It is often said that medicine is both an art and a science. In an imperfect world this is both inevitable and desirable. But it is extremely important that the two should not be confused with each other. In particular, because the “science” side of the equation has achieved overwhelming prestige and authority, it is often the case that that the “artistic” side seeks to claim the mantle of scientific certainty, Frequently this occurs when “experts” skate over the difference between scientific certainty and uncertainty. Often they do this by the selective use of data, in particular observational data and imperfect or cherry picked instances of non-robust randomized controlled studies that are the hallmark of real and convincing scientific certititude.
The controversy over the uses of PCI or CABG for left main coronary disease is a perfect example of this problem.
In his initial blog post on CVCT CardioBrief Bernard Gersh maintained that the scientific evidence in support of PCI was now sufficient to raise the status of PCI in scientific guidelines. But then Sanjay Kaul pointed out that by strict scientific standards the evidence was weak. The key point I want to highlight here is that in his response to Kaul’s post Gersh does not address the scientific issues raised by Kaul but instead bases his defense on “a clinical perspective.” Gersh illustrates my point:
Irrespective of the statistical issues pointed out by Dr Kaul, the point I tried to make is that there many other clinical factors that need to be taken into account when recommending a therapeutic approach to an individual patient. In my experience patients are not that interested in “interaction” terms or other statistical issues such as the p value but they do want an educated opinion.
In other words, Gersh makes no attempt to defend the scientific evidence supporting PCI. Instead he relies on the “artistic,” non-scientific perspective of experience and intuition. This is fine if he’s talking about advising patients. This is indeed the “art” of medicine. But educated opinions should not be the basis for scientific guidelines.
In the past I’ve argued that scientific guidelines should only be issued when there is an overwhelming scientific consensus. The most obvious danger is when guidelines are used to endorse policies and practices that are unhealthy. The best example I know is the case of the 80s campaign against dietary cholesterol, which may well have helped fuel the obesity and diabetes epidemics.
But, as I’ve written previously, there’s another sort of unintended consequence that is less obvious, though it may well be equally harmful. By their very nature guidelines present the illusion of successful science, the appearance that clarity and understanding has been achieved by the experts. When this happens, science is the victim.
The danger of guidelines is that we’ve declared premature victory. We’ve removed the incentives to perform new research. Any casual observer, looking at the hundreds of footnotes in a guideline, would be perfectly justified in asking why we should throw away more resources exploring a topic that is already so thoroughly well understood and uncontroversial.
The key to making new progress is to first acknowledge how much we don‘t know. Rather than trumpeting false knowledge and certainty, health organizations should instead publicize how much they still need to know. Such a move would instantly give them more durable and genuine credibility, and would also serve to educate the broader public about how science works, and why we need to perform more research.
It is perfectly possible that many LCMA patients, in consultation with their physicians, will choose PCI over CABG. It’s a reasonable decision. I suspect that I personally would make the same choice in this situation. (In my case this would be based on my belief that my long term outcome might well be better with CABG but my near term experience would almost certainly be easier, and less frightening, with PCI.)
But the doctors who guide the choice of their patients should not convey their clinical experience and personal opinion under the cloak of scientific authority and certitude. As a clinician it’s reasonable to say “this is what I would do if it were my family member.” But they should also acknowledge that the scientific evidence is far from complete, that it is perfectly possible that they could be wrong, and they should make an extra effort to present the opposing side.
Some people argue that clinical decisions should be based solely on validated RCT results, that clinical practice should be strictly based on the evidence. The problem is that there’s just not enough hard data or true science to inform decisions. The point to make here is that no one should pretend that this application of the art of medicine is actually based on hard science. Sometimes doctors need to be artists because there is no science to guide them. But the long term goal must be to expand the domain of knowledge informed by hard science.
Update (August 20):
Bernard Gersh sent the following response to this post:
“ I do not think that the art and the science of medicine are exclusive and an integral component of the “art” is to understand the strengths and limitations of randomized trial data. The next step is to take the trial data and integrate this into a comprehensive assessment of the entire patient.
In regard to the LMCA disease controversy we have 5 trials which demonstrate no statistically significant difference in terms of mortality ( the exception being all-cause mortality in EXCEL); differences in the rate of procedural MI depending upon which definition one uses and this is an issue which we have been struggling with since the CASS trial in the 1980’s.
Let me make the point again, the absolute magnitude of the differences are small but favor CABG. Furthermore, one needs to understand that these are highly selected patients among which there was clinical equipoise. To my mind, this means that there really is a role for PCI in selected patients with LMCA disease. Furthermore whether PCI ultimately gets a Class 1 or a 2A recommendation is not going to change my clinical recommendation for an individual patient. Guidelines are there to guide and have their own strengths and limitations. Moreover they are not legal statutes but represent a consensus. Having chaired and sat on several guideline committees, I understand the process.
Now to move into the area of clinical judgement for the individual patient. What is important? Age, life expectancy, comorbidities, extent and location of coronary artery disease, site of LMCA stenosis and amenability to PCI, LV function and patient preference.
Am I going to treat an 85 year old with chronic kidney disease and limited level of activity in the same way as a 50 year old diabetic who needs multiple arterial grafts? Of course not.
This is not confusing the science with the art but using the trial data we have and understanding the limitations of the data in regard to the trials themselves and their applicability to the patient who has come to me for an informed and educated opinion. This is the essence of being a physician. ”
Full Coverage of the PCI versus CABG for LMCA on CVCT Cardiobrief:
- Left Main Coronary Disease Trials: Close to Consensus by Bernard Gersh
- Left Main Coronary Disease Trials: Far from ‘Close to Consensus’ by Sanjay Kaul (with a response from Bernard Gersh)
- Don’t Confuse the Art and Science of Medicine: PCI vs CABG for Left Main Disease by Larry Husten