...the clear reporting lag? Yes, that's officially stated by the public health authority I believe. It's like the Sweden figures - every fool makes a graph and points out that deaths are falling once a week because they don't report deaths within 14 days well
-
-
Replying to @GidMK
The height of the blue line (symptom onset dates that eventually resulted in deaths) is roughly the same (certainly not exponentially higher) at Oct 30 vs Sept 30, which covers the 14-day reporting period. Honestly, what am I missing?
1 reply 0 retweets 0 likes -
Replying to @rubiconcapital_
Quickly eyeballing, looks like it's gone up by ~20% in that time. The reporting lag would also include the 29/30th, and symptom reporting is not a perfect measure. Would be more useful to look at day of death than day of reported symptoms
1 reply 0 retweets 0 likes -
Replying to @GidMK
Agree date of death vs. report would tell better story (date of death not published in daily CSV). Anyway, thanks for the debate and we can leave it there. My prediction is that natural herd immunity is nigh; heterogeneity was not included in epi models, and we see this soon.
1 reply 0 retweets 0 likes -
Replying to @rubiconcapital_
Look, again that is a statement that makes no sense whatsoever. If cases are increasing, herd immunity by definition cannot be "nigh". I honestly think you should look up some of this terminology before throwing it around - read an epi textbook!
1 reply 0 retweets 3 likes -
Replying to @GidMK
It doesn't take an epidemiologist to know we are not all identical dots bouncing around in the same box. We all have different immune susceptibility and contact susceptibility and this simply is not being built into public policy epi models and that is a fact.
1 reply 0 retweets 0 likes -
Replying to @rubiconcapital_
Modelling herd immunity is a fascinating and challenging topic, and it's true that some lower thresholds such as 40-50% may be viable in a fixed population. We have, however, now seen several examples of cities reaching above that threshold
2 replies 0 retweets 1 like -
Replying to @GidMK
Like NYC where infections may be > 30% of pop. In close contact areas (like a cruise ship!) epidemic spread can easily *overshoot* herd immunity thresholds. This is obvious. Certain pops will see outbreaks. But it is morally/ethically/intellectually bankrupt to assume no hetero.
1 reply 0 retweets 0 likes -
Replying to @rubiconcapital_
Mmmm, I'd argue it is intellectually bankrupt to argue about terms without even to bother looking them up. That being said, there have been many efforts to include heterogeneity into HI estimates - as I said, for a fixed population 40-50% is definitely reasonable for COVID
1 reply 0 retweets 0 likes -
Replying to @GidMK
Okay, so that begs an (perhaps *the*) obvious question: Given seroprevalence studies that drive IFR v CFR estimates, and considering, estimated T/B-cell immunity prevalence (coronaviruses 101), where are we (broadly speaking) vs. HIT? Either way, why is no one asking this Q?
1 reply 0 retweets 0 likes
People have been trying extremely hard to answer that question for months mate. The answer (in most places) is that we are very far from herd immunity with <10% infected
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.