Slack / Private Message Drop — page 152

Is the paper pre-printed? [2020-09-01 07:35:38] [Kristian Andersen] This is from China, so early in the pandemic - I believe the 273 was patients at that particular location that they could follow up on. They have neut titers in all patients from first and second episodes. 6 out of 273 is indeed very high - I'd say at the upper limit of what you might expect if pretty much everybody were susceptible within a three month period. The paper is preprinted, but I contacted the editor yesterday to see if they want to accelerate - IF the paper is as good as it looks at a first glance. I kinda prefer this isn't preprinted - it needs to be properly reviewed before going out since it'll create an uproar. [2020-09-01 14:21:12] [Eddie Holmes] Not entirely unexpected given some of the common cold coronas. The classic literature (by Phil Minor) was suggesting multiple infection episodes within a year. [2020-09-01 14:36:24] [Andrew Rambaut] But in this case it smells fishy. I don't believe there was enough SC2 around in Beijing for 6 out of their 273 patients even to have multiple exposures of radically different viruses, let alone actually become infected. [2020-09-01 14:40:15] [Eddie Holmes] Perhaps, but the official numbers from China are clearly large underestimates. [2020-09-01 14:41:56] [Andrew Rambaut] I agree but even so. One of the 6 is 2 years old. [2020-09-01 14:42:29] [Andrew Rambaut] (but I don't know how I know that because I haven't seen the paper) [2020-09-01 14:42:54] [Eddie Holmes] Excellent educated guess. [2020-09-01 14:44:04] [Eddie Holmes] There are two different things: (i) does the paper show reinfection; (ii) does it give a reliable frequency of reinfection because of sampling biases. [2020-09-01 14:49:54] [Kristian Andersen] Trying to get clarity on some of these questions - because both are important. I requested the raw data, so going to take a look at that first in case something got mixed up. If these six are actual my high risk (e.g. frontline workers assuming they're immune) than I can believe it. But if they're a random sampling, then it looks really unlikely - and yes, there's a baby in there too.... [2020-09-01 14:50:27] [Kristian Andersen] Or toddler - whatever - little human (I don't know kids...) [2020-09-02 06:52:22]