2 ms·

> The basic epi coming out of South Africa suggests that prior infection alone is wholly insufficient to prevent against infection with omicron The only study

by unmole 5y ago

> The basic epi coming out of South Africa suggests that prior infection alone is wholly insufficient to prevent against infection with omicron The only study I know of reported a possible reinfection rate of 1.27% for Omicron. While it's higher than Delta, it really doesn't support what you're implying. https://www.medrxiv.org/content/10.1101/2021.11.11.21266068v2 https://www.medrxiv.org/content/10.1101/2021.11.11.21266068v...

donavanm 5y ago
I don’t think that reading is correct. The headline 35k reinfections of 2,796k individuals is for all infections to date. The very recent reinfection data for the last few weeks, which is where omicron is circulating, is much worse: > Based on these analyses, we conclude there was no population-level evidence of immune escape associated with emergence of the Beta or Delta variants. In contrast, the number of daily new reinfections has recently spiked and exceeds the 95% projection interval from the null model, accompanied by a dramatic increase in the hazard ratio for reinfection versus primary infection. The timing of these changes strongly suggests that they are driven by the emergence of the Omicron variant.
- unmole 5y ago
  > I don’t think that reading is correct. You are right. > The very recent reinfection data for the last few weeks, which is where omicron is circulating, is much worse: I can't seem find the numbers that support this assertion in the paper. How many reinfections occurred during the period when Omicron is believed to have been circulating?
reureu 5y ago
The other commenter already pointed out that the 1.27% isn't really a meaningful data point around reinfection risk (beyond the time issue, it's also a proportion of positives, which doesn't get at what we're really interested in: incidence among those with prior infection). But, more broadly, I wasn't referring to a particular study, but rather the general public health surveillance that's happening in South Africa. There's an incredible rise in both cases and hospitalizations in South Africa, in the face of a relatively low vaccination rate (~30%, although this is 10x higher than most of the rest of Africa) with a relatively high estimated prior infection rate (vaccination + prior infection totaling ~90%). One hypothesis that would explain that is that prior infection offers minimal protection from omicron. That's the hunch that epidemiologists and others start exploring from with studies that will provide stronger evidence in favor or against. There's no answers right now, just data points that need to be taken in the context of uncertainty :-/

by unmole 5y ago

donavanm 5y ago

I don’t think that reading is correct. The headline 35k reinfections of 2,796k individuals is for all infections to date. The very recent reinfection data for the last few weeks, which is where omicron is circulating, is much worse: > Based on these analyses, we conclude there was no population-level evidence of immune escape associated with emergence of the Beta or Delta variants. In contrast, the number of daily new reinfections has recently spiked and exceeds the 95% projection interval from the null model, accompanied by a dramatic increase in the hazard ratio for reinfection versus primary infection. The timing of these changes strongly suggests that they are driven by the emergence of the Omicron variant.

unmole 5y ago
> I don’t think that reading is correct. You are right. > The very recent reinfection data for the last few weeks, which is where omicron is circulating, is much worse: I can't seem find the numbers that support this assertion in the paper. How many reinfections occurred during the period when Omicron is believed to have been circulating?

reureu 5y ago

The other commenter already pointed out that the 1.27% isn't really a meaningful data point around reinfection risk (beyond the time issue, it's also a proportion of positives, which doesn't get at what we're really interested in: incidence among those with prior infection). But, more broadly, I wasn't referring to a particular study, but rather the general public health surveillance that's happening in South Africa. There's an incredible rise in both cases and hospitalizations in South Africa, in the face of a relatively low vaccination rate (~30%, although this is 10x higher than most of the rest of Africa) with a relatively high estimated prior infection rate (vaccination + prior infection totaling ~90%). One hypothesis that would explain that is that prior infection offers minimal protection from omicron. That's the hunch that epidemiologists and others start exploring from with studies that will provide stronger evidence in favor or against. There's no answers right now, just data points that need to be taken in the context of uncertainty :-/