Delta progresses more quickly, and all else equal, that should make you less worried about R than you previously were. But I find it plausibly that it’s doing both: it’s scarily infectious and it progresses more quickly. Depending on the methodology of estimating R, it doesn’t necessarily lead to overestimates if one variant progresses more quickly. Your methodology could be to simply trace infections and count how many times a virus was passed on from the earlier generation’s host.
There’s some evidence of scarily high infectiousness. From Australia, we see that it apparently happened twice that there’s CCTV footage of people walking past each other (without mask) and infection happening within 15 seconds of passing. This wouldn’t be too surprising if that’s all the footage we had about scarily quick infections from major outbreaks in the US or the UK – large numbers imply bigger coincidences. However, the Australia outbreak is young and scarily quick infections happening twice implies that they must be rather common.
One thing I’m confused about is the UK’s recent case numbers: They seem to have slowed down a lot. It’s too early to tell and I think there’s a good chance the growth picks up again, and it’ll look very different next week. But there’s a chance that the UK is close to the ceiling already, which would be good news, and probably some evidence that experts overestimated the R of Delta variant, because many experts were predicting >100k or even 200k daily cases over the summer, and this may not happen if the apparent slowdown is real.
(Alternatively, maybe society is really segmented by now and many people are intent on not catching the virus, so saturation is only reached within populations that are comfortable with quite a lot of risk. This could still lead to a slowdown, but not for ideal reasons.)
The 15 seconds transmission (if it’s reasonably common and doesn’t involve one person coughing or sneezing in someone’s face) suggests a quite different infectiouness profile than other variants. You’d expect massive superspreader events from public transport, bars, and events. You’d expect very few people to be infected by family members (because they are exposed to so many people for 15 seconds). I’m not ruling this out but it sounds a bit unlikely based on reports of Delta spread so far.
I’ve seen a bunch of reports that feel intuitively consistent with this. E.g., superspreading around the Euro finals (“the Wembley variant”) and accounts from weddings that I vaguely thought were outdoors, but that could be a false memory. I’d imagine that people would continue to get infected by family members (at least ones that live together) because the secondary attack rate would be really high with such an infectiousness profile. And I’ve also seen some claims from Israel or Australia about near-100% secondary attack rates, but they seemed really anecdotal, so I don’t know. (I expect that it’s already possible to gain a lot confidence about all of this by doing a systematic screening of sources/reports.)
Delta progresses more quickly, and all else equal, that should make you less worried about R than you previously were. But I find it plausibly that it’s doing both: it’s scarily infectious and it progresses more quickly. Depending on the methodology of estimating R, it doesn’t necessarily lead to overestimates if one variant progresses more quickly. Your methodology could be to simply trace infections and count how many times a virus was passed on from the earlier generation’s host.
There’s some evidence of scarily high infectiousness. From Australia, we see that it apparently happened twice that there’s CCTV footage of people walking past each other (without mask) and infection happening within 15 seconds of passing. This wouldn’t be too surprising if that’s all the footage we had about scarily quick infections from major outbreaks in the US or the UK – large numbers imply bigger coincidences. However, the Australia outbreak is young and scarily quick infections happening twice implies that they must be rather common.
One thing I’m confused about is the UK’s recent case numbers: They seem to have slowed down a lot. It’s too early to tell and I think there’s a good chance the growth picks up again, and it’ll look very different next week. But there’s a chance that the UK is close to the ceiling already, which would be good news, and probably some evidence that experts overestimated the R of Delta variant, because many experts were predicting >100k or even 200k daily cases over the summer, and this may not happen if the apparent slowdown is real.
(Alternatively, maybe society is really segmented by now and many people are intent on not catching the virus, so saturation is only reached within populations that are comfortable with quite a lot of risk. This could still lead to a slowdown, but not for ideal reasons.)
The 15 seconds transmission (if it’s reasonably common and doesn’t involve one person coughing or sneezing in someone’s face) suggests a quite different infectiouness profile than other variants. You’d expect massive superspreader events from public transport, bars, and events. You’d expect very few people to be infected by family members (because they are exposed to so many people for 15 seconds). I’m not ruling this out but it sounds a bit unlikely based on reports of Delta spread so far.
I’ve seen a bunch of reports that feel intuitively consistent with this. E.g., superspreading around the Euro finals (“the Wembley variant”) and accounts from weddings that I vaguely thought were outdoors, but that could be a false memory. I’d imagine that people would continue to get infected by family members (at least ones that live together) because the secondary attack rate would be really high with such an infectiousness profile. And I’ve also seen some claims from Israel or Australia about near-100% secondary attack rates, but they seemed really anecdotal, so I don’t know. (I expect that it’s already possible to gain a lot confidence about all of this by doing a systematic screening of sources/reports.)