Ask Me MD: Medical School for the real world
Ask Me MD: Medical School for the real world
Joseph Francis, MD - Private Equity in Dermatology
Dr. Joseph Francis, MD, a practicing Mohs surgeon, talks about his statistical analysis of Medicare claims data and its intersection with private equity owned dermatology practices. Dr. Francis' study was published in
If you have questions or ideas for a show, send us an email at questions@askmemdpodcast.com. Hear the latest podcast at http://askmemdpodcast.com or through your favorite podcast directory.
Ask Me MD, medical school for the real world with the MD Dr. D.J. Verret. Greetings and welcome to Ask Me MD, medical school for the real world. I'm Dr. D.J. Verret and this week, we've had a series of interviews of researchers who have looked into the effects of private equity investments in health care. Today we're joined by Dr. Joseph Francis, a practicing Mohs surgeon in Florida, who has looked into the effects of private equity investment in dermatology. We'll talk to Dr. Francis about his research right after this. Welcome back to Ask Me MD, medical school for the real world. I'm Dr. DJ. Verret abd today we have the pleasure of talking with Dr. Joseph Francis, a practicing Mohs surgeon in Florida, about his research into private equity involvement in dermatology practices. Doctor, thanks for joining us.
Joseph Francis, MD :Thank you, DJ.
D.J. Verret, MD, FACS :So can you tell us a little bit about yourself and your background before we start talking about your research?
Joseph Francis, MD :Sure. I was born in Queens and raised in this rural area of Florida. Most of my teen years, my family situation sort of changed, I didn't have health insurance. So as part of this, you know, working poor gap where you know, you made too much money to be on Medicaid, but not really enough to afford health insurance. So I think that really had a big impact on me. There, I went to Duke for undergraduate school. Then after that, I worked for PricewaterhouseCoopers as a management consultant, where sort of I explored my interested in in programming, and my main goal really was to just be financially independent. Afterwards, you know, after a few years of that, I went to medical school at Howard University in Washington, DC. And I did my internship at Harbor UCLA, which are both sort of inner city hospitals. And a lot of my mentors were, I guess, you'd say patient advocates. And, you know, that had a big impact on me also, you know, my, I did that I did a dermatology residency at Medical College of Virginia, in Richmond, and then a fellowship with Richard Bennett, and Mohs surgery in Santa Monica, California. And, you know, working with Dr. Bennett probably had the greatest effect on me, personally and professionally, just in the sense of, you know, learning to look at, you know, research topics, well, less from an ideological perspective and more of like, a search for the truth. So, that's where I find myself today.
D.J. Verret, MD, FACS :So was medical school a second career for you? Or were you always focused on getting to medical school and just needed a path to get there?
Joseph Francis, MD :Well, I guess, you know, I'm, I wasn't like gung ho about it, I kind of just, you know, I kind of wanted to leave my options open, because I knew that, you know, medical school is a significant expense. And I wasn't sort of financially ready to put myself into a position where, you know, I'd be sort of like locked up in it for another, at least seven years without having an income or out without having my own being financially independent. So, these were things sort of, you know, they were kind of ingrained in me, all throughout undergraduate school and kind of sort of got in my decision making. But you know, it doesn't feel like a second career. Yeah. So, when I was in medical school, I kind of felt like I had the perspective of, you know, working as a consultant. I knew what you know, working in the corporate world and traveling were like, so I don't know didn't really take it for granted. I, you know, I studied a lot. Yeah. And kind of focused on being successful.
D.J. Verret, MD, FACS :So fast forward to today, obviously, private practice mode surgeon. But you also took an interest in doing research into the business of medicine. Tell us about what you're doing, and, and how you kind of came to that as well.
