Latest / The Medical Mind / Psych News Special Report: Addressing Cognitive Error in Psychiatric Practice

Transcript

Psych News Special Report: Addressing Cognitive Error in Psychiatric Practice · The Medical Mind

0:11
Hello everybody.
0:12
Welcome to Psychiatric News Special Report podcast.
0:17
I am the new host for the podcast.
0:20
My name is Dr.
0:20
Salman Aziz-Merza.
0:21
I am based out of Northern Virginia.
0:25
Do primarily child psychiatry, child adolescent psychiatry for my day job with the Nova Health Systems.
0:31
I also have a private practice working with adults.
0:34
I allow CalPsych.
0:36
I also do a lot of other things, things such as
0:40
working with the NBA through the um association mental health and wellness program.
0:47
And on the side, I've been doing a lot of content creation.
0:50
So you may know me through YouTube, TikTok, Instagram as the kick shrink where I do a lot of education and entertainment, trying to blend the two together.
1:00
I've been doing some articles for psychiatric news, great opportunity to do a lot of like the movie reviews.
1:08
Dr.
1:08
Prada has kind of stepped down from not the editor, but from the hosting the podcast.
1:14
So I've been asked to kind of step on in.
1:16
So this is a cool opportunity.
1:18
We have a wonderful first guest, Dr.
1:21
Paul Putman to talk about his recent special report on addressing cognitive error and psychiatric practice.
1:30
A little background and bio on Dr.
1:33
Putman.
1:34
Dr.
1:34
Putman earned his
1:35
BA in biology at Austin College in Sherman, and received his MD from the University of Texas Medical Branch at Galveston.
1:45
He completed his internship and residency in general psychiatry at the Medical University of South Carolina in Charleston, where he was the chief resident.
1:54
He's board certified in general psychiatry by the American Board of Psychiatry and Neurology.
1:59
He's been a practicing psychiatrist for over 30 years and has also performed phase one through four research studies
2:05
in psychopharmacology, published in peer review journals, and served as supervisor for the Austin Graduate Medical Education and UT Medical Branch Residency Program in Psychiatry
2:19
and lectured in the Psychiatry Department of Baylor College of Medicine, University of Texas Houston McGovern School of Medicine, and the Dell Medical School, University of Oc
2:29
and the University of Texas at Austin.
2:31
He also met with students at the University of Texas Medical Branch in Galveston.
2:35
and lectures for Austin College Master's of Medical Science program, as well as national, regional, and local medical meetings.
2:45
He's also the Associate Editor of the APA Open Source Journal, Psychiatric Research and Clinical Practice, contributes often to psychiatric news on topics consistent with
2:54
enhancing scholarly practice, and is currently serving a second term as a Doximity Op Med Fellow.
3:02
He has a first book, Roller Coaster, Finding and Treating Unstable Mood Disorders, was written for patients and their families in 2006.
3:09
His second book, Rational Psychopharmacology, a book of clinical skills, was published for the profession by APA Publishing in 2020, followed in 2024 by Encountering Treatment
3:21
Resistance Solutions Through Reconceptualization.
3:25
His fourth book, Thinking Again, Reducing Cognitive Errors in Psychiatric Practice, was released by APA
3:32
publishing in October of 2025.
3:34
So very well established with lot of books.
3:37
He's a distinguished life fellow of the APA and a fellow and former Laughlin fellow of the American College of Psychiatrists.
3:44
And he currently serves on the editorial board of the Psychiatrists in Practice Examination.
3:51
All right, Dr.
3:52
Pubman, welcome.
3:52
Did I leave anything off or we got everything mostly there?
3:55
no, that's probably more than you needed to say, but thank you very much.
4:00
So thank you, thank you for spending some time with us.
4:03
And writing this article, think it's kind of a distillation of your new book a little bit, A short of kind of summary and short of your book that's there, correct?
4:14
Yes, absolutely.
4:15
It's a synopsis and an introduction uh to topics that I, of course, go into greater depth in the latest book.
4:25
Nice.
4:27
It was interesting.
4:27
was reading the article and I found it, it struck me because the things that I liked about it was that it causes us to reflect on ourselves as psychiatrists.
4:42
You know, I'm primarily a clinician, right?
4:46
Then we do all the other stuff on the side.
4:47
We have, you know, 40, 50 patient facing hours per week between, um,
4:54
main job and then my private practice.
4:56
I'm relatively earlier in my career, right?
4:59
Seven, eight, eight plus years out of training.
5:03
But still like, I really enjoyed the idea of having to look within, having to look like at ourselves and what we're doing.
5:13
And then what the effects are of that on the field and on our patients as a whole primarily, right?
5:22
You've put your finger right on it.
5:24
uh Reflection is one of the main skills and tools that we have to protect ourselves from the very human cognitive errors that we all make over and over again.
5:36
In psychiatry training, of course, we learn transference, counter-transference.
5:44
We learn how to look at our emotions, and we learn how to...
5:48
try to uh temper the effect that that might have on any treatment interactions that we have.
5:54
But even though as scientists, we are very, very aware of how the human brain works and all the errors that it can make, unfortunately as a specialty, we actually focus on
6:07
cognitive error, probably less than other specialties.
6:11
So the important thing is broadening our reflection.
6:16
and not just looking at our motivations, but also being more aware of how our own brains work, not just our patients, and using that information ongoing every day throughout the
6:30
day while we're interacting with patients and treatment planning.
