The End of the "Syndrome" in Critical Care

From the book introduction showing how the RCT was corrupted inadvertently.

“The machine the rebels oppose is a massive infrastructure, its main tool is a science based on “Pathological Consensus”, a dangerous but seductive pathologic version of science, which masquerades as real science. Pathological Consensus emerged, as if out of the ether, in the 1960s. Its advocates soon dominated the landscape offering easy research and the promise of a simple path to the discovery of a one-size-fits-all protocols for large groups of diseases which they lumped together as Synthetic Syndromes.

In the 1960s, these machine creators had visions of massive breakthrough cures for large groupings of similar appearing diseases. No one knew it but the machine’s creators had subtly hijacked and corrupted the breakthrough method of the randomized controlled trial (RCT) which had been introduced, two decades earlier, by Austin Hall based on the work of the polymath genius Ronald Fischer. The creators inserted two new techniques into the mathematical DNA of the RCT method, so that a single trial could be used to test a single treatment for a large group of different, but similar appearing, diseases simultaneously

These two techniques were like hidden viruses destroying the normal function of the randomized controlled trial but opening the performance of these trials up to the masses of brilliant academics eager to perform them.

2 Likes

Hi Lawrence

Congrats on the book!

The key message I took from it: Progress in any scientific field is like building a Jenga tower. If you put rotten blocks at the bottom, you’ll never get it off the ground.

True experts take the time to critically interrogate the foundational work in their field. Researchers who ignore this history risk spending their careers building towers that collapse whenever a second storey is added.

Stagnancy in a field can signal that the problems at hand are wickedly complex. But it can also signal that leaders have ignored a rotten foundation.

2 Likes

Closely related to this is a presentation from @Sander from last year:

There is not much “Science” in “Science”.

@Lawrence_Lynn: have you considered that your arguments could be applied to many areas of medical research? An accounting of actual medical fact vs pathological consensus, is overdue.

1 Like

The world is run by people who gave faith that they know exactly what is going on”

“Illusions of skill are supported by a powerful professional culture”

People can sustain an unshakable level of faith in any proposition however absurd when they are sustained by a community of like minded individuals”

Near quotes from his talk. Excellent.

To which I would add that the The embellishment afforded by statistics derived from pathological consensus provides an impenetrable bubble. That’s why going deep and finding the apical errors in the spplication of Hills method is pivotal.

Could not have said that better. That’s why Wood and others operated as science police. There is no backup. Loss of science police and consequences for getting it wrong produces the present state in critical care.

No science police (self policing) results in a sustained pathological science which becomes pathological consensus, But who are the science police when everyone is a community seeking the favor of each other?

The question is how do you get the young to be science police. I was taught at Washington Univerity Stl by a member of the “ Antidogmatic Society”. I think the culture has changed to the “Hiw do I get a grant society”. There are no science police in that culture.

One more thing Erin.

My view is that we talk endlessly about these things but that’s not productive except perhaps as a catalyst for the young. My goal is more ambitious. I hope to induce or precipitate the deep debate and dialog required in critical care science. This was the reason for the book coauthored with @Rafael_Leite and David Lynn.

So far the leaders will not debate at a deep level. This is the reason we wrote the book for the lay public and the non critical care scientists because lumping scientists are not going to read it. The term “lumping scientists” is not disparaging, this is fundamental to the method by which they modify Hill’s technique.

There is always the theoretical “common driver” argument, which was the basis for the lumping in the 1960s, but a common driver of which diseases in which mix of the diseases under test?

Are not RCTs hard enough when diligently following the single disease model of Hill?

Yet no one will talk about this. It seems that the option of moving from the synthetic syndrome lumping model and going to the single disease model and then combining the diseases later for which a common response to treatment is identified, should, at least, be debated.

We should, at least debate whether the Petty/Bone RCT shortcut (lumping by threshold) modification of Hill’s method is valid and if so what are the parameters rendering the Petty/Bone RCT method valid? To my knowledge no statistician ever investigated the Petty/Bone RCT method Petty/Bone was presented to them as a single disease equivalent (I.e. Hill’s method) . In other words statisticians think they are using the RCT method of Hill and Fischer not Petty& Bone.

Yet the standard in critical care is the Petty/Bone RCT method. No one questions it. Perhaps no one even thinks to question this decades old standard method. Furthermore, I don’t think any of the statisticians are aware.