Joseph Francis, MD :So, well, that goes back to, you know, my fellowship director, Dr. Bennett, who was around 2014, I got a text message from him, I finished my fellowship in 2012. In 2014, I have a text message from him Joe, you know, they Medicare just released this data. And, you know, I think that, you know, we can take a look at, you know, the most surgeons in this data. And, you know, he had a couple of ideas. So, you know, we chatted, and we realized that, we can, we can basically what we could do was to take CPT codes used by dermatologists and try to sort of recreate a way that they practiced. And so we sort of tested that kind of just back and forth on the phone. And we said, You know what, I think we could actually do it. And that was kind of like, one big step that we took. So we said, well, let's take the billing codes, this. And that's what the, you know, the CMS data release basically was, was the first time that, you know, individual physicians were identified in a data set that looked at exactly how they're billing and what CPT codes they're billing and the actual counts. So we said, Well, why don't we just do a small scale project and look at most surgery fellowship directors and look to see how they're practicing low surgery. So we did that. And we, we said, well, how do we present this and say, Okay, well, let's do a histogram. So basically, what we did was a basic histogram, where the number of stages average number of stages, it would take, for them to clear a tumor. And the average was about 1.7. With a standard deviation of point three, however, you could see some people who averaged three or four stages to take out a skin cancer, which was kind of nuts. So I know, you know, you're familiar with most surgery DJ. So you can imagine somebody averaging three or four states, that's pretty high.
D.J. Verret, MD, FACS :It seems like just my feeling would be it's one to two stages for most people, and that, that usually takes care of it.
Joseph Francis, MD :Yeah, absolutely. Yeah, like average. 1.7. So we looked at fellowship directors, and then I said, Well, you know, I think I can, you know, process this entire data set and look at, you know, statistics for everyone, which I was able to do after, you know, lots of sort of data wrangling and everything and, you know, saw lots of interesting things there. You know, there were there was a big group of people that averaged one stage of low surgery, which, which, as you know, that's that's highly irregular averaging one stage, you know, if you're doing lots of cases. And, you know, then again, you know, there's also these people that are about, you know, there's a group there's a cohort of people that are, you know, three or four standard deviations above the mean. You know, as I got into it, I was sort of inspired by Jonathan Bluhm, who, when he they released the data, he put up a statement saying, I'm just going to read it because I like it so much. He says release of physician identifiable payment information will serve a significant public interest by increasing transparency of Medicare payments to physicians, which are governed by statutory requirements and shed light on Medicare fraud, waste and abuse. So I took I kind of took that as a challenge and said, You know, I can I can do this, I can sort of develop my own metrics, I have access to the data. So I was like, Okay, well, I'll take you up on that offer.
D.J. Verret, MD, FACS :Well, in and when we were talking offline, it was really interesting to me that the knowledge base that you came with and the skill set you came with to actually do that analysis, can you kind of share with our listeners, how you just the nuts and bolts of how you did it?
Joseph Francis, MD :Sure. So, you know, whenever I'm thinking of a project, so I kind of start I like to start with a very simple idea or a problem like, okay, let's say, you know, let's hear like listen to chatter. Someone says, well, there's a lot of people doing special stains on pathology specimens. And, you know, maybe there's, maybe there's something there. So there may be some labs that are, you know, sort of reflexively ordering special stains that are maybe unnecessary. So then I say well, Can I? Can I show? can I prove that? Or can I? Can I investigate this idea? And so I am pretty familiar with sort of the data dictionaries of all the publicly available datasets, and some of the ones that are not publicly available. So then I kind of delve into the these billing practices like fixes, let's say, for example, how would you build a special stance, so first, there probably have to be a three or five associated with it. So that's, you know, pathologic examination of tissue Technical and Professional component. And then, you know, there's an add on code for special stain. So then, you know, you can sort of develop a metric, sort of, like an equation to show, you know, what, can you like, an equation that, you know, includes these billing codes? Like, how can you prove this? or How can you develop them, you know, sort of a histogram to sort of look at this and see how prevalent it is. And then I use a program called R. And I write is, you know, sort of some basic code that, you know, from my pseudocode, when I was sort of planning it out, and, you know, with our such a powerful tool, and it's free, you can, I remember watching people, you know, at meetings, and suddenly, when the state of first kanaday, they go data that cms.gov, and they try to open up the, you know, Excel spreadsheet that contains, you know, all this data, it's like nine gigabytes, like computers, just like, you know, sit there forever, which is just can't be done, you really need, you know, some statistical program like our, that has the memory management, and, you know, the manipulation capabilities to do this. So, I guess that, I guess that answers your question.
D.J. Verret, MD, FACS :It does. I mean, it just kind of, it's very interesting to me that you bring together a skill set, both as a physician, knowing how to bill for it, the technical ability to use programming resources to actually analyze the data, and the understanding of the analysis to come to a conclusion. It's a skill set that I don't think many researchers normally have, and would would actually take multiple researchers, in most situations to come to the same conclusions that you do. And I suspect you're able to come to those conclusions a lot faster than other people looking into the same data sets are.