6:33
And as you said so well, reflection is one of the most important steps that unfortunately gets left out of everyday practice, but...
6:44
is one of the most important tools for reducing cognitive error.
6:50
Yeah.
6:52
It's something that, know, especially like I've started to work past few years working, we're in the process of getting a psychiatry residency over started next year at our, at
7:05
my, at at Inova.
7:07
And I've been working in the past few years with the pediatrics residents.
7:11
So as I've been doing a lot more teaching as, as a part of their GME requirements have changed and they're supposed to be getting
7:19
a lot more mental health experience because as we know, 75 to 80 % of psychiatric meds in America are prescribed by non-psychiatrists so that we have to have that other training
7:30
that's out there.
7:31
So they rotate with me and we spend four weeks of their intern year is spent with me as their training.
7:39
It's something that, again, I've had to reflect and we discuss more and...
7:43
And, know, I had an EMR change in my private practice as well.
7:46
So seeing some of like these old notes that I was writing, you know, when I was coming out of training and then comparing them to what I'm doing now, it's like, these are very
7:55
different, right?
7:56
These are totally different over the past six, seven, eight years that it's been.
8:01
Well, again, you're pointing out some very valuable steps to take.
8:05
uh Collaboration.
8:07
You we know that when we teach and we lecture and we write and we interact with others, we are looking at ourselves more carefully.
8:15
We are examining how we're thinking, what we're saying and what our conclusions are.
8:20
And we're also open to feedback from others.
8:23
We know that the more of us that practice in isolation, we tend to request second opinions less often.
8:31
We tend to question ourselves less often.
8:34
So increasing the opportunities for feedback and for interaction with our colleagues and students and para professionals is an excellent, excellent way for reducing the errors
8:47
that will impact our patient outcomes.
8:52
Tell me a little bit about, so for your article, what was the push, and then your book as well, right?
9:00
What was the push to write on cognitive error versus what a lot of, I think, critics point out with psychiatry is the knowledge gaps uh that may exist in the field.
9:12
Well, you know, I'm entering the sunset of my career and uh several years ago I decided I wanted to write a book about practicing uh and the best principles.
9:25
And that's what led to rational psychopharmacology.
9:27
And uh gratefully Laura Roberts and APA Publishing were so kind as to accept my suggestion and support me in doing that.
9:36
And then uh after that came out, Laura was saying, well, you know, these are great ideas.
9:42
Do you have examples of how they actually impacted patient care?
9:46
And I said, yeah, I do.
9:48
And so I gathered those from my practice and those of my colleagues.
9:52
And so then my next book uh for APA about treatment resistance was born.
9:59
uh as I wrote that, I began to think, well, you we're talking about cognitive error here and misconception.
10:09
And uh we need to understand more how we form concepts and how our brain goes on autopilot and does things we're not intending ah and things that are not helpful for clinical
10:19
reasoning.
10:20
uh And so I thought, well, I've just now got to write a book about the cognitive processes more specifically.
10:27
So it evolved.
10:28
uh And I was going to call this book uh in this article, Metacognition, but Laura begged me not to, I think wisely.
10:39
Because, you know, when you say that word, people's eyes just glaze over and you've lost them, you know, but, all it means is that you're thinking about how you think you
10:49
consciously look at what is the, uh, the paradigm that I'm using, uh, to do cognitive processing.
10:57
What is the clinical reasoning method that I'm using right now?
11:01
So that you're not on autopilot, but that you're aware that, all of them have some shortcomings.
11:08
and you wanna make sure that you're matching your cognitive reasoning efforts to the problem in front of you in the best possible way.
11:17
So that's how this all evolved and it just seemed like a natural outgrowth of starting with doing the very best you can in practice, you've got to include these cognitive steps.
11:30
Yeah.
11:32
And psychiatry as a field is so unique from the other medical specialties.
11:40
And it's again, the thing that sets it apart in regards to we don't have the blood tests, we don't have the imaging all the time, some cases, but not all the time that point to a
11:50
diagnosis, you know, and especially when I'm working with the P's residents, I was like, we have to think, we have to think more abstractly and to kind of
11:59
come to derive a diagnosis, integrating all the information that we have and then coming to a diagnosis and then instituting a treatment plan where a lot of other specialties are
12:09
more algorithmic.
12:13
And that's, you know, separates it.
12:16
What else about psychiatry or other other, is that part of what makes it a little harder for us to critique ourselves about it?
12:25
Cause we have the special psychiatric powers that come with our pipes and our cigars and the couches that we have in our office that make us aside from our psychiatric special
12:35
powers.
12:37
Well, we're a conceptual specialty, as you said.
12:40
We're not a hard data specialty.
12:43
Yes, of course, there is a database, and we all know that, and we all use it, and we all know the neuro-vegetative symptoms, and we all know the diagnostic validity that we've
12:53
been able to achieve.
12:55
But we're really still a conceptual specialty.
12:58
So we are dependent upon the natural processes that our brain uses
13:05
to create concepts.
13:07
And those are rife for error.
13:11
uh Our brains evolve to keep us alive and to pass on our genes.
13:17
They did not evolve to help us do complex clinical reasoning.
13:22
In fact, our brain pushes us in the other direction.
13:25
It fights it.
13:26
It wants to do the minimum amount of cognitive effort that we have to do to solve any problem.
