Evidence-based medicine

Reading an excellent article that describes the many problems with evidence-based medicine, I noticed there isn't really a discussion on the forum of the Evidence-Based Medicine (EBM) paradigm.

It's discussed indirectly pretty much all over the forum, but it certainly deserves a more directed discussion and criticism, given its excessive influence and power over our lives.

Frankly, to me, EBM is pseudoscience and a clear net negative. It's an idea that has merit, if well done, but it's never well done, and therefore is mostly harmful. It explicitly occupies the place that science should work in, and basically squats it indefinitely, until science clears it out with a chance discovery.

There are many aspects that are worth discussing, from its "pyramid" of evidence, to the issues with GRADE, the treadmill loop of research where a few academics often end up writing reviews of gradings of their own homework, and so on, and how by far its most common conclusion is that the same study should be done over again, usually pretty much identically.

Maybe this belongs more in clinical evidence. EBM isn't research, it's used instead of research to produce easy answers, and its academic loop is a disaster, stuck doing the same things over and over again. The process has become the product. There is no product other than the process.

 

Trump’s cronies aren’t what broke public health
The structural blind spots that undermine medical progress and how to fix them.
https://canadahealthwatch.ca/2025/04/18/trumps-cronies-arent-what-broke-public-health


The controversy raised a deeper question: how exactly is medical guidance made — and why does it so often resist change, even in the face of compelling evidence?
...
This exemplifies a recurring pattern of similar institutional failures, in which medical decision-making rejects outside expertise, replacing it with opaque processes biased towards inaction, that often seem strangely consistent with the personal opinions of those involved.

Just like the Task Force, the WHO relies on a simplistic framework called Evidence Based Medicine (EBM), a system designed to help front-line clinicians make quick decisions when they don’t have the time or expertise for a deep dive into the science.
...
Somewhere along the way, that distinction got lost. What began as a practical tool for non-experts was rebranded as definitive science. EBM’s dogmatic authorities claimed the right to override both science and clinical expertise, cherry-picking which evidence counts and which gets discarded.
...

...
Evidence can be down-ranked (or occasionally up-ranked) by a level or two based on evaluators’ subjective judgments, through processes such as GRADE — “Grading of Recommendations, Assessment, Development, and Evaluations.”

While often marketed to policymakers and the public as a rigorous scientific approach, even EBM adherents recognize GRADE is unavoidably prone to bias. Different evaluators produce different results.

The outcomes depend on who is selected to participate — and how much expertise they bring to the table. Different pickers, different cherries.
...
This is important, because any research can contain errors that require deep expertise to identify, and the clinical trials EBM practitioners hold up as their “gold standard” are no exception. There simply are no shortcuts to scientific expertise. You can’t find errors you don’t understand.
...
EBM often functions as a closed loop, producing studies of variable quality, marking its own homework, and rolling the results straight into guidance. Problems go unnoticed.

Basic mistakes accumulate. Inconsistent data get averaged into statistical noise. And no intervention — be it masking or breast cancer screening — ends up looking like it works.

The result is a celebration of do-nothing medicine: smug commentary on “overdiagnosis” and “overtreatment,” that never stops to consider whether the failure lies in the intervention — or the analysis.
...
Medical chauvinism has left public health leaders decades out of date on critical topics, and getting it right often seems to matter less than avoiding the admission that they got it wrong the first time.

 

Is this a thread for discussing the concept of EBM or the current implementation of EBM in various countries?

 

To pick one aspect, the peer review;

• just having 2/3 peer reviewers;
• usually the editor who decides who are the peer reviewers

seems like a huge risk of bias.

 

I think I’ve said this before — but I think the current model of peer review serves more to ensure ideological compliance, than to ensure quality.

Members on this forum practically find an error in data/transcription on every other biomedical paper. So what are the reviewers doing? It feels as if it is more that they read the paper and flag out sentences they don’t agree with, more than looking at actual methodological rigour and digging into the data.

 

Trying a general discussion, given how big of a role it plays in all this, and yet we rarely discuss it in itself. Anything that doesn't relate to a specific study.

Maybe it's not needed, but I thought the article provided a solid starting point, and although we discuss the specific flaws and errors with EBM as they are committed in studies, we don't really discuss it, uh, holistically.

 

I agree that it’s an important topic, but I feel like the two options I asked about are very different.

It would be like the difference between discussing the merits or democracy vs discussing how the democracy of the US etc. works today.

 

Probably would have a third discussion about if you can truly call western neoliberal states democracies ie. rule by the people…

 

There are plenty of those discussions to be had, but that would probably take us far beyond the no politics rule.

I’ll see if I can get back to this tomorrow and read the article rvallee linked.

 

My heart always sinks when I see the phrase evidence based medicine.

It is important we are clear in the distinction between evidenced based medicine, which is self evidently a good thing, and ‘Evidence Based Medicine’ which often is a profound misnomer. It often feels that the latter is a systematised way of protecting vested interests of a narrow ‘in’ group against good scientific evaluation.

 

oh yeah sorry never meant we should actually discuss that. just adding an example to your example of how one topic can lead to multiple quite different discussions

 

I read this 1992 paper that is one of the first mentions of EBM. It’s 5 pages and worth a read in the context of this discussion.

One thing that stands out to me is that they essentially just assume that humans have to be very good at evaluating evidence. Or creating evaluations of evidence that others can use.

So EBM has just moved the eminence and authority from individual people to institutions.

 

Oof. Very naive. Completely ignores human nature, or somehow places academics as being above this. Which they are very much not. In fact the professional culture demands a hefty dose of bias. Haven't seen much in the way of gaining such skills either. If anything, things might have gotten a lot worse since then.

For sure it's a good idea. Just not with humans involved, we're not compatible with this. Maybe there are other species out there capable of this. Vulcans, or whatever. But, yikes, no. Might as well be assuming that judges have good judgments simply because they are judges.

It's especially baffling considering that the idea would be of independent assessment, when it's part of academia that experts in a field are needed to evaluate other work in that field. It's very rare for people to be independent in those circumstances, and it basically ended with people mostly judging their own work, or the work of their close colleagues whom they work with. That's basically nothing like the original idea.

 

First I heard that EBM is not simply requiring that medicine be based on actual evidence (as opposed to hearsay and opinion). Hmmm…. Re-evaluating a lot of things said in the past couple of years ….

 

I feel like EBM might be the same as democracy in the sense that it’s the best bad option we’ve got.

 

I read the article as well, I liked how he explained the present crux of the matter:

(It's nitpicky -the problem as posed still stands- but I may disagree with a detail of his comment that EBM was "designed to help front-line clinicians make quick decisions when they don’t have the time or expertise for a deep dive into the science" or that it "began as a tool for non-experts". It may have evolved into this, but at quick glance, based on an elaborate 1992 paper by the EBM Working Group I have in my collection, EBM to me looks mainly like the promotion of habitually checking for publications in a computer database on a clinically-encountered topic, presented as a hyped-up Amazing New Paradigm In Medicine, at a time when both large-scale RCT publication and general computer use were still relatively novel.

From the EBM Working Group, chaired by Gordon Guyatt (1992),

"Evidence-based medicine de-emphasizes intuition, unsystematic clinical experience, and pathophysiologic rationale as sufficient grounds for clinical decision making and stresses the examination of evidence from clinical research. Evidence-based medicine requires new skills of the physician, including efficient literature searching and the application of formal rules of evidence evaluating the clinical literature.")
​

Edited to add:

That's the one.