Thread by @roby_bhatt

Eyre et al (medrxiv.org/content/10.110…) is a tour-de-force. Bravo!
- traced 150k contacts from 100k cases (! - NHS👏🏽)
- Pfizer vax'd index cases had ~5x (Alpha) or 3x (Delta) lower odds of spreading (this is beyond protection by preventing the index case to begin with)
🧵 1/6

2/6
- this despite similar cycle thresholds (Ct) in vax'd vs unvax'd index Delta cases, as others have shown
- Pfizer vax'd contacts had 16x (Alpha) or 10x (Delta) lower odds of being infected
- Ct is not infectivity
- Ct is not infectivity
- Ct is not infectivity
- #VaccinesWork

3/6
- they also saw waning, c/w other studies (incl imo the best studies, the post-hoc crossover analyses of mRNA RCTs). Here, they found 1.2x increased odds of transmission "for each doubling of weeks since 14 days after 2nd vax in index cases" (?), & 1.4x increase in contacts.

4/6
- interestingly, though Pfizer worked better than AZ against Delta overall, it also seemed to wane quicker, at least for protecting contacts (p = 0.002). Similar results seen elsewhere w/ J&J (less waning). Seems like support for 6 mo boost of mRNA vax (hopefully just x1)?

5/6
- contacts of asymptomatic index cases were less likely to test +, but this was more true for Alpha (>3x lower odds) than Delta (<2x lower odds)
- vax lowered measured VL in Alpha but not Delta cases
- higher VL transmitted more than lower VL, but vax helped at any VL (Ct)

6/6
Summary of a super-rich paper: Delta sucks, vaccines work less well against it that Alpha and wane a bit, but they still work really, really well (10x lower odds of being infected, 3x lower odds of spreading if infected).

Score another one for UK NHS (Test & Trace)👏🏽

#VaxUp

Btw, sometimes when I describe papers as "super-rich", it's code for "complicated".

This preprint isn't that. It's clearly written, & w/in limits of my ability to assess, very well done. It's super-rich b/c they traced 100,000 cases to 150,000 contacts FFS!!! Well worth a read.

Important point raised by @CrazyTalk : the odds ratios I cited throughout this thread come from Table 1 and are basically maximal activity (estimated at 14d post-dose 2); they wane from there.

https://twitter.com/roby_bhatt/status/1443871132357177346

Yet another good point (h/t @MarvinH2_G2): since this is a UK study, it probably reflects longer dose interval b/w Pfizer doses. No controlled data I'm aware of on how this would matter, but I might expect more waning w/ the standard 3 wk interval.

https://twitter.com/roby_bhatt/status/1443905333085327360

My takeaways:
- vaccines protect you from getting infected (but aren't perfect)
- if infected, vaccines reduce risk of transmitting to others (but aren't perfect - less good at this)
- both wane
- they're the most important intervention, but not the only needed w/ Delta raging

Anyway, glad ppl seem to be enjoying this thread. I've learned a lot from the discussion already, but I have to get to work on 5 grants due this month (that's at least 2 too many!!!)

Please, read the preprint! Layers of nuance I had to simplify #onhere

medrxiv.org/content/10.110…