EP129: Navigating Diagnostic Discordance in Dermatopathology with Dr. Gregory A. Hosler

In this episode of SEE HEAR FEEL, Christine J. Ko welcomes Dr. Gregory A. Hosler, president of the Dermatopathology Division of Sonic Healthcare USA. The conversation delves into his recent study on diagnostic discordance in dermatopathology, exploring its implications for patient management and highlighting the importance of ancillary testing and second opinions. Dr. Hosler also shares insights on emotional intelligence, emphasizing the significance of self-awareness and continuous improvement in pathology practice.

00:00 Introduction and Guest Background

01:54 Personal Anecdote and Career Journey

02:31 New Role and Responsibilities

03:38 Diagnostic Discordance Study

07:20 Improving Diagnostic Practices

10:55 Emotional Intelligence and Self-Reflection

12:42 Final Thoughts and Conclusion

Gregory A. Hosler, MD, PhD is President of the Dermatopathology Division of Sonic Healthcare USA (SHUSA). Before taking this role, he was at ProPath, which joined SHUSA in 2021, and at ProPath, he served as partner, board member, and Director of Dermatopathology. He is a Clinical Professor in the Departments of Dermatology and Pathology at the University of Texas Southwestern (UTSW). He has held numerous leadership roles in local and national societies, including serving on the Executive Board for the American Society of Dermatopathology (ASDP, 2021-present), as President of the Texas Society of Pathologists (2019), President of the North Texas Society of Pathologists (2015), and President of the Educational Foundation of the TSP (2020).

Dr. Hosler obtained his B.S.E. in Chemical Engineering at Princeton University and completed his medical degree and graduate work at UTSW in Molecular and Cellular Biology and Immunology (MD/PhD Medical Scientist Training Program). He completed his residency in Anatomic and Clinical Pathology at The Johns Hopkins Medical Institutions (JHMI), and his fellowship in dermatopathology, also at JHMI. 

Dr. Hosler is very involved in education, teaching at the fellowship, residency, and medical school levels. He is the Director of the ACGME-accredited Dermatopathology Fellowship Program at ProPath. He actively speaks, writes, and directs courses on dermatopathology. He is the editor and co-author of two dermatopathology textbooks - Molecular Diagnostics in Dermatology: practical applications of molecular testing for the diagnosis and management of the dermatology patient (Springer) and Diagnostic dermatopathology: a guide to ancillary tests beyond the H&E (JP Medical Publishers). He is also the author of the melanocytic chapter in Weedon’s Skin Pathology (4th, 5th, and 6th Eds, Elsevier) and many peer-reviewed journal articles and reviews.

Transcript

Christine Ko: [00:00:00] Welcome back to SEE HEAR FEEL. Today I'm very happy to be with Dr. Gregory A. Hosler, MD, PhD. Dr. Hosler is president of the Dermatopathology Division of Sonic Healthcare USA. Before taking this role, he was at ProPath as partner, board member and director of dermatopathology. He joined Sonic Healthcare in 2021. He is currently still a clinical professor in the Departments of Dermatology and Pathology at the University of Texas Southwestern. He has held numerous leadership roles in local and national societies, including serving on the executive board for the American Society of Dermatopathology from 2021. He has also served as President of the Texas Society of Pathologists in 2019, as well as President of the North Texas Society of Pathologists in 2015, and President of the Educational Foundation of the Texas Society of Pathologists in 2020.

Dr. Hosler obtained his B. S. E. in Chemical Engineering at Princeton University and completed his medical degree [00:01:00] and graduate work at the University of Texas Southwestern in Molecular and Cellular Biology and Immunology as part of the Medical Scientist Training Program. He completed residency in Anatomic and Clinical Pathology at the Johns Hopkins Medical Institutions and his fellowship in dermatopathology at the same institution.

He is passionate about education, teaching at all levels, including the medical school level. He is the editor and co author of two dermatopathology textbooks, Molecular Diagnostics in Dermatology: Practical Applications of Molecular Testing for the Diagnosis and Management of the Dermatology Patient, as well as Diagnostic Dermatopathology: A Guide to Ancillary Tests Beyond the H&E. He has many peer reviewed journal articles and reviews, one of which was recently published in April of 2024 in the Journal of Cutaneous Pathology.

Welcome to Greg. 

Greg Hosler: Thank you so much.

Christine Ko: Can you first share a personal anecdote?