13:32
And so we use uh
13:35
inborn mechanisms that are shortcuts and estimation rather than actual calculation and certainly not taking advantage of the broad experience of others that in this day and age
13:47
we have available to us uh through statistics.
13:52
Our brain doesn't get statistics at all.
13:54
We have to force it on the brain consciously and cognitively all the time in order to be able to take advantage of the scientific method.
14:03
So
14:05
psychiatry is a double-edged sword.
14:07
Yes, we have to be conceptual in order to conceive of what we're saying are problems and conceiving of how we can help people with problems.
14:17
That being said, the tools that our brains give us to do that are quite flawed.
14:24
Yeah.
14:25
And you brought up, or you talked about type one, type two errors, uh processing errors that are there and biases that was there.
14:34
it was, I was kind of reflecting and then looking back, I was like, even in our first kind of emails, when we were setting this up, like I called you Dr.
14:42
Putnam instead of Putman.
14:45
And I was like, that's a type one error because my brain went to the more common, more commonly kind of come across Putnam.
14:52
So I was like,
14:53
I've proven what you've discussed already, even our first interaction.
14:58
Yeah, fast and slow thinking is another way to say that the type one and type two processing has been very popularized from uh Tversky and Hanneman.
15:09
Hanneman wrote the book, Fast and Slow Thinking, and they did most of the work.
15:15
It basically, as you said, is there's an innate built-in structural mechanism for estimating, and we use heuristics, which are rules of thumb.
15:27
We estimate based entirely on our own individual experience, not on anyone else's experiences or database.
15:36
And that's a key thing to understand.
15:38
Whenever you're allowing the uh natural estimation to take over, you are only going on your own experience.
15:47
And you just have such a limited database, you're going to make errors over and over again about that.
15:53
And then the type two is more culturally evolved.
15:57
It's the scientific method.
15:59
It uses logic.
16:00
It uses probability.
16:02
uh It uses other people's data sets.
16:05
And of course, that's where we think we are all the time.
16:09
uh But most of us, even the brightest and best trained of us, naturally default to the simpler method.
16:18
And all of us spend most of our time there.
16:22
It's very tiring and exhausting to keep yourself in the type two processing.
16:27
Yet, that is something you need to constantly be asking yourself is which method am I using here?
16:33
When I come up with a rapid assessment and it's not an emergency situation, was I really using my full type two uh list of tools?
16:46
ah Or was I really just going to the old immediate
16:51
I've always done it this way.
16:53
I'm gonna do it this way this time, automatic, less conscious response.
16:59
Yeah.
17:00
Which I see, you know, I hear that all the time of this, like the worst, worst excuse or the worst reason for something is because this is the way we've always done it.
17:09
Right.
17:09
You know, this leads to all kinds of errors.
17:11
was like, because that's how we always do it.
17:13
And this turns into things like lack of progress, right.
17:18
Um, inability to move beyond certain harmful traps that we fall into.
17:23
then, yeah, it's, it's, it doesn't advance the field anymore.
17:27
If we just stick with our
17:29
how we always used to do it.
17:32
Yeah, we want to change memes.
17:33
We want to be using and demonstrating better cognitive strategies that others can witness and others can copy.
17:43
Because that's part of the part.
17:45
Part one is we copy each other all the time.
17:48
So if we're going to be copying each other, we want to be copying the good things.
17:51
We want to be promoting the good things for our patients and for our colleagues.
17:59
Are there specific kind of biases that you see that show up maybe more so in psychiatry maybe than other fields or in other parts of life?
18:08
Well, uh there's a long list and there's not an official list, but there is a long list.
18:16
And I would say for the purposes of today, overconfidence is probably one of the greatest problems that all physicians, including psychiatrists have, is we are bright, we are well
18:29
trained, we are experienced, we do know a lot about what we're talking about.
18:34
And so we tend as humans to
18:38
overvalue our knowledge, to overvalue our experience, and to overvalue our conclusions.
18:46
uh And as a result, we don't, one of the things we do is we get longer and longer in our careers is we become really, really attuned to positive feedback, but we really don't like
19:00
to listen to negative feedback.
19:02
ah And as I said before, we isolate ourselves more, there's less opportunity for feedback.
19:08
And so our overconfidence is going to lead us into treatment failure or even suboptimal outcomes more and more as our career progresses.
19:21
Yeah.
19:23
Another thing that I talk a lot about and kind of advocate a lot for is against almost to the way is big healthcare.
19:34
And we're seeing a lot of kind of like private equity moving in a lot of practices being forced to do quantity over quality of care and reducing the visits, you know, focusing on
19:49
documentation.
19:52
How do you feel that impacts all of this?
19:57
know, meaning things like speed, productivity, get your RVUs in, all of that.
20:03
How does that affect psychiatrists and cognitive error seeping in?
20:09
many ways, I'll point to at least two.
20:11
uh One is, of course, the amount of time you have available.
20:16
One of the mistakes that all physicians, including psychiatrists, make is that we diagnose way too quickly.
20:22
And the second you make a diagnosis, we know that our brain turns off.
20:28
Absolutely.
20:29
We quit trying.
20:30
We don't keep analyzing.
20:32
It just shuts off.
20:33
We have a problem.
20:34
And then we immediately start cherry picking data to support our conclusion.
20:39
This is what all physicians do.
