Elon Musk thinks the surge in coronavirus cases is due to testing errors, but a virologist is debunking that 'dangerous misinformation'

• Elon Musk said on Twitter that testing errors, rather than new infections, are causing the new surge of coronavirus cases.
• Virologist Angela Rasmussen rushed to debunk the claim, which she called "dangerous misinformation."
• COVID-19 diagnostic tests have relatively low rates of false positives, and spikes in hospitalizations show that transmission really has increased.

Elon Musk took to Twitter on Thursday to claim that testing errors, rather than increasing transmission, are driving the recent surge in coronavirus cases across the US.

"There are a ridiculous number of false positive [COVID-19] tests," Musk wrote. "This is a big part of why C19 positive tests are going up while hospitalizations & mortality are declining."

But Columbia University virologist Angela Rasmussen called that "false and dangerous misinformation," and the two got into an online spat as she debunked Musk's claims.

Diagnostic COVID-19 tests don't give false positives often enough to account for the recent spike of cases, she said. Since June 15, the seven-day rolling average for daily new cases has more than doubled, and the country's total case count has jumped from 2.1 million to more than 2.7 million.

Contrary to Musk's claim, hospitalizations are rising in places with new outbreaks. In Texas, for example, daily COVID-19 hospitalizations have more than doubled in the last two weeks.

After Musk suggested that all people who test positive should get tested again, Rasmussen said, "no one benefits if people with platforms allowing them to reach millions are spreading demonstrably false information and public health guidance."

That's when she got a response.

"Yes, Angela, please show us the graphs/data that prove your point," Musk said, seconding a follower who urged Rasmussen to "challenge the facts and not the credentials."

To bolster his argument, Musk shared New York Times graphics showing the spike in cases alongside a fall in daily death counts. The implication — that the rising case counts must be in error if death counts are falling — is misleading, since it usually takes several weeks for the coronavirus to kill people, and this uptick in cases only began in mid-June.

Death counts lag behind case counts by about three or four weeks, according to Rasmussen and other infectious-disease experts.

"Deaths will likely increase in the next few weeks as time passes and the cases diagnosed earlier in the surge get sicker," Rasmussen said. "The worst is yet to come."

There are reasons, however, why this new surge may not produce as many deaths as the US saw in April and May. But they have nothing to do with testing errors, as Musk suggested. Instead, experts say that because more young people are getting infected this time around, and they are less likely to die. About 80% of US coronavirus deaths through mid-June were people over 65.

Still, deaths are rising in some states. Arizona's seven-day rolling average of new COVID-19 deaths has more than doubled since the beginning of June, from an average of 15 deaths per day to 37.

In the same time period, Texas's average daily death count has increased by 53%.

Musk is wrong: Hospitalizations are not decreasing

Musk has a track record of sharing misleading theories about the pandemic. He previously suggested that deaths were being overcounted and that lockdowns were unnecessary.

In March, Musk also predicted that the US would be seeing "close to zero" new cases by the end of April. That ended up being the virus' first peak.

Responding to Rasmussen's debunk of his statements, Musk lashed out at the field of medicine itself.

"Something's messed up about medicine that's anti-science," he wrote. "In science, you question everyone, no matter who they are. Facts and reasoning are everything, but in medicine too much emphasis is on credentials, often by people who've accomplished nothing but a PhD thesis used by no one."

But Musk's own argument did not seem to take the facts into account. For one, he insisted that hospitalizations are declining.

As a rebuttal, Rasmussen pointed to the rising intensive care hospitalization rates in Arizona, which has one of the country's steepest surges in new cases.

Musk accused her of cherry-picking one state and called her argument "bs." He again shared the same New York Times graphics, even though those do not show hospitalizations.

In fact, hospitalizations have been increasing nationwide since June 17, according to data from the COVID Tracking Project. The project, run by reporters at The Atlantic, also notes that hospitalizations are likely undercounted, especially since Florida does not report those numbers and currently has one of the nation's biggest surges.

It can take weeks for a COVID-19 case to get severe enough to warrant a stay at the hospital, so hospitalizations usually lag behind case counts the same way deaths do. The CDC even notes this on the site where it publishes preliminary hospitalization data. That site shows a decline in the number of people hospitalized with positive COVID-19 test results from April 18 to June 20. But the CDC has no more data after that.

As for diagnostic COVID-19 tests, the likelihood of false positives is low. According to the Massachusetts Institute of Technology, the PCR tests used across the country "almost never" give false positive results.

Rasmussen said that the "bigger issue" is false negatives, which can lead to underreporting of new cases.