020
Cause of Death	ICD-10 Code(s)
Diseases of heart	I00-I09,I11,I13,I20-I51
Malignant neoplasms	C00-C97
COVID-19	U07.1
Accidents (unintentional injuries)	V01-X59,Y85-Y86
Cerebrovascular diseases	I60-I69
Chronic lower respiratory diseases	J40-J47
Alzheimer disease	G30
Diabetes mellitus	E10-E14
Influenza and pneumonia	J09-J18
Nephritis, nephrotic syndrome and nephrosis	N00-N07,N17-N19,N25-N27
Chronic liver disease and cirrhosis	K70,K73-K74
Intentional self-harm (suicide)	*U03,X60-X84,Y87.0
Essential hypertension and hypertensive renal disease	I10,I12,I15
Parkinson disease	G20-G21
Septicemia	A40-A41
Assault (homicide)	*U01-*U02,X85-Y09,Y87.1
Pneumonitis due to solids and liquids	J69

In general, combining similar causes of death into groups is valuable in understanding the prevailing patterns in how people die. For example, separating out each different type of cancer or heart disease would obscure the fact that together, those two groups of illness account for nearly 50 percent of deaths. While grouping similar illnesses may appear intuitive, grouping external causes of death (accidents, suicide, and assault) by their intent rather than method obscures a major cause of death for Americans: drug overdoses. Drug overdoses are included in the data but they are split between three aggregate categories above: Accidents, Intentional self-harm, and Assault. There were 4,865 overdose deaths in New York in 2020 which would make overdoses alone the eighth leading cause of death in the state above influenza and pneumonia, Alzheimer’s, and well ahead of total suicides and assaults.

Accidents were the fourth leading cause of death for New Yorkers in 2020 below only heart disease, COVID-19, and cancer. However, grouping all accidents together may not be valuable for policy makers who wish to reduce their lethality. Policies targeting motor vehicle collisions, workplace accidents, and drug overdose deaths are inherently different. While it is true that different types of cancer require radically different treatments, the way treatment and research into treatment is funded at a state and federal policy level is fairly similar, whereas the oversight of highway safety and overdose prevention are not.

New York Has a Relatively Low Age-Adjusted All-Cause Mortality and High Life Expectancy

Before the COVID-19 pandemic hit New York in early 2020, New York had one of the lowest all-cause age-adjusted mortality rates in the country ranking 49th of 51 in mortality. New York’s age-adjusted mortality rate was 616 in 100,000 which represents many years of both relative and absolute improvement in life expectancy since the 1990s. In 1999, New York ranked 32nd of 51 in mortality and had a 35 percent higher mortality rate at 833 deaths per 100,000 residents.

The COVID-19 pandemic, however, caused a stark reversal in this trend. From 2019 to 2020, the mortality rate in New York jumped nearly 30 percent to 797 deaths per 100,000 largely as a result of the COVID-19 pandemic hitting New York earlier than other states. New York saw an initial peak in COVID-19 deaths in April 2020 while the rest of the country did not see deaths seriously increase until the peak of the second wave in January 2021 which is outside the scope of the 2020 mortality data. Due to this early surge in New York, the state dropped in rank from 49th all the way to 35th which is the highest mortality rank relative to other states that New York has seen since the 1990s.

All-Cause Age-Adjusted Mortality Rate and State Ranking

There Are Persistent All-Cause Mortality Gender Disparities

Adjusting for age, men in New York are more likely to die from all major causes of death except for Alzheimer’s disease. This higher all-cause mortality contributes to the fact that men have a lower life expectancy than women. Notably, the two cases of death that most disproportionately affect men are suicide and accidents. Men are 3.5 times more likely to die from suicide controlling for age compared to women and 2.5 times more likely to die in an accident.

There are a number of reasons for the gender disparity in suicides and accidents. One group particularly at risk of suicide is veterans, who are disproportionately male. Additionally, men who attempt suicide tend to choose more lethal methods. In 2019, 55 percent of men who committed suicide used a firearm compared to only 31 percent of women. Gender disparities in accidental death also have complex causes. Men tend to have more physical jobs, engage in more risk taking, and are more likely to abuse alcohol and other drugs that can lead to accidental death.

Female:Male Age-Adjusted Mortality Rate in New York State
(Female = 1)

Conclusion

COVID-19 has highlighted the importance of analyzing and understanding high quality mortality data not just to understand how New Yorkers and Americans die, but the ways that they live and the health disparities that exist among racial and ethnic groups, as well as genders and geographies. COVID-19 was complicated by and magnified many of the chronic illnesses that make up the leading causes of death including diabetes, heart disease, cancer, and lower respiratory diseases.

In addition to this brief dive into all-cause mortality, we will further discuss firearms and overdose deaths. Firearms and overdose deaths are generally aggregated into the categories of assault, intentional self-harm, and accidents but we believe it is important to disaggregate these causes of death because of their unique policy implications.

ABOUT THE AUTHOR

Leigh Wedenoja is senior policy analyst at the Rockefeller Institute of Government