Still getting the hang of Twitter chatting but...
A2 - start with brushing off your journal club #skillz! There are so many points to make here in terms of identifying confounders and bias, but I'll start w/ one I get asked about a lot - immortal time bias! #ATSChat
A2 - Immortal time = period of time during which, due to study design, the study outcome cannot occur. Occurs when there’s a difference between the timing of entrance into the study cohort & timing of the exposure. Exposure = the treatment being assessed. #ATSChat (2)
A2 - This can bias the results in favor of the exposure (ie treatment) group ➡️ immortal time BIAS. #atschat (3)
A2 - Example:
Population = critically ill adults w/ COVID-19
Exposure = burgers
➡️ Dose = 1 burger TID x 24 hrs – must get 24 hrs of burgers to be in treatment group
Outcome = hospital mortality
#ATSChat (4) Image
A2 - Because of time bias, patients who die within 24 hrs are in the “untreated” group ➡️ gives a spurious survival advantage to the treated group. #ATSChat (5)
A2 - Also - let’s say burgers require GI approval & can only be administered in the ICU ➡️it may take a day or 2 for a pt to get burgers, especially if there’s a shortage of ICU beds (ie pts are critically ill but boarding in ED). #ATSChat (6)
A2 - Time to exposure is even longer ➡️pt has to live even longer (time it takes to get to ICU & get GI approval + 24 hrs of burger administration) to be considered in the treatment group #ATSChat (7)
A2 - In this diagram, it looks like 75% of untreated patients died vs 25% of treated patients. In reality, if you look at just the patients who actually lived long enough to have a chance of getting the treatment, there was only 1 death in each group. #ATSChat (9)
A2 - Another obvious bias NOT related to immortal time - burgers can’t be administered to patients who aren’t conscious or otherwise can’t take PO ➡️ these are probably sicker patients who end up in untreated group. #ATSChat (10)
A2 - this example is silly, and I really want a burger, but this happens all the time - ie - a patient must be on a particular med x 24 hrs to be in the treatment group...a patient must be able to take PO...(11)
...and after so much burger talk, I had to get one. Just as effective as #HCQ for treating #COVID19, but much more delicious. Image

• • •

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

Keep Current with Nida Qadir, MD

Nida Qadir, MD 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 @NidaQadirMD

18 Aug
Very excited to share our most recent publication about #CovidVaccine effectiveness, just published today by the IVY network. Especially relevant in light of today’s news about #boosters (spoiler - I’m skeptical about the need for boosters for all). 🧵

cdc.gov/mmwr/volumes/7…
➡️We looked specifically at vaccine effectiveness for preventing #COVID19 hospitalizations.

➡️We did NOT look at asymptomatic infections, or infections that did not result in hospitalization.

➡️Fully vaccinated = received 2nd dose of vaccine > 14 days prior to hospitalization
What we found:

➡️Vaccine effectiveness (VE) for preventing #COVID19 hospitalizations was 86% overall.

➡️Vaccine effectiveness was sustained for 6 months after full immunization - strikingly illustrated in this figure:
Read 9 tweets
16 Aug
Most young women I’ve admitted to the #ICU for #COVID19 have been #pregnant. Pregnancy is indeed a risk factor for more severe COVID illness, but the messaging about #pregnancy, COVID & #vaccines has led to a lot of confusion and preventable illness. Let’s break it down. 🤰🏽🧵
The reason I pretty much always hear from pregnant women for not getting vaccinated is, understandably, not wanting to put their pregnancy & their baby at any potential risk from taking a new & seemingly unfamiliar vaccine.
But forgoing vaccination exposes mom & baby to a much bigger risk - the known, real, significant risk of getting a severe #COVID19 infection. Both @acog & @MySMFM recommend that pregnant people get the #CovidVaccine in order to reduce that risk.

acog.org/news/news-rele…
Read 11 tweets
20 Jul
Thanks @CNN for sharing this story on @UCLAHealth's collaboration with @LAOpera! In this program, opera performers lead #COVID19 survivors through breathing and singing workshops - it's an innovative (and fun!) adjunct to pulmonary rehab.

cnn.com/2021/07/18/us/…
What does #opera have to do with pulmonary rehab? Well, would it be surprising to know that some of the work we do in pulmonary rehab is quite similar to the #breathing exercises opera singers do? The opera stars just happen to be doing them on an elite level.
For example, diaphragmatic breathing, which we use as a technique to help increase tidal volume, is also used by singers to increase endurance and help them sing longer on a single breath.
Read 5 tweets
19 Jul
Important questions about the type of ethical decision-making that needs to happen when resources become scarce - a situation no one wants to be in, but I worry may again become inevitable as #COVID19 cases continue to rise.
Should #vaccinated patients be prioritized over unvaccinated, all else being equal? I don’t think so, no matter how frustrating this is. Many unvaccinated are victims of misinformation. Also, if we did this, then where else do we draw lines related to personal health behaviors?
What about HCWs? When we were making our crisis standards of care document, there were discussions about the ethics of giving HCWs a tie breaker, all other things being equal. The rationale was that if they got better, they could go back to work & help others, maximizing benefit.
Read 5 tweets
25 Feb
Honored to present at combined medicine & surgery grand rounds @UCLAHealth today & a privilege to highlight the incredible work done by the UCLA #COVID19 front line. Also packed in a review of the evidence basis for COVID therapeutics & discussed COVID recovery.
A lot to cover, and admittedly, I haven’t been that nervous about a presentation in a long time. I’ll share some slides here, starting with a summary of #COVID19 therapeutics.
Reviewing the body of evidence is always a task, even if the last time you did it was a week ago. You’ll get >100,000 results in PubMed if you search for #COVID19. In one year, there are about as many results for COVID as there are for influenza over 30 years 😳
Read 12 tweets
20 Nov 20
Just published in @bmj_latest, an update to our @WHO living guideline on drugs for #COVID19 – this update focuses on #remdesivir & was informed by the results of 4 trials w/7,333 participants – ACTT-1, SOLIDARITY, Spinner (JAMA), & Wang (Lancet). tinyurl.com/y4svlcyn Image
Summary: based on currently available data, there is insufficient evidence that #remdesivir improves patient-important outcomes ➡️ thus, a conditional (weak) recommendation against the use of remdesivir in hospitalized patients with #COVID19.
A conditional, or weak, recommendation against #remdesivir means that, based on current data, the undesirable effects probably outweigh the desirable effects, but substantial uncertainty exists. With this in mind, we very much felt that trials of remdesivir should continue.
Read 8 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

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

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

:(