To understand why this is encouraging vaccine news, note that this is a *cluster* (overdispersed pathogen) *among the elderly* (vulnerable population) *who live together* (allows high attack rate). So way worse scenario compared with trials but vaccines *still* highly protective. https://twitter.com/BillHanage/status/1384968364460843016

I know the overdispersion can be hard to wrap one's mind around, but it's a crucial feature of this pandemic: it produces these clusters where lots of people get infected in contrast with those many cases (likely vast majority) where people don't seem to transmit onwards at all.

So when we look at vaccine efficacy against symptomatic disease in a *cluster* among *the elderly* in a *congregate setting* with a variant that worried us and see numbers this high: it's encouraging. Always, always, always: details matter in evaluating these studies.

Yes, this is exactly what the study is telling us (plus this was a variant that had caused worry!). Raw numbers without context, or trial/cluster efficacy comparisons without also considering the underlying distribution aren't that informative. https://twitter.com/botheritall/status/1384985382597533700

It is very important to study breakthrough cases, even if rare, and sequence them so we can keep an eye on what’s going on. Lab studies can only tell us so much. I just hope every such study doesn’t cause unnecessary alarms. This one showed particularly solid encouraging news.

Also: we can't directly compare efficacy from a *cluster among the elderly in a congregate setting* with a vaccine trial where participants aren't all living together and oversampled from the elderly. Even, if you concocted this, you'd find both in the same confidence interval.

I saw people deduce that this may imply weaker efficacy compared with the trial. That's not a thing we can conclude with a small breakthrough investigation but we can look at this as a case with really encouraging results. Nursing homes have fatalities from common cold outbreaks.

Here's the confidence interval calculation from the paper (though still note that this is an efficacy calculation for the elderly nursing home population in the same congregate setting, different than the trial population which was much younger and indepedent). CI is 65.6-94.7%.

I wrote a post for my newsletter, to try to explain why that particular study of a nursing home outbreak that affected even the vaccinated was really encouraging news. It's also an example of why we need more than "just the facts" to be better informed. https://www.theinsight.org/p/facts-are-pieces-of-a-puzzle-not/comments

My post on why that study of a nursing home outbreak in this thread that affected even the vaccinated was, in fact, encouraging news. It's also an example of why we need more than "just the facts" to be better informed. (Repost with the correct link.) https://www.theinsight.org/p/facts-are-pieces-of-a-puzzle-not