Share this page!

Leora Horwitz Profile picture
Twitter logo
Apr 9 • 4 tweets • 2 min read
Have been too busy enjoying seeing people in person and seeing so much great research at #SGIM22 to have been tweeting, but what a great meeting. @nyugrossman was out in force /1
Med student Kyle Smith had a wonderful oral presentation on how we have developed a method of finding people on oral anti psychotics who haven’t had a1c testing (no pics cause I was so busy watching!) with @SaulBlecker /2
T32 trainee Rachel Engelberg had a great poster on incarceration and health outcomes /3
new faculty recruit @SafiyaMD presented on digital health equity @devin_mann /4

• • •

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

Keep Current with Leora Horwitz

Leora Horwitz 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 @leorahorwitzmd

Leora Horwitz Profile picture
Jan 30
Fascinating article about how research into types and efficacy of traffic stops in multiple CT communities led to changes that both reduced disparities in stops and better targeted actual public safety issues. Some examples follow:
In Newington, 40% (1,608) traffic stops were for defective lights but found only 1 DUI. Dept switched focus to moving violations (defective lights ⬇️67%, moving violations ⬆️60%). Stops with DUI arrest ⬆️250%, from 18 to 63, and disparities substantially reduced: safer & fairer!
Hamden tried increasing stops for admin issues (lights, registration) to reduce crime in Black neighborhood but rarely found contraband (7%), no effect on crime, caused huge disparities. Switched to stops for hazardous driving: crime ⬇️5%, accidents ⬇️10%, found more contraband.
Read 4 tweets
Leora Horwitz Profile picture
Jan 4
A lot of chatter about hospitalization "with" versus "for" COVID, implying current hospitalization wave isn't "real." NY state is going to start trying to report the distinction; UK already does. Some thoughts, with exemplar data. /1
1st, not so easy to tell. Our health system calls "for" COVID: patients with problem list or clinical impression of respiratory failure with hypoxia (various codes), or x "due to COVID" or COVID positive is the only problem. Specific, but likely not very sensitive. /2
That is, people who meet those criteria are very likely being admitted for COVID, but others will be missed (e.g. diagnosis pneumonia, sepsis, COVID-related stroke/heart attack/PE). So, likely an underestimate. Still, if used consistently, may be useful approximation. /3
Read 8 tweets
Leora Horwitz Profile picture
Dec 22, 2021
Phenomenal preprint from South Africa on #omicron severity. Insanely fast analysis with multiple linked national datasets. Kudos to the authors. Results? You'll see headlines about reduced severity, but full story more complicated. My thoughts. medrxiv.org/content/10.110…
First off, methods. They link lab tests, case data, genome data and hospital data from across all of South Africa. (Wow!) They use a proxy for omicron (SFTF) and require Ct <=30 ("real" infection).
Then they run two comparisons: omicron vs not omicron Oct-Nov, and omicron Oct-Nov vs delta Apr-Nov, and compre frequency of hospitalization and of severe disease (=hospitalised + any of ICU/O2/ventilated/ECMO/ARDS/death). Outcomes assessed on 21 Dec (day preprint posted?!).
Read 18 tweets
Leora Horwitz Profile picture
Nov 11, 2021
Incomplete article in the @nytimes today re: vaccine effectiveness made me finally read last week's NEJM letter suggesting more waning of Moderna than NYT suggests. Will walk you through it. /1 nejm.org/doi/full/10.10…
This reports data from a Moderna randomized trial, in which 14,746 people got vaccine Jul-Dec 2020 and 11,431 got placebo. The placebo group then got the real vaccine later, between Dec 2020-Apr 2021. /2
In Jul-Dec 2020 the vaxxed group had way lower infections than the placebo group (11.8 per 1000 person-years vs. 148.8 cases per 1000 person-years); that's of course why the vaccine got approved. May-June '21, once the placebo group vaccinated, rates were equal. But... /3
Read 8 tweets
Leora Horwitz Profile picture
Nov 2, 2021
Took a Twitter break for several months to help get @NIH #RecoverCOVID up and running to study #LongCovid (AKA #PASC). Thrilled to report @UTHealthSA @UHHospitals have enrolled our first 7 participants! Congrats to them and cast of 1000s working night and day to kick it off. /1
#RecoverCOVID will follow >17K adults, >20K kids across nation + PR for 4 yrs to learn what #LongCovid looks like, how long it lasts, what causes it/increases risk, and, ultimately, what makes it better. Will also include autopsy data and electronic health records of millions. /2
You can find out more about it here recovercovid.org and can sign up to be told more about enrollment opportunities. /3
Read 8 tweets
Leora Horwitz Profile picture
Jun 10, 2021
Grateful to be part of the team @nyulangone leading the clinical science core for the @NIH long term outcomes of #COVID19 research, to be carried out by dozens of extraordinary investigators around the country. I take this responsibility very seriously. nyulangone.org/news/nyu-lango…
“Post acute sequelae of COVID” (PASC) threatens to have major public health implications for years to come, and patients are suffering now. We must learn more about what causes long term effects, and how they can be treated. For more info, see recovercovid.org.
Need to get my fellow PIs Stuart Katz and Andrea Troxel on Twitter, but meanwhile, thanks and admiration for them too, as well as so many of the faculty and staff @nyulangone who stepped up to kick this work off at light speed.
Read 4 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!

:(