Joseph Francis, MD :we Yeah, that anything, I think you hit the nail on the head there. But I mean, I'm hoping that there's more people like me, like, the, you know, I consider myself Generation X, you know, but I think, you know, the defining characteristic about millennials, I grew up with technology, but I guess, but I was kind of, I also grew up with technology, like I was, you know, I was messing around with computers and programming, you know, when, when I was probably like, maybe 10, or 11. But, you know, that was just kind of like, my hobby, because I wasn't any good at sports. But, you know, I, I'm hoping that, you know, that, you know, the latest generation of physicians can, you know, kind of use these tools, and, you know, maybe they have, you know, you know, patient advocacy backgrounds, or, you know, they're interested in you know, searching for, you know, these ideas, and they can do this type of work. But, you know, as you said right now it takes teams of people to do it. And the problem with that is that, you know, things get lost in translation. And there's nothing like, you know, caught like wading in the muck so you, you create this, you know, you build a, you know, data set that kind of, you know, it gets takes a while but part of the process is you, you create your own sort of custom data set where you can sort of just wade through and you can kind of look to see what's going on. Kind of like you know, when I go fishing again, same thing, you know, kind of waiting out there, trying to trying to get a feel for what's happening in the water. Now, you're gonna do the same thing with data. That's where you really get a lot of insights from it.
D.J. Verret, MD, FACS :It just be careful wading in Florida here. You have a lot of alligators out that,
Unknown Speaker :oh, yeah, we have real, you know, there's all kinds of things expressed.
D.J. Verret, MD, FACS :At the top of the show, I teased we talk about private equity. So you mentioned when you started delving into the data, you were looking at Mohs academic Mohs surgeons, most fellowship surgeons and and how many procedures they're doing. When did you kind of shift that focus into looking at private equity involvement in dermatology?
Joseph Francis, MD :Well, so, so there's a guy I met. I don't remember how I met him, to be honest, name Sailesh Kanda. He is, you know, very sort of, again, it's kind of ideologically opposed to the idea of private equity. And he sent a lot of research, you know, showing, you know, he's really, you know, not really good for medicine. It's not really good for the physicians, it's not really good for the field of Dermatology. And then he knows a lot of people agree with him, you know, There's a lot of articles, you know, Joshua Sharfstein, Gandhian soon, all these people that, you know, have this, you know, again, ideological views of how, you know, medicine should be practiced everything like that. And so, you know, we had, we decided to do Grand Rounds together at the University of Florida about private equity, because he felt that, you know, a lot of his residence, were going off to were for private equity, and then sort of later, you know, coming back and realizing, like, all these, you know, all these issues that, you know, maybe that maybe this should be part of their training, you know, like, let's take a look at, you know, dermatology business models and everything. So I said, Well, you know, how about, we joined forces, like, I'll talk about, you know, what I know about data, and you can discuss, you know, private equity, we can kind of fuse them together. So, you know, that, I guess that's sort of how, you know, we started to work together. And then I started to combine my observing to look through my outlier analysis and say, Well, you know, let's take a look at these outliers in detail, like, Are any of them associated with private equity? And, you know, let's look at it sort of overtime, and then I start to realize, like, wow, you know, a lot of these elements that, you know, you probably know, people in your community DJ, like, you say, Well, you know, I know how this guy practices, I've seen maybe some mutual patients, and they're sort of like, like a Bane on the on the community, you know, like, you end up having to, you know, fix their mistakes or deal with a lot of other issues, you know? Yes, I do know, exactly, ever happens, every doctor right? Now, let's say that, like, you know, these practice patterns that are sort of, embodied by these types of doctors. And this is what I consider the, you know, sort of the outlier group, you know, they're, they're usually, you know, sort of these doctors in the community that, you know, everyone has to kind of make up for them, you know, and, you know, makes us look good, right. But now, let's say, you know, private equity starts to get involved and buy them out, you know, basically what, what I realized that they were doing is, like, financially motivated physicians who are, you know, sort of exploiting the system at the expense of patient care, like, these are the kind of people that private equity was was attracted to, and they kind of want to take what they're doing and do it on a larger scale. And that's evidenced in the data.
D.J. Verret, MD, FACS :So let's talk about the data itself. Were there, why did you find basically where there were there some positive trends or negative trends or just kind of big picture? What did you see when you started looking at the data?
