Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

Why We Deceive Ourselves With Explanations

I have been thinking a bit more about Walt Trachim’s comment on Answer to What is this Dangerous Treatment and How Long Did it Take to Stop Using it and my response.

2. You know me well enough, I think, to know that I’m not satisfied unless I know WHY something happens. And this is no different; I would like very much to know why this works. I’m not asking if you understand why it works. I’m just saying that I would simply like to know more. And I would like to see more studies done. If I run across anything else in literature I will share it.

The problem is the desire to have an explanation. We ask for plausible treatments. Plausible treatments come with explanations, even if these plausible treatments are very harmful. Plausible means that the explanation makes some sort of sense.

Plausible appears reasonable.

Plausible means that it might work –

if the explanation is valid.

Might work.

What if . . . ?

Plausible means having a story that some people find convincing –

just as some people are convinced by the stories of con artists.

Our patients deserve better than What if . . . ?

What is better than plausible? Probable is better than plausible.

Probable requires evidence, but not an explanation.

What matters is that it works, not that it has a convincing story.

The worst treatments have some of the best stories. The worst treatments convince people to be satisfied with inadequate evidence, because there is a good story.

Epinephrine is plausible – but don’t ask for evidence that epinephrine improves survival. We stimulate the heart back to life. After decades of use, there still is not any evidence of improved survival with a functioning brain.

If your patients do not use their brains, then epinephrine may seem attractive.

Spinal immobilization is plausible – but don’t ask for evidence that spinal immobilization protects the spinal cord. After decades of use, there still is not any evidence of decreased disability with the use of spinal immobilization. We just need to manipulate the patient’s neck and back to fit a collar and backboard.

If your patients needed someone to come along and manipulate their spines, but a chiropractor was not available, then spinal immobilization may seem attractive.

Atropine is plausible – but don’t ask for evidence that atropine actually addresses a real cause of cardiac arrest. After decades of use, there still is not any evidence of improved survival with atropine.

Nothing changed.

There was no new research.

We just acknowledged the lack of evidence for giving atropine.

Atropine sulfate reverses cholinergic-mediated decreases in heart rate and atrioventricular nodal conduction. No prospective controlled clinical trials have examined the use of atropine in asystole or bradycardic PEA cardiac arrest. Lower-level clinical studies provide conflicting evidence of the benefit of routine use of atropine in cardiac arrest.34,295,–,304 There is no evidence that atropine has detrimental effects during bradycardic or asystolic cardiac arrest. Available evidence suggests that routine use of atropine during PEA or asystole is unlikely to have a therapeutic benefit (Class IIb, LOE B). For this reason atropine has been removed from the cardiac arrest algorithm.[1]

If your patients were just too relaxed to live, then atropine may have seemed attractive.

Maybe epinephrine, or amiodarone, or lidocaine, will be the next ACLS (Advanced Cardiac Life Support) drug to be removed from the guidelines due to lack of evidence of benefit (and for lack of evidence of safety).

Image credit.

Plausible means –

A duck!
An aquatic bird that, in medieval times, was used to find out if a person was a witch or not.

This stems from the very logical idea that if a person weighs the same as a duck, then that person is made of wood (because both ducks and wood float in water). And since wood burns (just like witches) then that person must be a witch, because witches are made of wood.

Therefore, the accused person(s) would be placed on a scale next to a duck, and if they balanced, the person(s) would be burned.

So, if she weighs the same as a duck, then she’s made of wood. And therefore… A WITCH![2]


Probable means –

Approach this from the opposite direction. Prove that the bad things happened. Prove that the accused person caused the bad things to happen. If that cannot be done, drop the charges. Either way, ignore the silliness about witches (even if there is good evidence, it is irrelevant).


[1] Atropine
2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science
Part 8: Adult Advanced Cardiovascular Life Support
Part 8.2: Management of Cardiac Arrest
Interventions Not Recommended for Routine Use During Cardiac Arrest
Free Full Text from Circulation with links to Free Full Text PDF

[2] A duck!
urban dictionary

Probably an urban legend, but it is the distorted logic of plausibility, that I am interested in.



  1. On the one hand, it’s really nice to understand why something works the way it does.

    On the other hand, it doesn’t matter how much we THINK we understand something if what we’re doing flat out doesn’t work.

    The problem arises when people can’t make the distinction between the theory of how something should work and how it actually works (or doesn’t) in the real world.

  2. The problem is not in looking for explanations. If we didn’t look for explanations, we’re left with either no progress (“if if ain’t broke, don’t fix it”), or blind trial and error (“one of these meds might work, let’s do human trials on all of them!”). Plausible explanations give a way to focus research.

    The problem is that people only take the first step and make a hypothesis, i.e. plausible explanation. What SHOULD happen next is doing studies to prove or disprove the hypothesis. THAT is where we run into all the problems with surrogate endpoints, bias, etc… People need to realize that finding an explanation is PART of the scientific method, not the endpoint.

  3. Am I a bad guy simply because I like to understand why I have the ability to do those things my license says I can do? I think not. But I feel a bit like I’m getting a bum rap because I simply want a deeper understanding of why things affect our patients in the manner that they do.

    I was trained as an engineer long before I started working as an EMS provider. Understand that engineers look for solutions to problems while at the same time working on gaining understanding as to why the problem that they try to solve exists in the first place. I have always at least made an attempt to understand the reasons for as well as the consequences of why things are done in certain ways. And if something can ever be done better, I try to find that as well.

    I was brought up using the scientific method as a means of working out problems. I’m re-applying it in the work I’m doing as a graduate student. Clearly it needs to be applied to problems associated with the practice of medicine; I’ve never suggested any different. And mpatk is right about what a hypothesis is: it is the first step of finding an explanation, plausible or otherwise. And that explanation has to be rejectable, otherwise it doesn’t fall within the bounds of what a hypothesis is. You’ve made that point yourself in the body of this post.

    I ask you, does it make me a bad provider because I want to have a better understanding of the significance of the actions I may or may not take on behalf of a patient in my care? I would like to think it doesn’t. And I try hard to be well-informed because I have always believed it’s in the patient’s best interest to be that way.

    For the record, your point is valid; very much so, in fact. But it isn’t the only one out there. And all I’m doing is asking what I hope are valid questions.