Wednesday, April 22, 2020

The Santa Clara Serology Study; Debate; John Ioannanidis with a Target on his Back

The simple story this week is that a new Stanford study of seroprevalence of COVID-19 antibodies show that past infections are far higher than most reports have indicated - Science, April 21, here.  See the Medrxiv.org paper by Bendavid et al., here, posted April 17.

The authors projected a seroprevalence of 2.5% to 4%, being 50-85X higher than the PCR confirmed cases.

For critiques, see e.g. San Jose Mercury News, here.  For some context and back story, TalkingPoints here.  There, John Marshall argues that New York City shows the lowest possible death rate is about 0.5%.  A Bloomberg article reports that the data are mixed and conclusions in either direction are hard to draw - here.  For another article on the debate with quite a bit of context, by Stephanie Lee, hereAnd Lee and others are discussed by a Columbia academician here.

John Ioannidis in NYT

What caught my attention - and my sense of humor - was today's NYT article, by Gina Kolata, with extensive quotes from John Ioannanidis.   Ioannanidis has spent over a decade as a high-profile, highly-quoted critic of other people's sciences, flaws in other people's studies, and generally finding that other people's science was not high quality enough.  One headline - "Ioannanidis, Making Science Look Bad Since 2005."

Here he's on the defensive for a change, and scrambling, and he's archived like a scrambling fly in amber in the NYT.  I couldn't help but feel he was newly in a scenario where the tables were turned.
  • “It’s not perfect, but it’s the best science can do,” said Dr. John Ioannidis, a professor of medicine at Stanford University and an author of the Santa Clara County report.
  • “We tried to look into the possibility of bias influencing the results,” Dr. Ioannidis said. 
    • Adding: "We did a very lengthy set of analyses.”  
    • (Umm...ok...lengthy data scrubbing?)
  • Nobody knows the truth — let’s be honest,” Dr. Ioannidis said of the prevalence figures. 
  • “But if I had to guess [!], I would say it is probably higher than our estimate," he summarized.
    • (Just what should lay people do while conflicting expert guesses fill up the media?)
  • For an update, an interview with Dr. Ioannidis on May 9, here.