20:40
ah And so the more you can make yourself create a differential diagnosis every single time, even when you work overconfidently, pretty sure you know what the answer is.
20:54
If you can force yourself to work through a rational differential, that's going to slow things down a bit, but ultimately it's going to give you the answer quicker.
21:06
with fewer failures and fewer suboptimal outcomes, which I think will be more satisfying not only to you and your patients, but to the system as well.
21:16
And the other thing I'd point out in these systems is you said documentation.
21:21
With the electronic medical records, there are many wonderful features of them, but unfortunately, one of them is it's so easy to copy and paste data.
21:31
And that is really a mistake.
21:34
What you want to do every time
21:36
is do a complete semi-structured interview, medical and psychiatric, and record all of the data at every visit, freshly.
21:46
You don't want to just copy.
21:48
Because we know that our memory does not remember anything, basically.
21:53
Our memories are recreations, and they are distorted by current exigencies.
21:59
They are not actual hardcopies of previous data.
22:05
So we are not remembering correctly.
22:07
If we don't record key data that can help us when we get to a treatment impasse and go back and say, yeah, I forgot that symptom was there, uh then we're not gonna have very
22:19
good outcomes.
22:20
So one is taking the time to do a differential and not do an immediate diagnosis.
22:27
And the other is making yourself record fresh data every single time and not just leaning on the electronic method.
22:35
Yeah.
22:36
think one of the keys that you brought up there was like just time, right?
22:42
You know, just having more time, spending more time.
22:45
And then the quality of your work is hopefully going to kind of improve with that.
22:51
You know, again, it's, one of the biggest feedbacks that I get from the pediatric residents where they're being forced to do 20 minute visits, 20 minute evaluations.
23:00
And then they're sitting with me they're getting, you know, 30 minute followups, 90 minute
23:04
evaluations and it's like, just the, they're like, my God, I've never like seen evaluations done like this before where there's, you know, those 90 minutes become almost
23:15
like two hours at times with we were spilling over where we're really getting into anything and everything and putting it all together.
23:23
And it's like, you know, like when somebody, and I think you kind of show this with their case example, you talked about how, which is something I really identified with was, you
23:33
know, uh
23:34
young woman who's gone through her whole life being told she has like anxiety and depression and you know, we're assuming kind of like these quickish kind of visits and
23:46
she's getting treated with antidepressants and then, you know, the antidepressants never work.
23:51
And then it becomes things like maybe she's got a mood disorder or something like a other mood disorder, or maybe she's got bipolar disorder, or maybe she's got borderline
23:59
personality and then she's
24:01
She gets on to anesthetics and mood stabilizers and nothing ever works.
24:06
And then somebody sits down, spends some time and is like, think you've had ADHD all along.
24:12
And again, I see this in my clinic all the time where it's like, they come and they message you a week later and there's been like, where has this been my whole life?
24:19
This has made a difference in more of a difference in one week versus a decade plus of failed treatment.
24:28
Yeah, we, we tend to overemphasize the chief complaint, which is of course a very important thing to begin with, but you've got to go beyond it.
24:38
You've got to do a thorough evaluation.
24:41
Most of our errors and suboptimal outcomes come from insufficient data.
24:45
Um, data that we're not aware we don't have.
24:49
Uh, so all of our efforts need to be focusing on making sure that we have complete assessments and a semi-structured interview isn't
24:57
excellent way to do that.
24:59
In this case that you mentioned, had the semi-structured interview been applied sooner, then most likely the correct diagnosis would have been found sooner.
25:09
It's comorbidities uh go undetected, medical and psychiatric, all the time if you're not doing an adequately broad interview each visit.
25:21
Yeah.
25:22
And, you know, it's, talked about the EMRs and medical records.
25:26
It's, you know, the, one of the main reasons, you know, that I've spent so much time, like optimizing my intake template to ask anything and everything.
25:34
And no matter what it was, like, you just go through everything and, you know, you take your detours during the evaluation to kind of get to where we need to go through, flush
25:45
through whatever else needs to be flushed out.
25:47
But there's so many times, especially again, in my work where
25:51
Again, kids are coming for anxiety, teens are coming for anxiety and it's like, Oh, it's because you don't understand school.
25:57
It's because you're struggling in school.
26:00
And that could be potentially like undiagnosed autism, ADHD, learning disorders, et cetera, that are there.
26:06
That's so many times again, we've, again, I've lost count hundreds, thousands of patients potentially who have been like gone down that antidepressant and as I kind of, kind of
26:14
cascade without being like, it should have been a stimulant on a lung and vice versa.
26:19
We see this plenty of times other ways too.
26:23
okay.
26:25
Also talked about abductive reasoning and very different type inductive reasoning and all these kinds of reasoning types.
26:32
Can you walk us through that a little bit?
26:34
Right.
26:35
Traditionally, it's been thought that the uh hypothetical deductive reasoning is what all clinicians have been using, uh where you start with a theory and then you make a
26:47
prediction based on it, and then you test the prediction.
26:51
uh Unfortunately, you need to make a lot of assumptions in order to do that, and many of those assumptions may be invalid.
27:00
The other method that's been talked about a lot is inductive reasoning.
27:04
uh where you create a rule from general observations and you're implying a causal link between the rule and the observations.
27:15
And unfortunately, you've got to make assumptions to create those linkages as well, and they may also be untrue.
