Share this page!

Roby Bhattacharyya Profile picture
Twitter logo
Dec 15 • 7 tweets • 3 min read
As I ponder whether to stay #onhere: can you share some tweets that changed how you view the world?

I'll start w one that pushed me faster than I otherwise would have towards a full embrace of open science:

If you're game, add yours to the thread below.
Ground rules:
1. tweets only, not links to papers/articles (those are an important function of twitter, but I've been surprised how much 280 characters can say)
2. not your own
3. pix OK
4. 🧵s OK
4a. "if you liked this 🧵 please retweet & follow" 🧵s discouraged
A prescient thread (OP is locked now bc of *waves hands*, so not sure if the QT will work) that made me reflect on one way in which academia and the review process can further marginalize scholars in the developing world, to the detriment of us all:
This 🧵, way outside my field, I keep coming back to. When it emerged that life expectancy is Roxbury is >30 yrs less than in Back Bay (two Boston neighborhoods)… did you realize this means death rates are nearly 8x different, 3.5 miles apart? I didn't.
The data viz that crystallizes the life-saving importance of vaccines in the Omicron era more than any other single graph I've seen. (Not sure the image will come through unless you click through the QT):
I'm sure there are more for me, and I'll add them as I think of them, but for now, please add yours.
Ooh, one more on life expectancy by age over the centuries, just b/c the graphic was 🤯 - check out pandemic flu in 1918!

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Roby Bhattacharyya

Roby Bhattacharyya Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @roby_bhatt

Roby Bhattacharyya Profile picture
Dec 17
Two papers this month w outcomes data for BA.5 bivalent booster:
- 30-50% effective vs symptomatic infection
- 73% effective vs hospitalization in those >65 yo (!)

In both cases *compared with ppl already 2-3x vaccinated, many w prior infection*!

BOOST UP!

Brief thread (1/11):
2/11
Study 1 (from early Dec): 360k outpts w/ respiratory infxn were tested for COVID (120k tested +). Those w bivalent (OG/BA.5) boost were less likely to test +, with vaccine efficacy of ~30% if prior dose was 2-3 mo ago, and ~50% if ≥8 mo ago.

cdc.gov/mmwr/volumes/7…
3/11
Study 2 (out yesterday): of 798 ppl >65 yo hospitalized for COVID-like illness, 381 tested + (cases) & 417 tested - (controls). 5% of cases, vs 14% of controls, had received bivalent boost = 73% effectiveness of booster at preventing hospitalization.

cdc.gov/mmwr/volumes/7…
Read 11 tweets
Roby Bhattacharyya Profile picture
Oct 30
Seeing resurgent claims that ventilators somehow killed people early in the pandemic. As best I can tell, this traces back to one JAMA paper that was badly misinterpreted, to the point that a correction was published 2 days later. Here's a slide from a talk I gave in May 2020. 1/ Original text: "88% mortality". Revised: "25%
Here's the paper (Apr 22): jamanetwork.com/journals/jama/…

And the correction (Apr 24): jamanetwork.com/journals/jama/…

I'll explain in case anyone still cares: authors rushed to publish NYC outcomes (👏🏽). Only pts either discharged or dead were adjudicated, & median f/u was only 4.5 days! 2/
Most pts intubated for COVID take a long time to resolve. So if you look only 4.5 days later, most will still be intubated; this was true, 72% still were ("unadjudicated"). Few will be extubated & discharged (also true: 3%). Some will have died (25%); this often happens early. 3/
Read 8 tweets
Roby Bhattacharyya Profile picture
Jun 19
Finally read all 4 FDA briefing docs on u5/u6 vax's. 🧵 on how I'm thinking about each, as an ID doc (adult, not pedi), dad of a 4yo, & close reader of primary COVID lit.

tldr:
- Mrna: ⬆️ efficacy sooner
- Pf: ⬇️ side fx, ⬆️ certainty re dose 3, ⬆️ safety info in kids 5-17

1/16
2/16
Efficacy data is fraught bc Omicron makes efficacy against any infection a challenging goal. But per observational data over past 6 mo, vaccines w comparable efficacy in observational studies (~40% vs symptomatic infection) remain tremendously helpful vs Omicron (see plots).
3/16
So I'll focus on "immunobridging" as a proxy for expected efficacy. This is commonly used in pedi vaccine assessment, bc severe outcomes vaccines prevent in kids are generally too rare to encounter in trials - but at nation/worldwide scale, this is a huge benefit of vaccines
Read 21 tweets
Roby Bhattacharyya Profile picture
Apr 2
Reviewed a few preprints on BA.2 immune evasion for a talk. My synthesis:
- BA.1 & BA.2 show similar immune escape from vax
- BA.2 is about as far from BA.1, antigenically, as Beta was from WT
- ppl w prior immunity, then BA.1 infxn, neutralize BA.1 & BA.2 ~equally well

Nerdy 🧵
2/6
 I like the "antigenic cartography" model for the variants (medrxiv.org/content/10.110…); I see BA.1 & BA.2 as ~ "antigenically equidistant" from ancestral spike, but kinda far from each other too, similar to the cloud of pre-Omi variants.

But, immunity is complex at this point…
3/6
 Folks whose first view of spike was BA.1 (ie, no prior vax or infection) may only be somewhat protected from BA.2 infection (similar to how Beta evaded D614G immunity). But, people with vax, pre-Omi infection, or both; *then* Omi infection, seem to have far broader immunity.
Read 20 tweets
Roby Bhattacharyya Profile picture
Mar 11
2 studies out this wk on masks in schools. One 9-state tour-de-force on transmission, another in Arkansas comparing before & after mask mandates for students & teachers. Both show assoc b/w school mask mandates & ⬇️ COVID spread; the AR one (& physics…) suggest causality. 🧵 1/8
2/8
 Starting w AR study bc I like the time-series analysis (cdc.gov/mmwr/volumes/7…): more cases in schools than local communities, but less so in schools w mask mandates. OK, but that could be bc those schools were also more cautious in general. That's where time-series comes in
3/8
 Comparing each school district w mask mandates vs itself, pre- & post-mandate: striking drop in school rates 1 wk later; community rates dropped too but ~5x less. Districts unlikely to change overall respect for COVID in a wk, suggesting mask mandates caused the difference.
Read 18 tweets
Roby Bhattacharyya Profile picture
Feb 2
This perspective on Omicron's intrinsic severity just out in @NEJM. Pleasure writing with @BillHanage.

nejm.org/doi/full/10.10…

Summary 🧵 below still applies (but the Fig is way prettier now, & writing clearer, thx to NEJM!).

cc @mgh_id @broadinstitute @CCDD_HSPH
An aside: I just started on service, so no time now, but I'm tempted to write a thread later abt my first time being so thoroughly copy-edited. Seriously eye-opening: I'm not sure a single sentence from the original escaped unscathed. (Lmk if you're interested in such a thing.)
We also updated it to reflect a bunch of new studies that have come out on severity. When the working paper came out, it was mostly theoretical; now, I do think immunity clearly a large part, but not all, of what's being oversimplified into an "Omicron is milder" narrative.
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

Send Email!

:(