How Accurate are COVID-19 Death Counts?
According to Johns Hopkins University, the worldwide count for COVID-19 deaths is now over 366,000, including 98,000 in the United States. Testing capacity has ramped up in the United States, with around 400,000 tests being conducted each day. With states beginning to partially reopen, the question remains, how many people actually have died from COVID-19 and furthermore, how can deaths be counted more accurately in the future to benefit the response.
Reporting of COVID-19 deaths has become an increasingly partisan topic. Conservatives and President Trump have questioned the way the deaths are being counted. They cite discrepancies in the methods states use to count deaths. Some states include deaths presumed to be caused by COVID-19 with the confirmed deaths. Other personalities have speculated hospitals are inflating COVID-19 counts for financial gain. Under the CARES Act, hospitals can charge Medicare and Medicaid 20 percent more for treating patients. However, these claims have been debunked, rendering them as conspiracy theories with little basis in fact.
While there is no evidence of COVID-19 deaths being overreported, there is very strong chance the death count is underreported. Dr. Anthony Fauci said before a Senate committee last week that the death count is “almost certainly higher.” Fauci cites that there are people dying at home who can not get a test. This would have been prevalent in March and April when the US testing capacity was extremely low, as the US did not perform 300,000 tests in one day until May 7. According to Forbes, in April, nearly 200 people in New York City per day were dying of COVID-19 in their homes, and their deaths were not attributed to the official death count.
Many have also pointed to “excess” death counts. According to the World Health Organization, excess mortality is “mortality above what would be expected based on the non-crisis mortality rate in the population of interest.” Basically, it is the difference between the deaths over a certain amount of time and the average historical deaths over that same period of time. This difference is relevant as if it shows a spike in deaths compared to the historical average for that period of time, it becomes clear there may be an underlying factor. In this case that factor is COVID-19. The CDC found earlier in May that the number of excess deaths in New York CIty shows an undercount of COVID-19 deaths by about 5000. This is a substantial number as this miscount in just the city of New York amounts to 5% of the reported death count. While excess deaths are not a perfect statistic, they can be a valuable indicator in showing the true amount of COVID-19 deaths, despite states and countries measuring deaths differently. For example, the United Kingdom is only counting deaths of people who tested positive for COVID-19 and then died in a hospital. This leaves out people who died in their homes or nursing homes as well those who died and did not get a test, or tested falsely negative for COVID-19. This strict definition necessarily undercounts the true number of deaths.
Statistics like excess deaths can also give us a better understanding of when COVID-19 actually came to the US. For example, there was a spike in heart attacks in New York City in mid-February. Scientists believe these heart complications were tied to COVID-19, meaning the disease could have been actively spreading for almost a month before there was a significant number of cases confirmed. Another attribute that could be assessed is deaths indirectly caused by COVID-19. According to Yale Medicine, over the past few weeks, hospitals have seen a decline in emergency visits. Some health professionals believe that people are not going to the emergency room because they fear getting COVID-19. According to Yale Medicine Chief of Neurovascular Surgery, Charles Matouk, MD, “The messaging to ‘shelter-in-place’ has left people thinking that if they have a serious condition, it’s not safe to come to the hospital.” In fact, overall visits to the Yale New Haven Health System emergency departments are down about 40%. Countrywide, there was a 38% drop in visits to the hospital for a certain type of heart attack. People are not having fewer heart attacks. COVID-19 has been linked to heart complications, so it’s likely that more are occurring, Instead, these people are attempting to live with their symptoms. This is a dangerous situation as early treatment of heart complications can save lives. If people are ignoring their symptoms to avoid coming to the hospital, there could be a rise in deaths indirectly tied to COVID-19. These deaths result in people refusing to go to the hospital, despite experiencing symptoms of heart attack and stroke, because they fear of contracting the disease.
It is extremely important to understand how many people currently have COVID-19, have had COVID-19, and have died from COVID-19. Not only are these numbers taken into account when deciding when and how to reopen economies, but correctly attributing someone’s death gives their family a sense of comfort. In order to remedy the undercounting issue, some countries such as France and Belgium have begun to retroactively attribute deaths to COVID-19 that were not originally credited to COVID-19.
The US faced similar issues when counting deaths during the H1N1 outbreak 10 years ago. According to the CDC, the number of deaths reported were likely half of the actual deaths due to H1N1. The Director of the CDC at the time of the outbreak, Dr. Thomas Frieden said that when trying to attribute deaths to H1N1, it is better to use science to estimate counts than actually count cases. Similarly, the seasonal flu poses the same predicament. Many people self treat the flu at home, so for every case confirmed, there are several unconfirmed. With people attempting to self treat COVID-19 at home, or avoiding hospitals out of fear, more people are dying at home without their deaths being attributed to COVID-19.
While it is impossible to correctly attribute every death, it seems very likely that the number of deaths caused by COVID-19 is underreported. The United States and the rest of the world must remedy this, for science and for our collective grief.
Works Cited:
Abdelmalek, Dr. Mark, et al. “How Accurate Is the US Coronavirus Death Count? Some Experts Say It's off by 'Tens of Thousands'.” ABC News, ABC News Network, 30 Apr. 2020, abcnews.go.com/Health/accurate-us-coronavirus-death-count-experts-off-tens/story?id=70385359.
CBS News. “H1N1 Deaths Underreported?” CBS News, CBS Interactive, 9 Nov. 2009, www.cbsnews.com/news/h1n1-deaths-underreported/.
Cohen, Joshua. “Underreporting Of COVID-19 Coronavirus Deaths In The U.S. And Europe (Update).” Forbes, Forbes Magazine, 17 Apr. 2020, www.forbes.com/sites/joshuacohen/2020/04/14/underreporting-of-covid-19-deaths-in-the-us-and-europe/#3660485282d7.
“COVID-19 United States Cases by County.” Johns Hopkins Coronavirus Resource Center, coronavirus.jhu.edu/us-map.
“Daily COVID-19 Tests.” Our World in Data, ourworldindata.org/grapher/full-list-covid-19-tests-per-day?time=2020-03-08..&country=USA.
Fichera, Angelo. “Hospital Payments and the COVID-19 Death Count.” FactCheck.org, 21 Apr. 2020, www.factcheck.org/2020/04/hospital-payments-and-the-covid-19-death-count/.
Lederman, Josh. “Official U.S. Coronavirus Death Toll May Be Missing Many Thousands.” NBCNews.com, NBCUniversal News Group, 7 May 2020, www.nbcnews.com/politics/politics-news/official-u-s-coronavirus-death-toll-may-be-missing-many-n1201441.
MacMillan, Carrie. “Hospitals Report Fewer Heart Attacks and Strokes Amid COVID-19.” Yale Medicine, 6 May 2020, www.yalemedicine.org/stories/hospitals-covid-fears/.
“Questions Raised over Accuracy of US Coronavirus Death Toll.” FOX6Now.Com, 16 May 2020, fox6now.com/2020/05/16/questions-raised-over-accuracy-of-us-coronavirus-death-toll/.
Ritchie, Hannah. “Excess Mortality from the Coronavirus Pandemic (COVID-19).” Our World in Data, ourworldindata.org/excess-mortality-covid.
Small, Alonzo. “As Flu Peaks, Not All Cases Tallied.” Delawareonline, 19 Feb. 2017, www.delawareonline.com/story/news/health/2017/02/19/flu-season-delaware-cases/98135590/.