This is not something that the syndrome science thought leaders can consider or perhaps even comprehend. They have been indoctrinated in the Petty/Bone RCT method and they think randomization allows the modification.

So action by leaders of the overarching philosophical and mathematical dimensions of science are required at a deeper thinking level than the thought leaders are capable of, within the constraints of their bias in their own clinical RCT realm. Specifically this requires those from outside the discipline at a higher academic level that the “realm leaders” cannot ignore.

Critical care syndrome science will aggressively debate vaccines, masking, the need for a 30cc/kg mandate for sepsis and almost anything. It goes on in twitter all the time. However they will not debate, for one minute, the “science of their own science”. That, they will not discuss. You will never see them here and I have often invited them.

So they cannot do it. They are too weak as a function of their bias. This also is not disparaging. Who is not intellectually weakened by their own bias? Therefore, the appeal now is to the public and to the scientists and philosophers of science itself. They are the only hope to get introspection and debate which is required to maintain the validity of any science.

Open debate and open minds are the two ingredients required to produce the almost magical self-correction feature which is characteristic of real science. We need that now for the world’s health.

1 Like

[quote] have you considered that your arguments could be applied to many areas of medical research? An accounting of actual medical fact vs pathological consensus, is overdue.”

Robert what other science groups are similarly affected? in the book we present this as a focal problem. Since it’s a book for the lay public we did not want to paint science with a broadly negative brush and much of medical science seems fairly solid.

If pathological consensus is present I do believe that going back into the history to find the apical error is the best method of addressing it. L

Minor typo: Fischer → Fisher

1 Like

Unfortunately, the critical care community is not prepared nor willing to pivot away from “lumped by threshold” syndromes like ARDS and sepsis.

After decades of replacing disease definitions by arbitrary thresholds, critical care has normalized the mistake and enthroned it as a dogma. As I said elsewhere, try to convince a physician that pneumonia is not pneumonia until the patient is hypoxic or crosses any other arbitrary threshold.

Threshold definition are not disease definitions. They won’t generate a treatment because they don’t point to a treatable cause.

Notably, it is easy to see what is wrong in this distorted form of science. Nevertheless, the field is paralyzed, testing interventions without a disease model.

In the book, we discuss how the field stuck in such mistakes. To begin with, consensus threshold definitions are the NIH standard for funding. Instead of sparking innovation, the agency now acts to prevent innovation in the field of critical care.

Moreover, there is a move that may run unnoticed in America. American influence was the vehicle to spread the apical mistake made by Petty and Bone everywhere in the world. There is no place in the world for questioning the dogma. We described it as a form of intellectual colonization.

We shouldn’t expect critical care researchers to change the course anywhere in the world. They will keep publishing and receiving grants inside the dogma.

Critical care is trapped in this conundrum and won’t fix itself from inside out. We bet on data scientists, scientists from other areas, and the lay public to call for change.

I invite you to find a expanded discussion in this link for the book

https://a.co/d/i8Zp32F

The Physician’s War : The Story of the Hidden Battle between Physicians and a Science Based on Pathological Consensus

1 Like

This study show exactly what the use of Petty/ Bone RCT method would produce. 1 out of 16 single center RCT were reproducible (that one is probably by chance).

Note also the treatment effect declined with multi center exactly as predicted by the Petty/Bone method due to increased dilution with different diseases which do not have the driver.

“ We identified 19 sRCTs reporting a significant mortality reduction in adult critically ill patients. For 16 sRCTs, we identified at least one subsequent mRCT (24 trials in total), while the interventions from three sRCTs have not yet been addressed in a subsequent mRCT. Only one out of 16 sRCTs (6%) was followed by a mRCT replicating a significant mortality reduction; 14 (88%) were followed by mRCTs with no mortality difference. The positive finding of one sRCT (6%) on intensive glycemic control was contradicted by a subsequent mRCT showing a significant mortality increase. Of the 14 sRCTs referenced at least once in international guidelines, six (43%) have since been either removed or suggested against in the most recent versions of relevant guidelines.

Conclusion

Mortality reduction shown by sRCTs is typically not replicated by mRCTs. The findings of sRCTs should be considered hypothesis-generating and should not contribute to guidelines”

I don’t know why failed methodologies should be used to generate hypotheses but they just can’t see that the Petty/Bone method is the problem.

