Official COVID-19 death counts reveal that more than 1 million deaths have occurred as a result of the pandemic. This number, however, is an undercount. Analyses of excess mortality — a tool used to assess the true mortality impact of the pandemic — have found that more than 170,000 deaths were hidden from official COVID-19 death statistics between 2020 and 2021.
These hidden deaths reflect a mixture of uncounted COVID-19 deaths and deaths that were indirectly related to the pandemic. COVID-19 deaths go uncounted when the coronavirus contributes to a death, but the death is assigned to a cause such as heart disease, respiratory disease, dementia or diabetes without any mention of COVID on the death certificate. This absence is contrary to Centers for Disease Control and Prevention guidelines, which state that death certificates should list COVID-19 whenever the coronavirus was a factor in the death.
Not all hidden deaths are directly related to the SARS-Cov-2 virus. Increases in external causes of death — like drug poisonings, homicides and unintentional injuries — reflect the devastating effects of the pandemic on social and economic wellbeing. And some of the increases in deaths from natural causes like heart disease and diabetes also stem from the pandemic indirectly, such as when patients avoided hospitals because they feared infection or when health care was delayed in overcrowded hospitals. Nonetheless, evidence is mounting that many of the hidden deaths across the U.S. represent uncounted COVID-19 deaths that the coronavirus directly caused or contributed to.
For example, investigative reporting has found significant irregularities in death certification practices across the U.S., especially in rural areas and across the South, where a higher proportion of COVID-19 deaths go uncounted. There is also concern that COVID-19 deaths have been hidden in communities with more Black residents, indicating a potential pattern of structural racism in the death investigation system.
Multiple systemic failures have contributed to the current undercounting of COVID-19 deaths. Like other aspects of the U.S. public health system, death reporting processes are decentralized. Who bears responsibility for a death certificate depends on where the death occurs — whether in a hospital or at home, as well as whether it occurs in counties that employ medical examiners or counties that rely on coroners. Once initial causes are recorded, the National Center for Health Statistics attempts to impose uniform standards in constructing a final set of causes, but all they have to work from is whatever was recorded by the death certifier. Due to this patchwork approach, inaccuracies occur in different ways in different settings.
Physicians, who typically fill out death certificates in the hospital setting, can sometimes lack the appropriate knowledge of their patients to accurately assign cause-of-death. Outside of the medical setting, death certifiers, most often coroners, lack access to patient history or post-mortem COVID-19 tests, making it difficult to certify deaths. Most coroners have no medical training, and unlike medical examiners, they are not forensic pathologists. Coroners’ offices also lack resources to investigate potential COVID-19 deaths in-person and often rely on information that families communicate to them. Some families, who have partisan views about COVID-19, may be unlikely to have pursued COVID-19 testing or to report a COVID-19 diagnosis.
And coroners, because they are elected, can have their own partisan bias. For example, a Missouri coroner is on record saying that he “doesn’t do COVID deaths.” In California, many coroners are also sheriffs, which presents a conflict of interest.
Early in the pandemic, the clinical manifestations of COVID-19 were unclear, and unfamiliar complications may have confused death certifiers. COVID-19 testing has also remained limited in many parts of the country. Rural residents in particular often live far away from the hospital, and their hospitals tend to have reduced capacity, resulting in lower testing rates. Testing rates are also lower in areas that have more people living with comorbidities such as hypertension, obesity and diabetes. This suggests that COVID-19 deaths in these areas may have gone unrecognized and instead been assigned to causes related to the comorbidities.
Undercounts of COVID-19 deaths changed how we responded and will continue to change how we respond to the pandemic.
Local and national government agencies use COVID-19 death statistics to inform the allocation of resources for prevention and mitigation efforts. Incomplete data can lead to incorrect policy decisions. Simply put, you can’t fix what you don’t count. Also, people make behavioral decisions based on their perception of COVID-19 risk. When deaths are undercounted, people may think their community is safer than it is and be less likely to get a vaccine or booster shot, wear a mask, or physically distance. Additionally, families whose loved ones’ deaths go unrecognized can lose out financially. FEMA has a funeral assistance program for families who have lost loved ones to COVID-19. One of the eligibility criteria, however, is that COVID-19 is listed on the deceased’s death certificate. Death certifiers who are undercounting COVID-19 deaths are excluding families from this program.
Reforms to the death investigation system are urgently needed. In 2009, the National Academies of Sciences, Engineering, and Medicine produced detailed recommendations about how to modernize this system. These proposals include federal funding to convert coroner systems to medical examiners and for training forensic pathologists, the establishment of a national working group to promote best practices in death certification, and accreditation requirements for all death certifiers. But these ideas will simply remain proposals on paper unless they are raised to the attention of federal policymakers and significant resources are devoted to implementing them.
The death investigation system’s inaccuracies have hindered our response to the COVID-19 pandemic. Modernizing this system is one of the most important steps we can take to improve our public health infrastructure and prepare for the next pandemic.
Andrew C. Stokes, Ph.D., is an assistant professor in the Department of Global Health at the Boston University School of Public Health. Dielle J. Lundberg is a research fellow in the Boston University School of Public Health. Elizabeth Wrigley-Field is an assistant professor in the Department of Sociology and Minnesota Population Center at the University of Minnesota, Twin Cities.