A peer-reviewed article in the medical journal Health Affairs predicts that the number of deaths in the United States caused by COVID-19 will increase at least three-fold by the end of the year, from 93,500 to a range of 350,000 to 1.2 million dead.

The US currently has nearly 1.6 million confirmed cases. There are 4.9 million cases internationally and more than 324,000 deaths. Countries including Brazil, Russia, India, Peru and Chile are emerging as new epicenters of the coronavirus pandemic.

The research was conducted by Anirban Basu, a professor at the University of Washington School of Pharmacy and its Stergachis Family Endowed Director of the CHOICE Institute. It was part of a broad effort to study infection and fatality rates in US counties for patients with symptoms. A survey of data collected through April 20 in 116 counties in 33 states produced an infection fatality rate between 0.6 and 2.1 percent.

The rate was centered at 1.3 percent, 13 times higher than the seasonal flu.

“COVID-19 infection is deadlier than flu—we can put that debate to rest,” said Basu in the study’s press released. He continued that the estimated number of deaths “is a staggering number, which can only be brought down with sound public health measures.” These include mass testing for the coronavirus, comprehensive contact tracing and safely isolating and caring for those infected. Such a program does not exist in the United States or the vast majority of the world’s countries.

Moreover, as Basu notes, his current estimates likely undercount the number of eventual dead from the coronavirus, because he conservatively calculated that only 20 percent of the US population will become infected by the end of the year. This assumes that physical distancing measures are maintained, which are now being ramped down at least partially in every state in the country.

As such, it is not out of the realm of possibility that, in the next seven months, 60 to 70 percent of all Americans will be infected. This is a worst-case scenario predicted by many epidemiologists, including Michael Osterholm, director of the Center for Infectious Disease Research and Prevention at the University of Minnesota. In this scenario, between 1.2 million and 4.8 million people in the United States will die of COVID-19.

This horrific scenario is already being played out. Several states that have begun to reopen, including Alabama, Arkansas, Maine, Minnesota, North Carolina and North Dakota, have already seen their number of new cases increase by more than 25 percent over the last two weeks. Since Texas began rolling back restrictions in April, it has seen a 55 percent increase in new cases.

And while Basu makes clear that his model will be updated as new data becomes available, this is already becoming more difficult. It was reported in Florida Today that the manager of Florida’s COVID-19 dashboard, Rebekah Jones, had been fired. In the days since being let go, the dashboard has crashed multiple times and data has been removed from the site without explanation.

In a public announcement of her removal by the state’s Department of Health, Jones warned that she no longer has any control over the data, including “what data they are now restricting.” She also made clear that she was the only one maintaining the database, which likely explains the reports of the system failing.

Ben Sawyer, a professor at the University of Central Florida and director of its LabX, responded by warning that there is “the worry that the scientists within government who can access the full data are being actively censored.” His colleague Jennifer Larson commented along the same lines, stating that, “We would not accept this lack of transparency for any other natural disaster, so why are we willing to accept it here?”

Both professors reportedly contacted the Florida Department of Health to regain their previous access to the data and were rejected on the grounds that the data are “provisional” and that they will have to wait until May 2021 at the earliest.

Despite this, the office of Florida Governor Ron DeSantis issued a statement saying, “The Florida COVID-19 Dashboard was created by the Geographic Information System (GIS) team in the Division of Disease Control and Health Protection at the Florida Department of Health. Although Rebekah Jones is no longer involved, the GIS team continues to manage and update the Dashboard providing accurate and important information that is publicly accessible.”

This is not the first time that there have been questions raised about the accuracy of the coronavirus case count and death toll in the state. In the weeks leading up to the state’s reopening on May 4, there were numerous reports that at least 10 percent of those who had died from the pandemic were not being accurately counted. According to the Miami Herald, health officials in Miami-Dade and Palm Beach counties were asked to restrict access to reports of COVID-19-related deaths, even as these areas saw a spike in their respective new cases and fatalities.

This has become an issue nationwide. The Centers for Disease Control and Prevention tracking for the deaths overall in the US has been delayed since the week ending February 8. This includes both the seasonally expected deaths and the excess deaths as a result of the pandemic. The agency reports that “only 60 percent of death records are submitted … within 10 days of the date of death, and completeness varies by jurisdiction.”

While some of this backlog can be attributed to the surge in deaths caused by the pandemic, no doubt a large portion of it is due to interference by the Trump administration. Earlier this month, White House Coronavirus Task Force Response Coordinator Deborah Birx told CDC Director Robert Redfield, “There is nothing from the CDC that I can trust.” She also claimed that the organization was inflating its death counts by up to 25 percent, even as Anthony Fauci, the nation’s top infectious diseases expert warned the Senate that the “the number of deaths are likely higher.”

The CDC reporting also raises questions about when the pandemic began killing people in the United States. The first official death in the United States was on February 6, but a single death would likely not cause medical examiners across the country to delay submitting death certificates. If there has been a genuine backlog in reporting deaths since that time, it suggests that the pandemic could have been far more widespread far earlier than currently known.