EBM, Guidelines – intended and unintended consequences


Category : Medical Rants

Who can argue against evidence based medicine? Who can argue with using evidence to develop guidelines? The key to practicing great medicine must involve using the best evidence to guide our protocols.

My son, while in college, was an English major. I remember reading his papers. He often used the phrase “on further reflection”. I often recall that phrase when considering these complex issues.

Frederick Nietzsche wrote “There are not data, only interpretations.” The problem with evidence based medicine is that medical decisions involve values.

Studies give us important information. They tell us how interventions impact disease, but they also tell us the side effects of the same interventions.

Some interventions are dramatic with minimal side effects. Other interventions modify the course of disease in less dramatic fashion and have greater side effects.

In the second situation we have to balance the positives and the negatives. Are the benefits worth the risks?

To make these decisions we have to implicitly assign values to the benefits and to the risks. Those values are and must be subjective. Those values are patient specific. For us to declare that everyone should receive a certain treatment implies that we can assign values for everyone.

If this problem was straightforward, then logically we could take the evidence are predictably develop guidelines.

The philosophy of logic demonstrates the inconsistency. We have too many examples of conflicting guidelines. Logically if s were a direct logically product of evidence, then differing panels should develop the same guidelines for the same clinical questions.

But differing panels develop differing guidelines. The only way to explain that phenomenon is to understand that evidence is never absolute. We must always interpret the evidence, and our interpretations involve values.

Thus too often our guidelines should not be rules. They are often not patient oriented. We cannot explain these observations otherwise.

The guideline movement is out of control. We are bombarded with long complex guidelines that address problems in a paternalistic fashion.

We need shorter, more focused guidelines. We need the honesty to provide the probability of benefits and risks. Guidelines should help us frame medical decisions for our patients. Guidelines should give us a framework. Guidelines are not, and should not be, rules.

We need to all understand that evidence requires interpretation and thus the evidence does not imply the same answer for every patient with the specific situation.

Comments (5)

I have stopped using the phrase ‘evidence based medicine’ and replaced it with ‘evidence informed medicine’ because the former suggests that the evidence is good enough to allow it to make decisions, while the latter clearly implies that the evidence (and the guidelines – which are simply the opinions of others based on their interpretations of the evidence) are one of the things that inform us during our collaborative work with patients.

This is one of my biggest hot buttons. Thanks for the good post. I plan to share it widely.

The other major issues involve application of evidence to frail, elderly or highly co-morbid patients. The risks and benefits are often unknown or even reversed for many of the most common patients we see. Is a frail 92 year old with ischemic CMP helped or harmed with ACE inhibition or high dose statin therapy? Do chronic disease guideline matter in the last 2 (or maybe 5) years of life? The patients in the single disease RCT trials are often materially different from the patients we care for in the hospital, but we often spit out trial data like robots (CHF = ACEI, CAD = statin, DM= medications) without any humility about the limits of knowledge and the potential for harm.

While there are likely some small successes with some (simple) EBM questions, the vast majority of complex patient care can only be optimized by heavy reliance clinical judgement, and by some active restraint of our desire to only use therapies “proven” in the literature.

I won’t go so far as to say evidence-based medicine was a scam, but it was overblown from the get-go. It was simply a new-fangled way to say “read the literature carefully”, something young physicians should be taught as a matter of course. The concept quickly took on a ice of its own. Now you see the parse “evidence-based” all over,as if it is the final word in any dispute.
It has been used as a political and economic gambit, which is a blatant misuse of the concept. To say that leaves you open to be called anti-science or a denier or a Luddite. Now clinical experience is looked on with suspicion, rather than analyzed for its strengths and weaknesses. Surely that is a mistake. Centuries of practice can’t be completely discarded because we have computers.
“Evidence” comes from the Latin “videre”- to see. But we all know that evidence, even eye-witness evidence” can be misleading or dispositive.
Also – just coincidence, the man who coined the term, David Sackett just died.

The other problem with many of the “guidelines” is that they were set by panels filled with people who had conflicts of interest regarding the medications they were recommending.

[…] EBM, Guidelines – intended and unintended consequences […]

Post a comment