Joseph Francis, MD :Okay, well, okay, so there's a few things. So with regards to the biopsy outliers, I mean, there's obvious trend that, you know, these these doctors, and, you know, to be honest, you know, I kind of was like, we know, how do you average 11 biopsies per patient, you know, or how do you average, you know, six or seven biopsies per patient per year. I mean, that's like a ridiculous number. So I kind of, you know, when I was practicing in West Palm Beach, you know, a lot of them were down there in Palm Beach County, so I kind of went out of my way to meet these people and kind of understand them, or try to understand them, and a lot of them were, you know, they're like sociopaths, you know, like, they have this idea. They believe that what they're doing is right, and you really can't convince them otherwise. So, that was that was one thing, you know, that they were sort of preferentially, you know, selling to private equity or being selected to join privately.
D.J. Verret, MD, FACS :So, when and let me interrupt you for a second. When you say 11 biopsies per patient, you mean, every patient they saw had, on average, 11 biopsies per year or
Joseph Francis, MD :so. So, okay, so there's limitations on you know, that statement. So this is I guess I'll just say it fully. So averaging 11 biopsies per patient per year on fee for service Medicare, meaning that the average fee for service Medicare patient that walks in the door gets, you know, 11 biopsies. I know, that's hard to take in, right?
D.J. Verret, MD, FACS :Yes, that's why I was that that's why
Joseph Francis, MD :You thought I mispoke.
D.J. Verret, MD, FACS :Yes, I really did. Because I, you know, I do a lot of skin cancer reconstructions after Mohs surgery. And we have some patients that are in every three or four months having things taken off. But the vast majority of patients may be in, you know, at most once a year, or not even that frequently so to think that every one of those patients would have gotten 11 biopsies per year is is quite astounding, and I think would be a significant outlier. Just from my Gestalt of seeing those patients,
Joseph Francis, MD :yes. So what's, you know, what's what's even more discerning about that DJ is, you know, why was this person able to do this for so long? Without any repercussions? So where's the, you know, where's the government? Where's the, you know, data analysis that the government is doing? Where are the auditors? I don't know, I don't know how to answer that question. But what I do know is that, if you can, if you can go and sort of get away with this for so long, you know, that kind of opens up an opportunity, because, you know, a business perspective, you know, if you can do this, and without any repercussions, then, you know, that's a good business opportunity, you know, because, you know, you could create a, you know, a practice of large corporate practice system that, you know, employs these sorts of things, and or, like a self referral based, you know, practice system that, you know, that does this, and, you know, it's totally fine, because no one's looking.
D.J. Verret, MD, FACS :Now, now, were you able to link those outliers in biopsies to private equity owned? practices?
Joseph Francis, MD :Correct. So that was the patient, its future considerations for dermatology in the 21st century. paper that, you know, if you, if you want, I mean, it's, it was, you know, pulled by the journal because of pressure from, you know, private equity interests within the aad, or, you know, so so we assume, but, um, and then there was a New York Times article about it, and that paper was put back, you know, after being accepted, the paper was put back online. But that there's a, there's a table on there that I'm hoping somebody with, with a good understanding of, you know, you know, CPT codes, and, you know, the structure can, you know, can really look at it and analyze it, and see, you know, what I did, and you know why it's important, but yeah, I mean, it does that. The other thing that I found was, and it's also in the paper, the same paper is, there was a group of so there was a private equity backed, mobile dermatology group that, I mean, it was a mobile dermatologist with no dermatologists, it was just basically just physician extenders that were going around to nursing homes in the state of Michigan, and we're doing intralesional injections on patients in the nursing homes, and you know, I think was like, very high percentage of them had Alzheimer's disease. So it was like 70 to 80%. And intralesional. injection is usually, you know, steroids, you know, so there's really no, I mean, they're billing sort of hundreds of thousands of dollars of intralesional injections, on these Alzheimers patients, and then we're sort of traveling around to the different ones. And, you know, doesn't really make any sense to do this, like, I mean, so there's, there's a lot of little things like that, and the data that you can see, you just have to really, really focus in on it. And like, you know, that, you know, you can find that kind of stuff you can find, you know, once you start linking all these things together, you know, you can start to look at, well, what percentage of patients have that see this person I've Alzheimers and, you know, you know, sort everything and you can, you can find little trends like this. Actually, that's, that was sort of incidental thing that wasn't something we found without even looking without even trying to look at it. How we're really silly with private equity.