Some of that arises from the general problem of (1) using p-values and (2) not assessing evidence for clinical significance. If RCTs used Bayesian \Pr(\text{efficacy} > \text{trivial}) things would be different.

I understand that is true for testing treatment of a single disease, especially based on your teachings. However, if the problem is the lumping of, for example, 30 different diseases under a synthetic syndrome by use of a set of consensus thresholds, how are the priors to be determined and would not the priors change with each new mix of diseases captured in each trial.

Is it really possible to do trials testing a treatment for a variable set of diseases captured by a set of thresholds (the Petty/Bone RCT method)?

We have no evidence that the multi center RCT are correct for these synthetic syndromes either. They are simply perceived as the standard, however they are virtually always negative (we would say due to dilution of any driver being targeted by the treatment.)”.

Is there any evidence that the technique of Fisher/Hill can be modified to apply to a variable set of different diseases captured by a consensus set of thresholds (The Petty/Bone shortcut)? Shouldn’t we have the statisticians provide a position paper on whether the Petty/Bone shortcut, which was made by clinicians, is valid mathematically.

We saw what happened when COViD pneumonia entered the Mix of ARDS. The clinicians then excluded COVID pneumonia and called their old syndrome Non-COVID ARDS but what other outliers are hidden in the mix of NonCOVID ARDS? . How would the priors be adjusted for the mix? How would the adjustments be made with the next new virus?

The alternative seems to be to go back to the mathematically sound single disease methodology with frequentist or Bayesian approach. Is that true or is there still a path for testing treatment of diseases groups lumped by general nonspecific (non PCR, etc.) thresholds.?.

2 Likes

~50 years later they question the counting science of sleep apnea. (The AHI). How many hundreds perhaps a thousand RCTs were done with this simple counting.

Fake science. They act like it’s a revelation. I told them the AHI would not work in the 1990s. We should be ashamed for allowing our science to be fake.

image

Physiologic Consequences of Upper Airway Obstruction in Sleep Apnea - CHEST https://journal.chestnet.org/article/S0012-3692(24)00708-6/fulltext

Abstract

“OSA is diagnosed and managed by a metric called the apnea-hypopnea index (AHI). The AHI quantifies the number of respiratory events (apnea or hypopnea), disregarding important information on the characteristics and physiologic consequences of respiratory events, including degrees of ventilatory deficit and associated hypoxemia, cardiac autonomic response, and cortical activity. The oversimplification of the disorder by the AHI is considered one of the reasons for divergent findings on the associations of OSA and cardiovascular disease (CVD) in observational and randomized controlled trial studies. Prospective observational cohort studies have demonstrated strong associations of OSA with several cardiovascular diseases, and randomized controlled trials of CPAP intervention have not been able to detect a benefit of CPAP to reduce the risk of CVD. Over the last several years, novel methodologies have been proposed to better quantify the magnitude of OSA-related breathing disturbance and its physiologic consequences. As a result, stronger associations with cardiovascular and neurocognitive outcomes have been observed. In this review, we focus on the methods that capture polysomnographic heterogeneity of OSA.

1 Like

Interesting discussion to read through (or at least skim!). I have little experience with RCTs but frequently use EMR data for my work. One of my PhD supervisors was a critical care specialist and around 4 years ago advised me not to pay too much attention to syndromes in the data because they “aren’t real.”

The intention at the time was to develop some diagnostic models for different root causes and try to link them together using model averaging. However, the only up-to-date diagnostic information we had for most patients was syndrome-based rather than ICD-based, and we decided to abandon the idea.
Incidentally, I did see this was published several months ago: https://pubmed.ncbi.nlm.nih.gov/38687499/

3 Likes

Perhaps these are reasons for taking a different approach when a device like CPAP is developed; first show that quality of life and functional status are improved by the device before attempting to find evidence for reduction of risk of clinical events. Benefits of CPAP for quality and function are huge.

3 Likes

Absolutely agree. I remember when the first CPAP devices were wooden boxes. I have been taking care of sleep apnea patients since the 80s. Counted apneas with a thumb counter.

So sad to see a field of medical science with so much potential be so oversimplified.

An amazingly intriguing example of 40 years of managed science, i.e. top down medical science by pathological consensus

1 Like

Your mentor saved you by telling you the truth. So many mentees have been led down the pathological consensus path to produce nothing but failure. You should thank your mentor.

2 Likes