The sepsis science monopoly
youtu.be/C75RHvh1Lpw?...
The sepsis science monopoly
youtu.be/C75RHvh1Lpw?...
I asked Grok to “Summarize REMAP CAP using two DAGs”
But Grok only looks at work I did on X and discussions at datamethods and provides a
Well organized summary. Grok is a sophisticated text review parrot. Which is nice.
discourse.datamethods.org/t/the-petty-...
Grok does fairly well. Can you provide an example of failure ?
02.06.2025 20:46 — 👍 0 🔁 0 💬 1 📌 0
showing that lumping dilutes significant effects and masks weak ones
2. Highlighting heterogeneity (disease subtype: C1) and confounding biases (severity: C2), necessitating stratification by etiology…
Now Quoting Grok 3:
“Conclusion
Performing DAGs prior to REMAP-CAP would have led researchers to recognize the error of lumping viral and bacterial pneumonia by:
1. Revealing distinct causal pathways for strong antibiotics (bacterial DAG) versus weak antivirals (viral DAG)…
Note:
Grok refers here to the standard method in critical care which lumps multiple diseases by a set of opinion based non disease specific criteria to test a single treatment
This method was used in REMAP CAP lumping viral & bacterial pneumonia & excluding pneumonia types acquired in hospital
Thread:
This is a conclusion of Grok 3 re: formal symbolic causal modeling prior to RCT.
(Referring specifically to the recently published REMAP CAP RCT testing hydrocortisone for severe community acquired pneumonia (CAP)
a nee era symbolic causal modeling as a @NIH required for grant
Did you notice that no one cited in these sepsis review articles actually discovered anything.
They just make up the rules and synthetically shift the paradigm.
Learn the history of sepsis science here.
youtu.be/BTgEU07FI9o
Read the whole linked article not the abstract. The guessed Sepsis 3 platform of failed PettyBone sepsis “science”
Time for young to lead Abandon guess of Vincent which like Bone’s SIRS guessed when they were young.
Don’t waste your career on 20th century guesses
pubmed.ncbi.nlm.nih.gov/9824069/
Young need to start afresh. In 1998, SOFA was found to be so nonspecific for sepsis they changed the first word of name from “Sepsis-Related” to “Sequential”. Did you know?
Each study they laud “heterogeneous syndrome”. A meaningless term. Can’t build science on an ambiguous & variable platform
You can’t start with PettyBone science & lumped set of different diseases triaged by Vincent’s 1996 guessed threshold set, then phenotype the improperly lumped diseases.
All critcare fellows should watch this brief video.
Study Science not the latest opinion of a task force!
youtu.be/BTgEU07FI9o
Yet they virtually always have been in PettyBone science with ARDS and CAP trials.
13000 ppl died last year due to influenza pneumonia and we have no substantive RCT to tell us if steroids help or harm because our 35 yr old standard are “PettyBone RCT mimics” not Bradford Hill RCT.
Of course not. Influenza lacks the secure safety of antibiotics to offset the potential reduced viral clearance due to corticosteroids. The DAGs are markedly different so they can’t be lumped in the first instance.
29.05.2025 17:59 — 👍 0 🔁 0 💬 1 📌 0
I have been blocked by the thought leaders because I argue they were indoctrinated in PettyBone pseudoscience (as we all were)
Really though did you think we could lump influenza pneumonia and pneumococcal pneumonia for a REMAP CAP hydrocortisone RCT?
In the link I explain why REMAP CAP would not be expected to reproduce CAPE COD. (It’s not reversion to the mean)
We clinicians have to be smarter. We fooled the statisticians into thinking our synthetic syndromes were disease equivalents in the Bradford hill sense.Time to end PettyBone science
The formal symbolic causal modeling is the spice for the pul crit care physician
The CI extension of that may be a bridge too far in our complex environment.
The synergy between this modeling exploits our knowledge of pathophysiology, used fir RCT interpretation and RCT design.
a.co/d/eaISGSp
Formal Symbolic causal modeling is the way to optimize the design of these RCT and to speak of them in more formal objective terms.
Clinicians need to up their game bringing formality to their physiology expertise interpreting these trials.
It shows why 35 years of PettyBone RCT mimics failed.
No one will discuss or retweet PettyBone RCT mimics. But your generation needs the courage to break out. Study and teach Pearl’s symbolic causal modeling
Send your pul fellows to see this alternative view explaining REMAP CAP at datamethods. 6.8k views
discourse.datamethods.org/t/the-petty-...
The discovery went from narrow (IPF) out. Not starting with lumping
IPF -> PPF
The discovery would likely have been negative & abandoned if they started with PPF because of signal dilution
ARDS & sepsis PettyBone RCT mimics use triage for participants guessed threshold sets
hubs.la/Q03p8LgD0
Critical Care Quote of the day.
“…formal causal modeling is a means to add much needed rigor to determine the DoE [design of experiment]. Had causal modeling been required to acquire the [ @NIH ] grants in the past, it might have saved the crit care science decades of wasted research & careers”
Yes I was consulted on a case of renal cell carcinoma that embolized to the Tricuspid valve where the tumor was stuck. He had a large ASD/PFO which saved him.
Significant hypoxemia unresponsive to FIO2 of 1 but not, as I recall, hypotensive.
Or bring your excellent points to Vanderbilt Datamethods forum.
discourse.datamethods.org/t/the-petty-...
The problem is more fundamental and relates to the “RCT mimic” itself. The conflation of synthetic syndromes for disease equivalents
Join discussion questioning whether or not formal symbolic causal modeling should be required in the application for an NIH grant for an RCT.
x.com/patientstorm...
As pressure from the NIH grows to generate reproducible RCT outputs in critical care, and with potential defunding of non reproducible result generating RCT.
See this brief video relating to this pressing issue relating to critical care research funding and defunding.
youtu.be/BTgEU07FI9o
The history of “lemon juice science” and whatever treatment you fancy for the lumped set of diseases which meets the latest amazing international task force generated (guessed) threshold set.
youtu.be/BTgEU07FI9o
The PettyBone RCT mimic is an endless cycle of grant generation in critical care.
This video tell the history of “RCT mimics” and “Synthetic Syndrome “ in critical care experimentation science.
youtu.be/BTgEU07FI9o
Watch this alternative view of research of ARDS, Sepsis, & CAP. Share this with others in training. Only 6% of single center RCT are reproducible & protocols based on multicenter RCT are often reversed.
This is a history of critical care science your mentors never told you.
youtu.be/BTgEU07FI9o
We have to push back. No retreat. Yet we must clean our house
Let’s bring the elephant in the room debate (see link) & reform critical science
If we are afraid of deep debate & introspection we lose the high ground with the new NIH who are discovering these failed dogma
youtu.be/BTgEU07FI9o
Sorry to hear that. Yet we need everyone to push back. No retreat.
However, we have our own house to clean. Please watch this brief video from and help us reform critical care science. Debate the defunding of the task forces with your fellows.
youtu.be/BTgEU07FI9o
In critical illness, we know patients are “sick” & we use terms like “significant inflammatory process” but we aren’t able to quantify it. Proud to be part of this work where we showed the role of neutrophil extracellular traps in organ injury #sepsis ccforum.biomedcentral.com/articles/10....
19.05.2025 16:59 — 👍 3 🔁 2 💬 0 📌 0