Welcome to my blog, a place to explore and learn about the experience of running a psychiatric practice. I post about things that I find useful to know or think about. So, enjoy, and let me know what you think.

Saturday, November 30, 2013

What's In A Name?

Interesting article in Clinical Psychiatry News, Psychoactive Drug Nomenclature System Devised, by Mitchel Zoler.

Joseph Zohar, David J. Nutt, David J. Kupfer, Hans-Jurgen Moller, Shigeto Yamawakie, Michael Spedding, and Stephen M. Stahl have all been busy coming up with a new way to describe psychoactive medications. It's a 5 Axis system (I guess they were homesick for DSM-IV), and the plan is for it to be web-based, so it can be in a constant state of revision.

The idea behind it is that, according to the example in the article, you have a patient with an anxiety disorder, for whom you prescribe, say, celexa, which is an antidepressant. So the patient wonders why you're telling him he has one condition, and giving him a drug for another.

Here's my list of the axes-but you can view them in the article, in a nice table, complete with two examples:

Axis 1- Class and Relevant Mechanism
What neurotransmitters it affects, and what it does at the molecular level.

Axis 2- Family
This seems to overlap with Axis 1.

Axis 3- Neurobiologic Activity
Includes neurotransmitter effects, and broader physiologic effects.

Axis 4- Efficacy and Major Side Effects
What conditions it's good for, and what else it does on the experiential level.

Axis 5- Indications
Formal FDA approvals, I think.

The Pros:

The physicist, Richard Feynman, had an interesting father. He would say to young Feynman, "You see that bird over there? I can tell you its name, but first, I'll tell you what it does."

Names, categories, titles, they're all important, but they often don't give you that much information about whatever they're naming. "Sparrow" doesn't tell you anything. You need to connect that word with a certain type of bird.

So it's a good idea to describe drugs by what they are and what they purportedly do. You end up with a lot more information than you get from "SSRI" or "Prozac".

And on the flip side, "antidepressant" only tells you one thing that a particular drug does. If it does other things, you won't know it from that one word.

I quite like the idea of putting all the relevant information out there right from the start.

The Cons:

Who is this for? I can see where it would be useful for medical students studying for pharmacology exams. Or for psychiatrists, to tweak our memories, or help us clarify things to our patients, although personally I'd like an Axis 6 with Cyp450 enzyme data.

But what about non-psychiatrist prescribers. Could this be a sneaky way to promote off-label prescribing? If you look at the chart in the article, Axis 4 for amytriptyline is "anxiety;chronic pain", not "depression", which is listed in Axis 5 as "Major Depressive Disorder".

And does listing the known molecular mechanism of action, e.g "Increases synaptic 5-HT and norepinephrine" right next to the conditions the drug is used to treat, e.g. "Depression", sneakily imply that too little synaptic 5-HT causes depression, and more synaptic 5-HT is what "cures" depression? Because we just don't know if that's true.

Is this system for patients? Will it really be helpful, or more helpful than just telling these details to patients as part of discussions surrounding medication?

The question of who this system is for, as well as the plan to make the classification system web-based are particularly relevant, in light of the fact that I couldn't access the original article. Well, I could, for $35.95.

My feeling is that this idea is being presented as an ongoing collaborative model, but really, it's proprietary. And forgive me if I have concerns about where they're heading with it. This quote is from the Zoler article:

"This is the first step in a long process. We’re trying to wake up a 50-year-old nomenclature into a living document that will be used over the next decades and may someday merge with DSM [Diagnostic and Statistical Manual of Mental Disorders]," said Dr. Stephen M. Stahl, a panel member and psychiatrist affiliated with the University of California, San Diego.

What will it mean to "merge with DSM"? Will new diagnoses be created based on how medications are being prescribed? Or with what frequency they're being prescribed? After all, David Kupfer, one of the authors, was the chair of the DSM-5 task force.

And how will this affect getting FDA approval for medications? If a drug is being used off label at a high frequency, will this be a way to extend a patent? In other words, is noting the usage of a drug a way of establishing that said drug is effective for a different indication than the one it was originally approved for. 

My main concern stems from the memory of those little books I got from drug reps when I was a resident. Nice little pictures with colorful neurotransmitters happily making their way around a bunch of big receptors. All written by Stephen M. Stahl.

The final paragraph of the Zoler article:

Dr. Zohar said he has been an adviser or consultant to or received research support from eight drug companies. Dr. Nutt said he has been an adviser or consultant to eight drug companies. Dr. Stahl said he has been an adviser or consultant to more than 30 drug companies. Dr. Goodwin said he had been an adviser or consultant to 12 drug companies.

Bottom line: I think more information about medications is a good thing. But not if it's misleading. And not if there's an ulterior motive to presenting the information. Hopefully, this type of system can be used for the good of patients.

1 comment:

  1. Have not read the original article yet, but I think that the Drug Facts and Comparisons approach may beat this one. It classifies on the basis of predominant use and then looks at on and off label uses and the evidence for off label use and gives you the documentation levels for off label use ranging from 1 (good) to 5 (poor). In actual practice, my rule is that there has to be some evidence for what I do and these levels generally correlate with what I pull up on Medline. The Medline search is my go to procedure. I would also think that any study like this there days would need to correlate with Ki and receptor data:


    as well as the recent huge bioinformatics study that looked at all possible interactions with drugs at all known targets: