Here's Why The Coronavirus Death Toll Is Likely Underreported

President Donald Trump thinks the coronavirus death toll figures are being inflated, with officials around him nodding in agreement, Axios reported this week. The outlet suggested the president may soon launch a sustained public attack on the numbers as the 2020 election gets closer.

Over at Fox News, political commentator Sean Hannity picked up a video casting heavy doubt on the figures. In the footage, produced by conservative activist group Project Veritas, several funeral home directors are shown opining that hospitals are defrauding the government by overreporting deaths from COVID-19, the disease caused by the virus.

Fueling the fire, conspiracy theories and misinformation have permeated social media, falsely suggesting the pandemic was manufactured for the purpose of controlling people.

But the crisis is devastatingly real. And far from being exaggerated, experts believe the true number of people who have died from COVID-19 is actually higher than what’s been reported ― more than 80,000 in the U.S. so far.

Here are some reasons why the coronavirus death toll is most likely an undercount — not an overcount.

There was a massive testing shortage early on in the crisis.

We know the virus does not affect everyone the same way. Some people can become infected and pass it on without exhibiting any symptoms. The first confirmed case in the U.S. was a man who had a fever and a dry cough, but not the shortness of breath that has been widely reported.

At the beginning of the coronavirus epidemic, the Centers for Disease Control and Prevention held the nation’s health care system on a tight leash. In late January and early February, the agency was severely limiting the testing criteria patients needed to meet to certain symptoms paired with recent travel to China (or contact with someone who recently traveled there). Testing was expanded at the end of February to people who’d recently been to other places, including Italy and Japan, but you generally still needed to have certain symptoms, like a fever.

Tests were doled out so sparingly in part because there were so few of them early on. The agency botched the process of a nationwide rollout by designing its own test instead of using one of the existing designs other countries were using. The CDC’s test was flawed, and fixing it took precious weeks while the agency refused to permit labs to find other solutions.

With so few tests, the agency essentially had to ration them. There are certainly more people who got sick with COVID-19 toward the beginning of the crisis than are part of the official record, and some of them probably died. Just how many more is unclear.

We don’t know exactly when the virus arrived here and where it went.

Precisely how long the virus has been spreading in the United States and what paths it has taken are also unknown factors. The country saw its first confirmed case in late January, when a Washington state man who had returned from visiting family in China sought urgent care. New York state got its first confirmed case over a month later, on March 1. Residents were initially told the virus migrated to them from China, too, but researchers out of Northeastern University now say it came over from Europe.

While restrictions on travel from China went into place in late January, it was several weeks before the Trump administration blocked travel from Europe, where the virus was wreaking havoc on Italy.

New York Gov. Andrew Cuomo has estimated that more than 10,000 residents may have contracted the disease by the time his state confirmed its first official case. A CDC analysis released Monday suggested the virus’ death toll in New York City is likely several thousands higher than what has been reported.

“There are probably lots of unreported deaths in March in the big cities in the northeast and maybe Southern California,” Dr. Eili Klein, a professor of emergency medicine at Johns Hopkins, told HuffPost.

“There are probably very, very few unreported deaths in other parts of the country,” he added.

It can be difficult to decide how to fill out a death certificate when someone with the virus had a major underlying illness, and states aren’t all doing it the same way.

“If you have somebody who dies of a heart attack and tests positive for COVID-19, is the death due to COVID-19? Or is the death due to the heart attack, and the COVID-19 was completely just a bystander?” Klein said. “If you have somebody who is unwell already, and this precipitates a more rapid decline, you know, how do you measure that impact?”

What exactly killed a patient can sometimes be a tough question for doctors and medical examiners in charge of filling out the certificates in normal times. In the current crisis, figuring out how many people have died because of the virus is complicated by inconsistent reporting practices between the states.

“Those numbers are only as good as the death certificates that they’re based on. And that depends upon the people who are filling them out,” Dr. James Gill, Connecticut’s chief medical examiner and vice president of the National Association of Medical Examiners told NBC News.

Florida’s health department, for example, has only been adding confirmed COVID-19 cases to its death toll. Medical examiners there have been keeping more detailed records, but the state has been trying to keep them under wraps. Meanwhile, New York’s health department began adding in presumed COVID-19 deaths last month as the number of people dying at home skyrocketed.

Even though testing capability has improved vastly across the country since February, tests are prioritized for the living ― meaning that it can be difficult to confirm a patient had COVID-19 after their death.

The CDC’s tally can be influenced by the Trump administration.

As a division of the Department of Health and Human Services, the CDC falls under the purview of the executive branch. A May 13 report from the Daily Beast suggested that Trump and some members of his coronavirus task force are pushing CDC officials to change how the agency works with states to report COVID-19 deaths. The changes could lead to fewer deaths than the agency previously reported by excluding those without confirmed lab results, according to five officials with knowledge of the matter cited in the report.

But there are other organizations keeping track of COVID-19 numbers.

Many news outlets have been publishing coronavirus statistics compiled by Johns Hopkins University, which uses its own data along with data from the Red Cross, the Census Bureau’s American Community Survey and the Bureau of Labor Statistics. Also reliable is The New York Times’ tally, compiled from state and local agencies and hospitals, according to the Yale School of Public Health. Both update more than once per day.

With so many variables, public health experts say that looking at “excess deaths” may actually be the key to understanding the virus’ true impact.

Based on historical data and population trends, we can generally predict how many people are expected to die over time. Subtracting those “expected deaths” from the number of actual deaths gives you “excess deaths.” Epidemiologists believe this metric, measured on a weekly basis, “may be the most objective and comparable way of assessing the scale of the pandemic and formulating lessons to be learned,” according to an article in the medical journal The Lancet.

The CDC produces data on excess deaths, highlighting how many are COVID-19 cases in a series of interactive charts on its website. Perhaps unsurprisingly, the U.S. has logged thousands of excess deaths over the past several weeks, and data is still being collected.

In many states, the number of recorded COVID-19 deaths is actually smaller than the number of excess deaths from any other cause during certain weeks. New Jersey documented around 5,500 excess deaths over a four-week span between March and April, but only about 2,200 of them were categorized as COVID-19 deaths, according to a New York Times analysis. That leaves a gap of some 3,300 deaths.

Several factors could be at play. People may be scared about going into hospitals during the crisis, meaning that they may put off seeking potentially lifesaving care. Hospital resources, including care for non-COVID-19 conditions, may also be stretched too thin. In New York City, the number of people dying at home is six times higher than normal, according to a mid-April ProPublica report. In Detroit, the number is almost four times higher.

“Is that because people were getting infected and the disease is causing more problems?” Klein asked. “Or was that because people were not going to the hospital and then dying at home or calling EMS late? It’s really hard to disentangle those two questions.”

COVID-19 has been known to have strange effects on the body, such as blood clotting and dangerously low oxygen levels, and patients can go from feeling relatively fine to deteriorating with alarming speed. Higher rates of suicide, domestic violence and substance overdoses in these anxious times may also explain some of the gap. Adding to the puzzle is the likelihood that, with so many Americans staying home, some causes of death, such as traffic accidents, may have plummeted.

All told, a full picture of the virus’ deadly impact will probably not be known for several more months.