27:22
So uh both of those methods are very, uh one is top down, the other is bottom up, and they're very rigid.
27:32
and they can never absolutely prove that your reasoning is correct.
27:38
The much better method that works very well for psychiatry and in fact all of medicine is as you said, abductive reasoning, where basically you make an observation, you make a
27:51
hypothesis, and then you test it, and then you gather the data, and then you go back and you change your original model.
28:00
And that's what's missing so often.
28:02
ah So often we'll start somebody on what we think is a treatment for major depression and the first antidepressant doesn't work and the second antidepressant doesn't work and the
28:13
third antidepressant doesn't work.
28:15
And instead of going back and saying, maybe this isn't major depression, let's look at the outcome.
28:21
Let's look at the symptoms again.
28:22
Let's go back and look at my original assessment.
28:25
Let's see if I was right about that.
28:27
ah We just keep with blinders.
28:30
This is major depression and we got to treat it.
28:33
uh So that's what abductive reasoning does for us.
28:37
It's a never ending process.
28:39
You go back, you revise your original model, because all models are wrong.
28:46
We just want to see if they're still useful.
28:48
So we go back, we revise our model, we test again, we go back with that data, and we just keep going over and over again.
28:55
Another way of saying it is creative, iterative hypothesis.
29:01
In other words, a differential diagnosis.
29:03
That's exactly what I'm talking about.
29:05
So it's very, very important to not just say, oops, that failed.
29:11
Well, it must be the treatment or the patient has treatment resistance or something else.
29:15
No, it's my model's probably wrong.
29:18
So how can I keep evolving to change my model until I get a pretty good assessment of everything that's going on with that patient?
29:27
What other diagnoses have I missed?
29:29
uh Did I, as with our case example, did I miss the primary diagnosis?
29:34
If we don't keep going back and questioning our initial assessment, we're probably not gonna work our way out of that impasse.
29:44
I use an analogy a lot when I'm working with patients and again, like in training and stuff too is if somebody comes, if you want a car garage and a car repair shop and
29:56
somebody drops off the car and says the car is not working and that's all the information that they've given you, car doesn't work.
30:04
You can do things like put gas in the tank or change the spark plugs or whatever.
30:11
And these are all potentially valid fixes for
30:14
broken car, but until we do like the diagnosing of what the actual problem is, the battery is dead or something else, right?
30:24
Like all those valid potential treatments are not going to work, right?
30:29
It's we have to apply, we have to look through and diagnose and revisit what's there and then apply the proper treatment.
30:37
then hopefully we get better treatment outcomes, right?
30:40
The car starts working again.
30:42
It keeps coming back to information gaps.
30:45
It all comes back to information that we don't have either because we misunderstood, we didn't hear, we didn't listen, we didn't ask.
30:54
All of those things happen to all of us often.
30:58
And so that's why we've got to be constantly aware that our goal in clinical reasoning is to ask better and better clinical questions to fill in information gaps that we identify
31:11
and that we can oh
31:12
that we can solve.
31:14
Yeah.
31:15
You wrote a book on that treatment resistance.
31:18
Give us uh your thoughts brieflyish, I guess, on that topic.
31:23
I'm not a fan of the word treatment resistant as a whole.
31:28
Well, that's great to hear because I'm not either.
31:32
It never occurs to me in my practice to come up with the term treatment resistance.
31:38
To me, it's like we said with that Dr.
31:40
Preseany, I just haven't gotten it right yet.
31:43
How can I keep going back and changing my analysis of the problem so that I have a different model and that gives me different information to fill and that gives me
31:53
different treatment opportunities?
31:56
uh I think that our patients are being very disturbed by the oh treatment resistance concept.
32:04
Almost 50 % of treatments right now are being labeled treatment resistant, and yet two thirds to three quarters at tertiary care centers of these cases have easily identified
32:18
new treatment options that can readily be applied.
32:22
So there just isn't data to support that there's an entity out there called treatment resistance.
32:28
And I know there are some people who like to use the term because it points them in a new treatment direction.
32:33
If they think that it's treatment resistant, they say, well then I'm going to use ketamine or I'm going to use a transcranial magnetic stimulation.
32:41
Okay, I'm not saying you can't use those things, but what I'm saying is the concept is insufficient.
32:47
It's an invalid concept.
32:49
So.
32:50
only go to those other treatments if you've gathered other data to improve your model saying, yes, those are treatments that will be valid for me, not just relying on uh
33:01
something that's not a reliable or consistent or verified, validated diagnosis in our profession.
33:09
Yeah.
33:10
It was, you know, like that struck out struck a chord with me.
33:15
Cause like, can remember a case of like this, you know, pre-teenish boy who's being staying in my office and he had been labeled by his outside provider as treatment
33:27
resistant bipolar disorder.
33:30
And I think all of us trained psychiatrists are like, well, I first of all doubt this cause a pre-teen boy should not be bipolar.
33:39
just based on statistics and everything like that.
33:41
So that was like the first kind of thing that was there.
33:43
But by the time I was seeing him, had been tried, he came to me on like five or six different medications and had previously tried like probably close to a dozen medications.
33:53
And he's sitting in the room and just like sedated out of his mind, you know, and just like lethargic and just being like, I don't know.
34:01
And he's just spitting these kind of phrases that he's been told that he has treatment with bipolar disorder.
34:10
I mean, within a few minutes, I was like, this kid has ADHD and autism.