Greg Hosler: So this is going to sound like a bio sketch, but [00:02:00] bear with me. Beginning in 2006, I was a dermatopathologist, and I took on the role of fellowship director in 2012, and then the director of DermPath in 2015. ProPath became part of Sonic Healthcare in 2021. And at the end of last year, I was asked to take on this new role of President of the Dermpath Division. So for some perspective, as a director of dermpath at ProPath, my primary role was signing out cases. I did have some admin duties and was involved in teaching and writing and did some research, but make no mistake, I was signing out a lot of cases. In this new role, I've significantly reduced my routine work and most of my dermpath practice is focused on consultations. In addition to this, a major part of my new job is trying to get our practice into the new digital age of pathology. So what that means is implementing a new LIS, digital workflows, AI. And while I was always involved in running our practice at ProPath, this is a new level. So large scale laboratory operations with everything [00:03:00] that goes into it, including the IT, the billing, contracting, HR.

Christine Ko: Yes. So are you learning on the job or do you have experience in knowing how to do all of these various things?

Greg Hosler: Yeah. As part of ProPath, I was involved in running that practice. I wasn't a sole leader of that practice. We were partnership. And so there was some experience there, but this is a different level. And like many things, I look to mentors to help guide me. I do not profess to be an expert in this area but it's super exciting, challenging me in a lot of ways. I definitely lean on some of the senior folks who've been doing this for decades.

Christine Ko: Your recent article in the Journal of Cutaneous Pathology. It's on diagnostic discordance, meaning when one person, one good dermatopathologist says something is one thing and another, at least one other, but sometimes more than one other, gives another diagnosis for the same thing at a different point in time with both people or all the people not knowing, [00:04:00] right? They're blinded to what someone else said. Can you talk a little bit about diagnostic discordance?

Greg Hosler: Yeah, sure. A couple of points here to a lay person. They think we look at a microscope or a screen and we see like a plus sign or a minus sign and then render a diagnosis. What's lost in that is all the nuance and the difficulties in rendering a diagnosis. We sometimes hem and haw, we show it to colleagues, we get additional testing, and we may ultimately come to a specific diagnosis. But once we do, the clinician and the patient go down a specific management pathway and what's lost is all the nuance that went into that.

The premise here of this manuscript was to shine a new light on the concept of diagnostic discordance. This isn't a new concept per se but what we went out to do is look at this in a new light with a large dataset, a large group of dermatopathologists, and with an angle of how does this affect patient management with these different [00:05:00] diagnoses, a large number of cases, over 3, 000 cases and nine sites. These cases were evaluated by multiple dermatopathologists. Just in summary, what we showed that there was about 24 percent of cases where there was some level of discordance, which can seem quite high and maybe a little bit unsettling for most, but it is in line with other research studies. And then we took those cases of discordance and we looked at, what is the impact on management? And this is really a virtual exercise. Knowing that benign lesions, often nothing is done. Malignant lesions, there are guidelines into the management, like the size of the excision or send a lymph node biopsy, things like that. And then these middle ground cases where there's a wide variability in how those are managed.

So taking those, and taking all the different diagnoses that the pathologist rendered, we looked at how that would impact the management of these patients. And you can imagine that some patients would be [00:06:00] over treated, if they were diagnosed with malignancy. Some might have been undertreated. We found about 7 percent of these cases fell into this category where there was a significant impact on the path those patients would have gone down management wise, had these been real cases and real follow up. 

Christine Ko: 7 percent. Yeah, it's a lot lower than 24%. So maybe one in four of my cases, if someone else looked at it, they would have a slightly different opinion. You could argue that maybe we have too many names in dermpath that like, it's all really the same thing. But for 7 percent, maybe averaged on the lower side and say one out of 20 of my cases, the patient would have something else done for them. That seems really... 

Greg Hosler: It's a little scary, right? So of that 7%, roughly half of those would be over treated, meaning that somebody gave them a diagnosis of malignancy, but probably, all [00:07:00] they needed was follow up; and the other half of that, the other three and a half percent significantly undertreated, meaning they were given a benign diagnosis. So yeah, when you start looking at the data you start to see that there's a great need for other opinions or other testing or whatever that next step for you may be. 

Christine Ko: Do you have any ideas on how we can practice better based on this diagnostic discordance that does exist?

