Hacker News new | ask | show | jobs
by m-ee 2002 days ago
What evidence are you saying is “quite solid”. So far everything has just a been inferred from the epidemiology and the error bars are extremely wide. I wouldn’t call that solid at all.
2 comments

So, when the news broke a week ago, I was wearing my skeptic hat just like for the handful of previous announcements about "new strains". And some initial reading suggested that there was nothing to worry about. But the one data point that was so eye-popping that I thought it had to be a misprint was that 60% of the cases in London were for the new variant. It wasn't a misprint, and it wasn't isolated to London.

That is the single strongest set of data here. In the areas where both the new variant and the older ones can be distinguished (via sequencing or via the fortuitous FN from the commonly used PCR test), the new one is spreading substantially faster. This measurement is not sensitive to confounding factors such as environmental or behavioral changes, because those would affect all variants equally.

The thing that the TWIV guys seem to love repeating is that this is just a founder effect. I just don't understand where they're getting that idea from. Yes, that's almost certainly been the explanation in the earlier cases we had where one variant become predominant. But that happened in low prevalence environments. It's basically inevitable that if there are few cases, one variant or the other will become predominant just by random chance. That was not the case in the UK. In mid-October, when this variant was still basically non-existent, they were at 15k confirmed cases / day. This variant is not a founder, it was a very successful invader.

The other evidence is weaker, because it's less direct and the data is noisier. (Increased viral loads, many of the changes happening in parts of the genome that were already expected to be of biological interest, correlation studies showing that areas where the new variant is predominant have higher growth rates when controlling for other factors). That data would not be conclusive by itself. But all of it is directionally consistent with the main data point of concern, and strengthens the case.

Agree 100%. I can’t see any alternative hypothesis that explains the UK sequencing data. Here’s a good graph from a seemingly reliable source: https://twitter.com/mugecevik/status/1341094871822680067?s=2...

What could explain that data other than increased transmissibility...?

What you consider a solid evidence?
Some actual lab work? I’m not a biologist so I can’t give you specific assays. The TWiV team might, I’ll submit a question. If you look at the data behind the NERVTAG report it’s extremely noisy, and is circumstantial. Nothing in there directly measures the viruses ability to spread.
The UK is a giant lab right now and the results of the experiment are in: https://twitter.com/mugecevik/status/1341094871822680067?s=2...

What could explain that data other than increased transmissibility?

We’ll have to wait for actual science to come in to answer that. They’re just making inferences from epidemiology and gene sequences. None of this actually does anything to confirm their hypothesis.
30% of people infected not able to walk a block without losing breath and being exhausted for a day — that’s enough evidence to say the risk profile has changed.

Infection rate does not change outcome of recovery (unless it’s resource constrained like vents or meds) and to that point — we’ve walked back the immediacy of vents because they were actually harmful to recovery. So technically the resource management argument should be reduced in weight by some factor.

Where is that 30% coming from?
(Sorry GP) asked a personal opinion question and I offered one —

My gut feeling tells me this pandemic is suffering from a similar phenomenon in user interface latency. The research shows that if the latency is under a few hundred milliseconds, we believe that we are in control of the device and our cause and effect matters. However, if the latency is longer than some hundred millisecond mark, we think the device is under someone else’s control and that our cause has no effect on the system — though in fact, it is the direct cause.

To me, a pandemic kinda follows the same thought — if the disease isn’t having a direct, somewhat immediate effect (intuition says a couple days) to our actions, then it’s hard to issue broad, population level behavioral change that comes with direct costs to life and property - it’s just beyond the timeline for us to accept our actions (not wearing a mask or not distancing) as meaningful inputs to the pandemic system.

Personal opinion: I have no trouble wearing a mask even after having covid, but the current gov’t approach is 100% not the way I’d approach this issue. I would approach the mask mandates and business shutdowns this way only if the disease symptoms were severely impacting (so that’s my symptom description with walking a block) a large set of the population (so that’s the 30%, no age brackets needed) — and to me, that number is about 1 in 3. Now the counterpoint to this, is that if there is a tipping point in the proposed pandemic that once it gets to a certain spread, it becomes that 1 in 3 statistic - but I digress, you get the point I believe.

Edit: to tie this all together — there’s a lot of talk about increased sanctions from this new variant — but if the severity of symptoms do not change, only the transmissibility - it is a dangerous seed to sow amongst an already incredulous population.