AlanSubie4Life
Efficiency Obsessed Member
France Discovers Undiagnosed Coronavirus Patient From Last Year
Bloomberg so I won't link, but the headline says it all. Frenchy with cv19 admitted back in late December. Just like the far less contagious Swine Flu, we're gonna find out in retrospect that this spread far wider and earlier than anyone projected.
Certainly an interesting find. I'm interested to see where this particular virus genome fits in in the overall tree. It did sound like they already suspected this to some extent, based on the phylogenetic tree they've developed (the French genomes do not correspond with the known travel import cases). Definitely want to get the sequence on this one to rule out any possibility of a false positive (I think unlikely that it is).
It's totally possible (though I think generally unlikely, so would be unlikely to have this happen in all countries around the world) for Wuhan to seed outbreaks elsewhere this early (say, December sometime). It started in Wuhan in late November, so it's possible that an export could have occurred before the number of cases became noticeable there. But since there were so few cases in Wuhan at that time, it would likely only throw a spark to one or two places elsewhere in the world (if any).
But note that France's timeline obviously had to be early than the US - their outbreak got bad earlier than ours. Since known imports to the US started in mid-January (accompanied by unknown imports undoubtedly), it seems entirely possible that France was a few weeks earlier.
And remember that the total number of cases may not be altered that much by the earlier behavior in an epidemic. Remember, the rapidity of spread is stochastic at the beginning. The behavior of the first few imports can dramatically affect how large the epidemic gets. For example, the first introduction could be a superspreader. Or, on the other hand, it could be a solitary individual with few contacts. This will dramatically affect the size of the epidemic down the road.
I doubt this will affect the overall number of infected cases much, though undoubtedly there is uncounted mortality and plenty (in absolute terms) of undiagnosed cases in January in France. But since they're early in the explosive exponential growth, there are relatively few of them compared to the total number of cases now.
We should just rely on good serological testing to determine the prevalence of antibodies.
I imagine the Nextstrain folks will be taking a look at this - keep your eyes open for their analysis.
Last edited: