Nick Mark MD Profile picture
Mar 3 9 tweets 5 min read
New embarrassingly bad study of ivermectin is a textbook example of “confounding by indication”

In short a retroactive database review found that people w/ COVID did worse if given remdesivir (only given to sick inpatients) than ivermectin (only given to well outpatients)🤔
1/
There are many problems with this “study”.

First it’s generous to even call it a “study.” It’s an *abstract* of a retrospective database review. The whole “article” is less than a page (see below).

It wasn’t pre-registered.

sciencedirect.com/science/articl…
2/
The methods are weird.

They found 1.7 million people w/ COVID. Out of this group they identified 1,072 who received ivermectin (not exactly widely used) & 40k who received remdesivir

The groups were very dissimilar: IVM was 10 yrs younger. They don’t report any comorbidities
3/
Despite the groups being wildly dissimilar, they *somehow* adjusted for potential compounders. (They don’t say how)

Crucially they did NOT adjust for hospitalization status. This means they compared COVID outpatients (on IVM) to COVID inpatients (on remdesivir). 🍎to🍊

4/
The authors are interesting…

The first two are medical students, the 3rd is a plastic surgery resident.
(Nothing wrong with this, though it is weird for the MS not to be affiliated w/ SOM)

The fifth & sixth authors are plastic surgeons. Neither has expertise in ID or COVID
5/
The 4th author @JoseGonzalesZa1 is the head of the ID fellowship at UMiami. He is an actual expert on this topic.

Based on his other work, I’m pretty sure he doesn’t think ivermectin is an effective therapy as this abstract concludes. See translation of one of his articles👇
6/
All of this makes me wonder - did the authors all even read/approve this abstract?

Did the peer reviewers think it was odd that plastic surgeons were doing research on ivermectin for COVID?
7/
I wonder if this abstract was the work of an eager group of students/residents who were inadequately supervised by more experienced researchers

That seems like the only way to explain such an obvious confounder being missed

This must have slipped past peer review at #IMED21
8/
Bottom line:
-this abstract of a retrospective review is fatally flawed due to the confounder of hospitalization (+more). There is no evidence that ivermectin reduces mortality
-it won’t stop fringe docs cashing in on selling ivermectin from promoting this “massive study”
9/9

• • •

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

Keep Current with Nick Mark MD

Nick Mark 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 @nickmmark

Mar 3
The FLCCC’s “data analyst” (who has no pharm training) is dosing ivermectin horse paste

Only problem: he makes a FATAL math error

A tube of horse paste contains 6gm (6080 mg) of ivermectin at a concentration of 18.7 mg/ml

@TwitterSafety suspend this guy before he kills someone ImageImage
This guy actually got both calculations completely wrong.

In the ACTIV6 trial an 80 kg person receiving 0.6 mg/kg of ivermectin would receive:
80 kg * 0.6 mg/kg = 48 mg per day

(Somehow he calculated 0.24 mg so he was off by 200x; another huge error 😱)
Bottom line:
- don’t take animal meds EVER; it’s very easy to make a fatal mistake
- don’t take medical advice from a guy with zero training who can’t multiply two numbers correctly
Read 4 tweets
Feb 12
In confronting COVID misinformation I’ve mostly focused on inpatient treatment (this is my area of experience).

Recently I saw the FLCCC Long COVID “protocol” & oh boy is this some crazy non-evidence based prescriptions: HIV meds, steroids, diuretics, & of course ivermectin…
1/
First off, Long COVID is definitely “real” & can be severe.

Many studies have found persistent changes in immune cell phenotype & function, months after COVID infection.

Many great docs (@WesElyMD & others) are actively researching long COVID to improve our understanding.
2/
What concerns me is FLCC presenting “protocols” as proven treatments for long COVID.

Throwing 20 medications (9 are prescription 🟥) at a problem with minimal (or no) evidence is irresponsible.
nature.com/articles/s4159…

As we will see, this is both unethical & likely harmful.
3/
Read 34 tweets
Feb 2
Do you want to improve working conditions for residents/fellows in WA?

The WA legislature is voting on HR1764 which would allow interest arbitration in collective bargaining.

This would go a long way towards correcting the power imbalance between administration & residents.
1/
Under the current system of negotiation, residents can *request* certain changes: safe working conditions, sufficient PPE, hazard/overtime pay, etc.

Administrators (who are often non-clinical) just say NOPE.

Because residents can’t/won’t strike they don’t have much leverage.
2/
When we formed the UWHA/@RFPU_NW resident union we spent YEARS fighting with administrators over these issues.

Ultimately we got lucky. Pressure from the city council & media attention helped us break the logjam.

There ought to be a better way.

3/
dailyuw.com/news/uw-medica…
Read 8 tweets
Feb 1
Instead of proposals to cap RN pay I’d love to see legislation that:
1️⃣mandates safe RN staffing ratios nationwide
2️⃣enacts a “50 state medical license” w/ straightforward reciprocity
3️⃣imposes limits on executive compensation for any hospital/org that bills CMS
For those doubting these reforms are possible. A couple points:

There already is a federal law calling for “adequate numbers of licensed RNs” 42CFR 482.23(b) The issue is that this is too vague.

15 states have passed laws that go further. CA & MA explicitly stipulate RN ratios
The CA law, enacted in 2004, mandates 1 RN to 5 med/surg patients & 1:2 for ICU patients.

After implementation RNs cared for one fewer patient on average. There was a decrease in hospital mortality & increased RN job satisfaction. ncbi.nlm.nih.gov/pmc/articles/P…
Read 6 tweets
Jan 21
The ivermectin crazies are now recommending hydroxychloroquine too.

Their “protocol” includes a dangerously high dose of diuretics & recommends high dose steroids in people not on supplemental O2.

This has crossed the line from (mostly) harmless nonsense to actual harm.
Supporting Evidence:

A 2021 Cochrane meta-analysis (the 🥇standard) concluded that HCQ “has little or no effect on the risk of death and probably no effect on progression to mechanical ventilation. Adverse events are tripled compared to placebo…”

cochranelibrary.com/content?templa…
A more recent meta-analysis in @NatureComms that included unpublished studies went further, concluding “that treatment with hydroxychloroquine is associated with increased mortality in COVID-19 patients”

nature.com/articles/s4146…
Read 9 tweets
Jan 20
Just 3️⃣simple acts of compassion can go a long way

Important RCT just published in @TheLancet shows that a family support strategy consisting of 3 extra family meetings can substantially reduce family grief 6mo after the death of a loved one in the ICU
thelancet.com/journals/lance…
1/
Pre-pandemic, ICU mortality was often ~20%

In the US, 30-60% of ICU deaths were preceded by a decision to withdraw life support

We are experts at providing comfort focused care for patients but in many cases, families feel grief months afterwards

What can we do about this?
2/ ImageImage
The COSMIC RCT studied this.

The intervention was 3 family meetings held following a decision to withdraw life support:

1️⃣family conference to prepare relatives for the imminent death
2️⃣ICU-room visit to provide support
3️⃣meeting after death to offer condolences & closure

3/ ImageImage
Read 7 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!

:(