34:15
use my own biases in a way to kind of be like, this is what's going on.
34:19
You know, and, and during the course of treatment with me, we've gotten them off these medications and got the diagnoses confirmed by neuropsych testing.
34:27
And again, just reducing the medications is down to like two medications now and thriving, doing well.
34:34
But again, like it's, it's these people, patients, unfortunately get pigeonholed with this label and then.
34:41
meds and meds meds and like you said things like ketamine etc TMS targeting thrown out them as like this is the thing that's going to save you and it's like we have to look back
34:51
and see we're probably treating the wrong thing.
34:55
Absolutely.
34:56
As we said, you created a differential.
34:58
You went back, you changed your model, and then you went back and you tested it.
35:02
And you went back and were able to prove that those alternative diagnoses were correct.
35:07
You know, when we say someone has treatment resistance, we might mean something really technical among ourselves, but can you imagine how that sounds to patients?
35:15
I mean, it just sounds hopeless.
35:17
It's like, I don't know what to do.
35:18
You don't know what to do.
35:19
oh It does not serve our patients.
35:25
at all.
35:25
really strongly recommend we not use it.
35:28
Yeah.
35:29
And I often say to like, patients is like this, it's an insurance based kind of term, I think, just to kind of get approval for things like ketamine, spravato spray, or TMS or
35:40
other things like that.
35:42
Right.
35:43
And the FDA has used it for approvals too.
35:45
And I just don't think that's the right way to proceed.
35:49
Yeah.
35:49
So how can so realistically and, and I, and I think I have luxury of time, you have luxury of time and your practice where we're able to kind of like space out appointments and
36:02
really advocate for that.
36:04
somebody who's in like, let's say a less than ideal situation, they're the only psychiatrist for hundreds of miles around and they're being booked back to back and
36:13
they're supposed to be seeing dozens and dozens of patients a day.
36:18
How can they do this?
36:20
How can they apply abductive reasoning to their work when they're just getting slammed with so much?
36:28
Well, you when I wrote the book, Rational Psychopharmacology, some people said, gosh, this is a lot of stuff to do.
36:36
How do you do this in a 15 to 30 minute med check?
36:39
And I said, believe me, I've done it most of my career.
36:41
I mean, you can do it.
36:42
You just have to know what you're going to do.
36:45
You have to plan and then you have to practice.
36:47
So what I would say to someone in that situation is go with your templates.
36:53
Make sure you have very good semi-structured interview templates.
36:56
and force yourself to work through them.
36:58
ah I mean, of course you want to give some leeway for following the patient's direction, but nevertheless, you've still got to go back and answer all these questions.
37:08
Just like we always ask about suicide, you've got to ask about a lot of other things every single time.
37:13
So ah I think if you can discipline yourself to use these templates early on in your practice, and you will get much, much better at using them more efficiently,
37:26
your patients will come to expect them.
37:29
And in fact, you know, I've had patients say, you know, you forgot to ask me about caffeine today, you know, I said, you're right, did, I lost it.
37:36
Because they like that you're being thorough.
37:38
They really feel like you're digging and you're paying attention and you're trying to get the right answer.
37:44
They will work with you, they will come in and say, you know, you always asked me about alcohol and yeah, I did have some alcohol this time.
37:52
they will bring data for you and reinforce that.
37:55
So you work together on that.
37:57
And it helps a lot with the therapeutic alliance and it helps a great deal with filling in your database.
38:03
And then when you've got a really good database, force yourself to make a differential.
38:08
And I think those are all things that can be done in time sensitive environments.
38:14
Of course, you always want to be assessing for safety and urgency, uh but most of the time, even in a time
38:21
sensitive environment, you don't have to have a diagnosis after the first session.
38:25
And if you don't have to, please don't, because we don't want that brain turning off.
38:30
We want you to keep thinking creatively and come until you come up with some decent answers.
38:39
Yeah, I tell the trainees too is like, diagnosis is written in pencil, right?
38:43
It's not in stone, not in ink.
38:45
It's in pencil with plenty of eraser marks all around it and leave your rule outs and your verses and keep that all there in that initial diagnoses part.
38:55
Absolutely.
38:57
Yeah.
38:58
How can, I mean, you also describe expertise as a risk factor.
39:03
And what are some signs that like a clinician is slipping into expert mode errors?
39:11
Well, we know that for the first six years of our practice, we're struggling to do the hypothetic, the adaptive reasoning.
39:19
We're struggling to remember biomedical information and put it all together and to create good hypotheses.
39:27
But we know that most of us, after about six years of practice, our brains reorganize into expert mode.
39:35
So if you've been practicing six years or longer, you're in expert mode, okay?
39:40
So we know that you have switched to what we call illness scripts, where you uh match the features of a diagnosis to the consequences of a diagnosis automatically.
39:56
And that sounds very familiar because that's type one thinking.
40:00
You're back from the scientific method to the built-in human error mechanisms that are going to pattern match.
40:09
we overdo pattern matching as experts.
40:11
We narrow our focus ah and we tend to only think about things we've done before and things we've seen before.
40:19
So if suddenly we seem to be diagnosing a lot of bipolar disorder, suddenly everything's bipolar disorder.
40:25
ah So just accept that if you've been practicing for six years or longer, you're doing this.
40:33
And so you need to really, really try to cognitively catch yourself doing it.
