The Machine is Broken
I should probably begin by stating that I am not surprised R.G. took a stance largely in contradiction with my original post. After all, it’s quite an easy question to ask-what could possibly be wrong with practising medicine based on hard evidence?
While there is quite a bit of merit in some of the points R.G. makes, I still have a few qualms. These may well be a matter of nuance, but I will flesh them out anyway.
R.G., referring to EBM says, “It is old wine in a new, statistically significant bottle”. This may be true prima facie, but EBM has actually brought a deluge of conflicting information, recommendations which are vague at best, and plain confusing at worst. It is less a ‘statistically significant bottle’ and more akin to a tin can full of static.
R.G. goes on to state that doctors are human beings who make mistakes and this should be accepted as part of a medical system that includes EBM. This is correct, but one of my major problems with the EBM as an implemented system is the quality of evidence available to the average doctor. I mentioned the fact that negative studies are not routinely published. R.G. only indirectly addresses this. In a sense he shrugs it off as a chronic ailment of all industries that depend on the results of academic studies. It’s the big machine that’s broken. It is par for the course.
The fact that something is widespread does not make it acceptable.
R.G. also attacks the point made in relation to the doctor I argued with concerning word of mouth claims about the safety of ibuprofen in COVID-19 positive patients- ‘I would say that if you want to push back against EBM then you would need to have a stronger example than something based on fragile evidence’. It’s impossible to disagree with his point about levels of evidence. It is quite clear what should be accepted as strong evidence and what should not.
Maybe one solution would be to provide better education to doctors in terms of how to interpret evidence with regards to EBM. This doctor is not the first to change her way of practicing medicine based on little to no real evidence. It is quite possible that many doctors are not familiar with the levels of evidence R.G. cites in his blog. This is a serious problem.
The rebuttal provided by R.G. also fails to account for my original point pertaining to the sheer volume of information available on any one issue. While it is correct that a doctor should be expected to interpret information and not simply regurgitate conclusions provided by any given study, it is important to note that providing a deluge of information that is likely to change within a few months is not productive.
The unfortunate result of this is that physicians, when presented with large amounts of conflicting evidence, will invariably do what they want to do. Evidence be damned. This goes directly against the original purpose of EBM. Instead of becoming an aid to clinicians, what EBM has done is create a paralysis of choice. Clinicians, backed into a corner by the flood of studies and conflicting results invariably either do what they are accustomed doing, or make a decision based on another line of logic. Ironically enough, this line of logic often follows the ‘plausible mechanism’ method I outlined in my first post.
There is too much information in front of me, therefore I will decide based on what should work in this situation.
The Omega-3 fatty acid issue is a prime (but not only) example of this problem. In July 2016, this article published in Medscape news claims that omega 3 from fish and plant sources may lower the risk of fatal heart attacks.
In October of the same year, Medscape news published another article suggesting that there was no cardiovascular risk reduction associated with dietary fish.
January of 2017 produces another article, published yet again by Medscape news showing the omega 3 lowers heart disease risk, more so in hyperlipidemia (high fat levels in blood).
In May 2018, the American Heart Association recommends fish (to increase omega 3 intake) twice weekly for cardiovascular health.
Fast forward to July 2018, and Medscape publishes another article claiming that omega 3 disappoints in terms of cardiovascular protection.
It is worth noting that all of these articles were written based on what would be deemed as high quality evidence.
One might respond that consumption of Omega-3 is optional. Omega 3 isn’t harmful, and maybe one could look for other ways of bettering cardiovascular health. The lack of consensus won’t provoke too many headaches for a physician at any rate. Maybe this is an issue one can ignore.
However, it doesn’t take long to think of the wide range of everyday foods that have been linked with an increased risk of cancer. Finding out whether I need to avoid a common food because it is dangerous would be of great value. Alas, there is no consensus here either. Physicians and patients are yet again paralysed, while tonnes of information exists on this subject. One cannot help but wonder who really benefits from the conduction of all these studies.
A pattern emerges. The clinician is expected to utilize his judgement, but this is clearly not possible if ‘the best available evidence’ is about as constant as the seasons. This is yet again a problem not with the idea of EBM, but with the implementation. It is unreasonable to expect a physician to make any kind of informed decision in this type of circumstance.
R.G. accuses doctors of wanting EBM to be a crutch, but I do not think this is the case. Doctors want a system which provides them with evidence that is trustworthy. The medical profession has no problem with the IDEA of EBM. Doctors would like to be able to say “I made this decision based on x study or group of studies”, and feel confident while doing so. It’s difficult to construct a sound clinical decision when your foundations are so weak.
The machine is broken.
