Skip to Main Content

Long Covid has begun appearing on death certificates for a small percentage of people who have died during the pandemic, but that tiny fraction of records only hints at the whole story, two experts told STAT, while another has doubts about drawing any conclusions from it at all.

Death certificates listing long Covid as a cause of death numbered 3,544, representing 0.3% of the 1,021,487 Covid deaths in the United States from January 2020 through June 2022, according to an analysis issued Wednesday by the National Center for Health Statistics, part of the Centers for Disease Control and Prevention. To compile the report, researchers pulled a variety of descriptions of long Covid — including post-acute sequelae of Covid-19, long haul Covid, post-Covid conditions — from text entered on death certificates. Until recently, there was no specific diagnostic code for or agreed-upon definition of the symptoms that linger after someone’s acute Covid infection has faded.

advertisement

Eric Topol, director and founder of the Scripps Research Translational Institute, dismisses death certificates as unreliable across the board, not just in Covid. “It’s especially clouded because long Covid isn’t known to kill people directly,” he said. “It’s a chronic condition that’s got a lot of ambiguity as to the multiple systems. It’s a mosaic of meaning, so quantifying the deaths is really difficult.”

Other experts say the CDC numbers are at best an undercount and at worst represent a “myopic” view of people living with long Covid. They, like Topol, were not involved in the research.

“The quality of what [the researchers] did is high; the challenge is that the data they have provides only a floor of an estimate of the mortality burden of long Covid,” said Harlan Krumholz, a Yale cardiologist who has been working with long Covid patients. “I am confident that long Covid, in addition to what it does to quality of life, is also causing deaths.”

advertisement

The report notes disparities in long Covid deaths that mirror the pandemic as a whole. In addition to the death rate being highest among adults 85 and older and among men compared to women, non-Hispanic American Indian or Alaska Native people had a death rate that was more than twice as high as people from other racial or ethnic groups. That tracks with American Indian or Alaska Native people being far more likely to die from Covid or have long Covid as a contributing factor to their deaths, likely because they had less access to health care, the authors wrote.

There were exceptions: Despite having higher Covid-19 mortality rates, non-Hispanic Black and Hispanic people did not have higher long Covid death rates than non-Hispanic white people. “These differences may be due to higher mortality among non-Hispanic Black and Hispanic populations, resulting in fewer Covid-19 survivors left to experience long Covid conditions,” the researchers wrote.

The disparities are a sobering insight from the report, but David Putrino, director of rehabilitation innovation at Mount Sinai Health System, is troubled by what he sees as its emphasis on severe acute infection.

“We're still being incredibly myopic about our viewpoint of what long Covid is. And so we're not studying all of the ways that it is causing loss of life,” said Putrino, who started one of the first long Covid clinics in the U.S. “It's very, very clear that the majority of these cases were related to individuals who had severe acute disease and were struggling to recover from severe acute disease. The truth is, there are many, many more — orders of magnitude more — folks who initially did not have severe acute disease, but went on to develop these highly debilitating symptoms.”

That leaves out people who develop long Covid symptoms after milder infection. It also misses two causes of death after long Covid that might not necessarily appear on death certificates: vascular events, like heart attacks following Covid’s known blood clotting abnormalities, and suicide.

“Mental health issues often fly under the radar, but we know that folks with long Covid are taking their own lives,” Putrino said. “It's very rare for their death records to be labeled as related to long Covid. It's important to investigate all of the ways that Covid is continuing to cause the loss of life and loss of function.”

In response to STAT’s request for comment, mortality statisticians at NCHS said via a CDC spokesperson that death certificates are specifically designed to establish causation, pushing back at criticism of them as unreliable, but acknowledging that sometimes they’re not completed properly and sometimes the cause of death is uncertain or unknown. As for the severity of disease, the death certificates captured cause of death as a contributing factor, regardless of how bad the initial Covid-19 infection was. And suicide was not one of the main underlying causes for long Covid deaths.

Long Covid can lead to death in multiple ways, including suicide, Krumholz said. It may exacerbate other conditions, it may cause problems related to its central symptoms, such as fatigue, shortness of breath, cognitive problems — which in turn can lead to vulnerability. It may cause loss of income from missing work, which can also lead to vulnerability. And it may lead to suicide. “All of these scenarios would be related to long Covid,” he said.

The CDC researchers acknowledge limitations, principally the lag in long Covid being assigned an ICD number by the bible of diagnoses known as ICD-10, making it too late for the analysts to use. And another caveat: “The investigation only included death certificates that listed Covid-19 as a cause of death, and as a result may underestimate deaths where prior Covid-19 infection was not confirmed or suspected but may have contributed to the death.”

Topol is skeptical of drawing a straight line of causation between long Covid and death, based only on death certificates and not on a thorough review of a patient’s illness. “When you look at the whole list of symptoms and signs, you don't see a direct link to death. You see fatigue and breathing difficulties and headaches and brain fog,” he said. “People suffer disability, but it isn't known as a killer.”

What's confusing, Topol added, is that there are other symptoms across many different organ systems that can occur in people who don't have the symptoms of long Covid.

“If somebody had a stroke and they've had Covid a couple of months prior, and it was thought that the stroke was a complication of Covid, I don't know that I would sign them off on their death certificate,” he said. “I would say they had a stroke. I can't prove cause and effect. I wouldn't think of it as long Covid. I think of it as a possible Covid complication.”

A limitation throughout the pandemic has been how to diagnose long Covid, how to estimate its prevalence, and how to treat it. Putrino hasn’t seen any deaths among his clinic patients, but he worries about people who aren’t getting better. “The majority of folks experience at least some significant improvement in our clinic, if not complete and total recovery,” he said. “But there is a percentage that we don't know how to treat just yet.”

Krumholz said the CDC report serves one purpose: “It is useful to focus attention that deaths are occurring, but people should not take this small number to be any more than the floor for any estimate,” he said. “The number is undoubtedly much larger.”


If you or someone you know may be considering suicide, contact the 988 Suicide & Crisis Lifeline: call or text 988 or chat 988lifeline.org. For TTY users: Use your preferred relay service or dial 711 then 988.

STAT encourages you to share your voice. We welcome your commentary, criticism, and expertise on our subscriber-only platform, STAT+ Connect

To submit a correction request, please visit our Contact Us page.