COVID-19 Was Killing Americans Earlier Than We Previously Thought
Plus: U.S. death toll from COVID-19 surpasses 45,000, Trump threatens Iran via Twitter, and more....
New data contradict current wisdom about COVID-19 on the West Coast. Santa Clara County, in California's Bay Area, might be where the first known COVID-19 deaths in the U.S. took place.
This regrettable distinction formerly belonged to Kirkland, Washington, where someone died of the coronavirus on February 29. But autopsies have now revealed that COVID-19 was responsible for the February 6 and 17 deaths of two people in Santa Clara County. Another person in Santa Clara died of COVID-19 on March 6.
"These three individuals died at home during a time when very limited testing was available," said the county's statement, noting that at the time, "testing criteria set by the CDC [Centers for Disease Control and Prevention] at the time restricted testing to only individuals with a known travel history and who sought medical care for specific symptoms."
"As the Medical Examiner-Coroner continues to carefully investigate deaths throughout the county, we anticipate additional deaths from COVID-19 will be identified," it added.
Santa Clara County Executive Jeff Smith suggested that the news may mean COVID-19 was spreading around parts of the U.S. for "a lot longer than we first believed." Contradicting current wisdom on the matter, Smith said the virus most likely hit some U.S. communities "back in December."
Recent studies from Santa Clara and Los Angeles counties suggest there may have been more COVID-19 cases in California than was previously known. As Jacob Sullum noted here Monday, that would actually be a somewhat positive thing, making the death rate from COVID-19 much smaller than we had previously thought.
"In contrast with the current crude case fatality rate of about 4.5 percent," writes Sullum, "the study suggests that 0.1 percent to 0.2 percent of people infected by the virus will die, which would make COVID-19 only somewhat more deadly than the seasonal flu."
Good news! Right?
Well, maybe. Some scientists are now calling those studies into question.
Yesterday, Will Fithian, a statistics professor at the University of California, Berkeley, laid out the issues in a long Twitter thread (start here).
"Before journalists publicize any more results from this group, they should know that the confidence intervals reported in both studies have no known statistical provenance as of now," Fithian tweeted. "The calculations are not questionable; they are either wrong or unknown."
More on potential problems with the Los Angeles and Santa Clara COVID-19 studies here.
Fithian has been in touch with the authors, and he notes that they have "demographic information they have not yet shared, so it's conceivable a more refined analysis" will come. But for now, it's best to be a bit skeptical about anything related to this study. Apparently, the authors have a history of making questionable assumptions in their research:
In a WSJ op-ed last month, these same two guys used "fraction of NBA players who tested positive for COVID by March 19" as an estimate for "fraction of the COVID-positive population in cities with NBA teams." Craziest thing I've ever read.https://t.co/uhcVLUxngX pic.twitter.com/Ne1R4JL3cP
— Jeremy Cohen (@deepcohen) April 21, 2020
Overall, the U.S. death toll from COVID-19 has now surpassed 45,000.
Reported US coronavirus deaths:
8 weeks ago: 0 deaths
7 weeks ago: 9 deaths
6 weeks ago: 31 deaths
5 weeks ago: 111 deaths
4 weeks ago: 704 deaths
3 weeks ago: 3,834 deaths
2 weeks ago: 12,895 deaths
1 week ago: 26,033 deaths
Right now: 45,039 deaths
— Ryan Struyk (@ryanstruyk) April 22, 2020
The Trump administration said on Tuesday that an executive order temporarily closing all U.S. borders would be signed on Wednesday, though officials could still not offer details about what that would look like.
The move, as we noted yesterday, makes no sense outside pursuing the president's pre-coronavirus policy agenda and making his base feel all warm and fuzzy.
There's "no evidence we've seen that immigrants are associated with the spread of Covid-19 more than anyone else," the Wall Street Journal editorial board weighed in with a Tuesday evening op-ed. And "Trump's economic case is even weaker."
- Trump's latest attempt to distract people from how badly his administration is handling the COVID-19 outbreak may get us into another Middle Eastern conflict (playing the hits never sounded so bad…):
I have instructed the United States Navy to shoot down and destroy any and all Iranian gunboats if they harass our ships at sea.
— Donald J. Trump (@realDonaldTrump) April 22, 2020
- The U.S. House of Representatives will vote today on whether to allow members to vote remotely. Meanwhile in the Senate, Majority Leader Mitch McConnell has rejected a proposal (from Kentucky Republican Sen. Rand Paul) to do the same.
- Typical of reporting on the groups organizing anti-lockdown protests, The New York Times gets a bit breathless over the fact that these conservative groups are—gasp—using "their social media accounts and text and email lists to spread the word about the protests."
- Lots of layoffs have been hitting digital media companies.
- "New York City's coronavirus cases aren't correlated with neighborhood density at all," notes Slate.
- Reason's Brian Doherty on "what each side of the COVID-19 debate should understand about the other."