Greg Hosler: Yeah it's a challenging area. One thing that's important to know is that ancillary testing is available. This manuscript, this study was based on H& E, and oftentimes when someone has a challenging lesion, they do additional testing. We have immunohistochemistry, we have molecular testing. They do come with a cost, literally, they can be very expensive. But I do think ancillary testing does play a role. The other thing that plays a role is show a colleague. Maybe that means walking down the hallway to a friend or showing something virtually to get a second opinion, or maybe a formal consult where you [00:08:00] send it off for an expert opinion. One of the main take homes from this is, don't be a hero. Overconfidence is the pathologist's greatest enemy. So having a little humility in the way we practice, understanding what we don't know. Sometimes you got to wave the white flag a little bit and say, I did the best I could. I need some other eyes on this or some other testing. Lastly, as far as just practical management, is communicating the level of difficulty of a case to the clinician. Whether it's a phone call or it's in the report, but discussing with clinicians the challenges of a case and what's best for the patient, I think that's best practices.

Christine Ko: I love that. Yes. I think when we put foremost what's best for the patient, then things become a little easier. There shouldn't really be any shame per se in there being diagnostic discordance. If really one in four cases, there's potential for diagnostic [00:09:00] discordance, it just shows that even at a high level, there is slight differences of opinion. 

I'm a big fan of Daniel Kahneman's book, Thinking, Fast and Slow. And then the follow up book, Noise, also applies to dermpath as well. What he refers to as noise is actually multiple different opinions for the same thing. So exactly your study, this diagnostic discordance. He gives other examples, like in law, it would be one judge giving one sentence versus another judge would give the same crime, the same person, a different type of sentence. Or the same judge actually will give a different, harsher sentence for essentially the same crime to someone else, like later in the day, when they're tired.

So there's this noise, basically discordance. His conclusion is that we have to face this and accept it and also realize that actually the noise matters in the sense that generally the average of opinions is actually the one that's more correct.

Greg Hosler: I love your example of the [00:10:00] judges because isn't that exactly what we do? I think there's even studies on it. You wake up early in the morning, booming out melanomas. And then, in the afternoon you're calling everything atypical, and it's just you get tired, your brain gets tired. So definitely, day to day or times within the day changes how we practice, just like you're describing with a judge's sentence.

Christine Ko: Yes, absolutely. When I talked to Arnaud de la Fourchardière, he gave a good point a lot of his cases are consultation cases like yours, so quite difficult. He has a setup where he will look at a case three different times before signing it out. Tim McCalmont, I think he calls it his set aside, where he leaves it till the next morning or a couple of hours or something like that. 

Greg Hosler: I definitely do that. 

Christine Ko: And you mentioned like our ancillary tests and stuff. And I know I do. I know a lot of my colleagues do. You order an extra test in order to give yourself extra time too. That's right. 

Greg Hosler: You know the tricks.

Christine Ko: Yeah. I'm going to segue a little bit then to other things that I talk about on this [00:11:00] podcast: emotional intelligence, deliberate practice, metacognition, thinking about our thinking. Do you have any thoughts that you want to share about any of those topics? 

Greg Hosler: I can share a couple of thoughts regarding emotional intelligence. I've always been data driven, and I think this is really no different in some ways. The problem that I realized as I started going down this path is that I didn't have any data about myself. I needed to take notes about myself, I can feel my ears getting warm when somebody talks. I do tons, in my new position, I do tons of remote zoom calls. I started taking my own little notes, like action items or followup items. One thing I started doing is also trying to get the temperature of the room. The temperature in the room and how people are feeling are important to document. Sometimes we do talk about things where people do get emotional. Emotions run high. 

Christine Ko: Yeah. I'm fascinated by this idea that you were collecting data on yourself, if your ears are getting warm or something. So [00:12:00] then would you then also see what you did next or create something for yourself to do next when you would feel that? 

Greg Hosler: If you don't document it, you forget. And I just feel like if you don't document it, remember it, it's just going to repeat itself. And I think when you have that data, you can immediately address the situation after the meeting, or maybe you just make a note. And then the next time that meeting is up, it's just in the back of your brain so you can take preventative action and things like that. 

Christine Ko: Yeah. I think that health care is a stressful place. Pathology operates at a different pace, I think, than the emergency room, say, but still, surgeons, patients need their diagnoses sooner rather than later. Do you have any final thoughts?

Greg Hosler: What I would say, just thinking about this paper and, my transition, sometimes I think about what gets me out of the bed and to work every day. And I would say for me, the big thing is just participating and pushing our specialty forward. I think this is [00:13:00] important for everybody to have some sort of cadence where they evaluate what gets you out and and what really gets you going.

Christine Ko: Thank you so much for your time.