40:39
Try to use counterfactuals.
40:41
Try to think, okay, well, if I was wrong, what could I have done that would have made this wrong?
40:46
How can I go back and try to catch my error?
40:50
I've always tried to be a general psychiatrist.
40:53
I don't always diagnose the same thing over and over again.
40:56
But nevertheless, I get a lot of referrals for bipolar disorder.
41:01
I like treating that.
41:02
I'm very happy with the outcomes.
41:04
And so I could be known as that person.
41:07
Okay.
41:09
This scares me because I don't want to be diagnosing everybody that comes into the office as bipolar.
41:16
And so what, what I've been trying to do is to prove beyond the shadow of a doubt that I don't have, that this patient doesn't have bipolar disorder.
41:26
probably go further than most to do that.
41:30
And I think that that's kind of a good example.
41:32
ah You want to make sure that you're not diagnosing the same thing over and over and over again.
41:38
and try very hard not to be known as the ADHD specialist or the autism specialist because people are going to point their data towards you to help reinforce that bias.
41:51
Yeah.
41:52
The old saying of everything.
41:54
If you're a hammer, the whole world is nails, right?
41:56
Everything looks like a nail.
41:57
um If you're in charge of like saying, here's how I want to train the future generation of psychiatrists redesigning psychiatric training from scratch.
42:11
Would there be specific cognitive skills that you would add in there or reflections or exercises that you would want to have for all?
42:20
the next generation of psychiatrists entering the field.
42:24
Yes, I would want to focus less on memorizing the DSM and more on developing critical thinking skills and the habits that go along with those.
42:38
I would want to help them uh create reflective journals.
42:42
ah The surgeons always have a uh list of all the surgeries they've ever done.
42:48
Well, we probably need to have something similar.
42:50
We need to have uh a journal where we can
42:55
One, take the time to reflect on what we've been doing in our practice in any given cases.
43:02
We want to take time to review old records, even records from years ago, active patients and old patients.
43:11
You'll find great opportunity for learning from yourself from that way.
43:16
And then of course, we want to encourage uh consultation, uh formal consultation, not hallway consultation.
43:24
but formal consultation with each other and many, many case discussions that focus not only on the therapeutics, but heavily on the diagnostics.
43:34
uh And I think that we've got to build more awareness that we are infallible and we are going to make cognitive errors all the time.
43:48
And therefore we need to be uh looking for them.
43:52
We need to create an environment among learners and colleagues where it's not shameful to have made an error, but it's part of the process and we need to come to each other openly
44:05
to discuss them with each other so that we can learn from each other.
44:09
The best way to get feedback is to listen and to ask for it.
44:14
If you don't do that, you're just going to be in isolation and you're just going to keep repeating.
44:19
the same patterns over and over again with that narrow focus and you're gonna make more and more mistakes.
44:27
I have the benefit of like, again, working in the group practice for my, with the hospital system, every meet every Monday, we have Monday meetings with the docs and we discuss
44:39
cases and that's great.
44:42
And then my prep practice, I don't have that, you know, so I have, you know, I have the opposite of that.
44:48
I can't run it with anybody.
44:49
I have a couple of docs who are doing some part-time work with me and you know, we can kind of maybe run some cases here and there, but like, again, situations where
44:57
people are in private practice, you know, and they're operating out of their home offices or operating alone, or again, shortage areas, what are some strategies that they can use
45:07
to get that kind of peer support, that collaboration that's there?
45:12
I think that we need to be much more open to asking for second opinions.
45:18
We need to do it much quicker.
45:20
ah At every impasse or if we don't even know where to begin treatment, go ahead and get a second opinion from someone that you trust.
45:29
uh Explain to the patient why you think this is good medicine, discuss it with the provider, discuss it with the patient thoroughly, and that gives you lots of soundboards
45:41
and feedback.
45:42
Even if you don't agree with what your colleague has said, it opens the door, it opens your mind to further possibilities.
45:52
And I think that that is a really, really good habit to get into that again, I think your patients are going to like and respect as well.
46:00
So that's a really good way from keeping yourself from being isolated.
46:04
You may not have a coffee pot to gather around, but nevertheless, you can consult on the same patients uh in depth.
46:12
Yeah.
46:13
I've been encouraged by seeing like a lot of the social media stuff in regards to psychiatrists being able to connect with each other, therapists connecting with each
46:23
other.
46:24
There's a Facebook group of psychiatrists where, and I, a lot of times you see things where there's an anonymous kind of case report or anonymous kind of someone who's running
46:33
a case to see like, Hey, what would you do in these situations?
46:36
And like, it's great, right?
46:38
Cause I think, you you see the benefits, the idea of
46:42
the spirit of collaboration that's there, alternative viewpoints to kind of be like, I didn't think of that before.
46:50
And I think that's something that we have to kind of use a lot more is like, especially for people who may not have access to, you know, five other docs that they're working
46:59
with.
47:01
My only concern about that is I also see in those settings, people say things like, I've got a treatment resistant bipolar disorder, what should I use?
47:10
And people immediately start telling them without asking any questions whatsoever.
47:16
So, I mean, we do want to make sure that when we're getting consultation, we're getting real consultation, that the people have been able to examine the data uh thoroughly from
47:26
their own perspective and that they're not just spitting out.
47:29
quick answers, but apart from that, I agree with you.
47:32
Yeah.