D.J. Verret, MD, FACS :Well, and so my next question is, how are you able to link the physicians with the with a private equity ownership? Obviously, the Medicare data simply gives you NPI number, physician name and the procedure, it doesn't necessarily show that it's linked to a corporate entity? How were you able to make those connections?
Joseph Francis, MD :Good question. So, one, you can do it manually, which is, you know, you can do if you have a small once you've Once you've created like a group of outliers, you can you can associate them you can just look up the lookup the physician find out where it works. There's another data set, it's called. So what I like to do is I like I alluded to before, I like to link lots of different data sets. So another data set called physician compare where, you know, the, the key is not unique, but it's keyed by NPI number you can make it unique. And then you can link you know, this physician compare data set with with the, you know, Medicare physician payment data set. And that physical paired data set will tell you where that person works and you know, every office actually that they work at,
D.J. Verret, MD, FACS :and then how what's the what's the connection then between the office they work at and private equity because I'm assuming it doesn't say KKR in the physician compare data set, right?
Joseph Francis, MD :Oh, so yeah, so that is sort of, that's where Dr. Khan comes in. So he has, you know, his data set that he created of, you know, all of all physicians and you know, all physician groups, and you know, sort of what, you know, private equity group they work with.
D.J. Verret, MD, FACS :So it was actually and we keep it updated. So it was combining across it sounds like three large data sets, including one that's that Dr. Panda said, that's probably fairly labor intensive and manual to to create those relationships is that
Joseph Francis, MD :well, yeah, I mean, you also have to understand that, you know, there are like, I think, you know, last time I checked, I think something like five to 10% of dermatology practices are owned by private equity. And that's as a whole, there are certain large markets, you know, where it's more than that. So, you know, some certain large metro areas, even in Florida, you know, there can be dominated by, you know, private equity, at least in dermatology.
D.J. Verret, MD, FACS :So, we talked kind of about some, some very interesting, but negative findings. Was there anything positive or striking that you saw when when you looked at the data that you didn't expect?
Joseph Francis, MD :Yeah, I? Well, I mean, you know, 95% of people are honest. And I don't I don't think that, you know, I don't think that dermatologist that word for private equity are dishonest, I don't think so. I think a vast majority of them are, and I don't like to denigrate anyone that does work for private, and I don't, I don't think that it's right to do that. And I think that, you know, you can criticize the system, but based on facts, and, but, and there are definitely, you know, a lot of people in private practice, even in academics that, you know, practice dishonesty or are outliers. So, I don't think that you can kind of pin it all on, you know, corporations and private equity, but I think you can look at their, their, their sort of structure and the way that they do things and who they're associating themselves with interest conclusions from that.
D.J. Verret, MD, FACS :Looking ahead, do you have any plans for future research? It sounds like, like I said, you have a very unique skill set and ability. Is there anything that that intrigues you at this point you're going to start looking at?
Joseph Francis, MD :Well, yeah, like I mentioned, I'd like to look at pathology. I'd like to sort of look at, I'd like to look more at trends, like, what's happened since, you know, we have publicly available data, like what's happened since 2012. And, you know, I've found ways to sort of look at, look at data over the past, you know, six years and try to look at trends, like, you know, what happens after, you know, physician is acquired by private equity, like, what ends up happening to that practice? I think that's, that's an important thing to show. If you're ever if you're really gonna, you know, try to disorder draw conclusions. I think that there's, there's lots of work that still needs to be done. You know, just because, you know, just in not necessarily with private equity, but just looking at looking for the truth that's in the data. There's so much, there's so much to be discovered there. I mean, when it's, you know, sort of frame the right way that can have not just like, you know, lasting effects on on possible reimbursement or, but could you know, could have, you know, lots of different facts that on the way that we practice technology in the future.
D.J. Verret, MD, FACS :fascinating stuff. Joe, I really appreciate you coming on the show.
Joseph Francis, MD :Well, thanks for having me.
D.J. Verret, MD, FACS :We've been talking with Dr. Joseph Francis, a practicing Mohs surgeon in Florida, about his research into private equity investments in dermatology. You're listening to Ask Me MD, medical school for the real world. I'm Dr. D.J. Verret. Until next time, make it an awesome week.
Announcer :Thank you for joining us for another episode of Ask me MD medical school for the real world with Dr. D.J. Verret. If you have a question or an idea for a show, send us an email at questions at ask me Md podcast.com.