47:34
And I know you had also talked about like doing some stuff with Doximity and I know like one of the hot topics is AI as a whole.
47:41
I've been using the Doximity's GPT to help kind of synthesize some of like my shorthand notes into cohesive narratives and then playing around with it a little bit more.
47:55
Your thoughts on just like AI, large language models, any of those kinds of things and helping us out or again,
48:03
bringing in other alternative perspectives potentially.
48:06
Yes, I have a whole section in this book about that and I've written several articles for Psych News about that.
48:14
My first thing is I think we all need to understand that generative AI is not a search engine.
48:21
Please don't use it as a search engine.
48:24
Even the improving models.
48:26
The bottom line is that humans make cognitive mistakes at a measurable degree.
48:34
AI makes
48:35
cognitive mistakes at a measurable degree and at the current levels of the kinds of AI that we're having and Generative is the most popular one right now.
48:45
Yeah, we're making errors at about the same rate ah That one of the things if you're doing a search engine Search you can look at the original data and see how valid that appears to
48:58
be Unfortunately with some uh companies with AI you don't know where the data came from
49:05
uh There are steps to improve that and give you that and that is getting better.
49:09
But right now, just be aware that AI is just as open to human error and uh confabulation uh as we are as humans.
49:21
So unfortunately, it's not going to save the day anytime soon.
49:28
yes, it's definitely needing a lot of double, triple checking.
49:31
So just as we check that, we have to check ourselves as well, right?
49:36
Yes, absolutely.
49:37
Neither one of us are safe is what I'm saying.
49:40
Yeah.
49:41
Yeah.
49:42
Are there, are there just kind of like looking back?
49:46
I think you kind of referenced it a little bit, but like any kind of mistakes that you find yourself kind of making or catching yourself here and there with.
49:54
Well, of course I make all of them too, being human.
49:59
Probably the one that frustrates me the most is making assumptions.
50:05
I am very into not making assumptions and I'm always correcting my family and friends about making assumptions.
50:10
And then I turn right around and do it, you know, and that's throughout my life.
50:14
And it really frustrates me when I do that because I know better.
50:19
And that's the point.
50:20
We all know better.
50:22
And I said, we're well trained, we're well meaning, we're very bright, and we still make these human errors because that's the way our brains function.
50:30
So uh just be aware that you're going to make the same mistakes over and over and over again.
50:36
The goal is anticipating that, planning for it, having a plan to catch them and to respond.
50:43
ah Because unfortunately, all get frustrated with this every day.
50:51
then I'll carry over one thing from my old podcast is because we want to keep our psychiatrists alive and well, and we know that their burnout is a big thing for us, what
51:03
are some ways of doing self-care for yourself?
51:06
Some ways that you do some self-care.
51:09
Yeah.
51:09
And I mentioned this in the most recent book too.
51:12
Of course, if we're impaired in any way, all of our errors are going to go sky high.
51:19
So of course, taking care of ourselves is essential for taking care of our patients, as you so wisely point out.
51:28
need, psychiatrists are actually pretty good at this because this is what we do.
51:35
We're really good at creating boundaries.
51:38
And so you do need a boundary between your work and between the rest of your life, a healthy boundary.
51:43
uh You need to uh create time for rest and for uh relaxation and reflection.
51:51
Reflection is part of your work, but it's also part of your recovery.
51:56
We know that if you stretch your vacation out throughout the year, you are less likely to show signs and symptoms of burnout or stress.
52:05
uh
52:06
We know that uh having close relationships with other people outside of your uh career is going to reduce your stress.
52:17
We also know that getting together with your colleagues informally, such as at medical meetings or uh medical society meetings or departmental meetings, actually has a big
52:30
factor in reducing stress and burnout as well.
52:34
uh
52:36
The basic wellness things that we all know about and talk about, actually doing those is very, very important for not only your own health, but also the success of your treatments
52:49
with your patients.
52:51
Yeah.
52:52
So everybody go to your holiday parties, holiday work parties, cause it's a good thing for everybody.
52:58
all right.
52:59
Dr.
52:59
Putman.
53:00
see I did it again.
53:01
Dr.
53:02
Putman.
53:03
What any kind of like parting words or plugs for yourself or any other kind of like, yeah.
53:09
Wrap it up a bit.
53:11
Well, just ask us, we all are in this field because we want to help people and we want to do the best job that we can to do that.
53:20
And just be aware that we unwittingly make lots of mistakes and that that's human and that's natural.
53:29
And just like we help our patients understand that about themselves, we need to understand that about ourselves as well.
53:35
So let's just anticipate that this is going to be happening.
53:39
and that our goal is not to make sure it never happens, but to be prepared for it when it does and to be able to identify it.
53:47
And if we can do that, then I think we are really fulfilling our role in the Therapeutic Alliance in a much better way, because then we truly are giving our patients our best most
53:59
of the time.
54:01
Yeah.
54:02
Thank you again so much, Dr.
54:03
Pubman for your time.
54:05
Thank you for your special report.
54:07
Again, you can find it in the January edition of Psychiatric News, addressing cognitive error in psychiatric practice.
54:15
The book is out.
54:17
That is also available.
54:18
Remind us the name of the book again.
54:19
is.
54:20
Thinking again, reducing cognitive errors in psychiatric practice.
54:25
Yes, make sure to buy the book.
54:26
So, alright, until next time, be safe and be well.
54:33
you