inviting me. It's a pleasure to be here. For those of you that heard this story, you'll have to bear with me. In my fourth year, we entered what was called clinical clerkship, and that's when we entered the hospitals, and I really had no idea what I wanted to do as a specialty. In fact, if anything, maybe I thought the internal medicine was where I should go. As luck would have it, my first rotation was neurosurgery, and I was enthralled. I had no concept of what surgery had to offer for patients, but at the end of the rotation, I sat down with the neurosurgeon who was responsible for what were called the clinical clerks, and he was a man who was known for his bedside manner, but not normally in the way you would want to be known for your bedside manner. We talked through the rotation, and he said to me at the end, so what are you thinking in terms of a career choice? I said, neurosurgery seems really interesting, and he paused, and he looked at me, and he said, you have to be really smart to go into neurosurgery. So I stand before you, a humble orthopedic surgeon. I love orthopedics in no small part because I think 80% of what orthopedic surgeons do is among the most life-changing, quality of life-enhancing, cost-effective interventions in all of clinical medicine, but there is about 20% of what we do that maybe we shouldn't do, and 20% that we don't do that we should do, and that's probably where I spent the majority of my career. So that leads me to the title of my talk. It's no longer okay to say I practice differently than everyone else, and it is a provocative title. I don't mean to provoke you, but I do want to challenge you to think about the way we think. And the controls are there. So here I am. I spent most of my career as a children's orthopedic surgeon doing foot and ankle surgery at a hospital in Toronto. I spent two and a half years at the University of Oxford practicing at the Nuffield Department of Orthopedics and Musculoskeletal Research Science, and I have no conflicts. So when you think back on the practice of orthopedics, there are obviously some fundamental changes, more than a hundred years ago, the introduction of general anesthesia. We were the leaders in surgery in introducing what's become known as minimally invasive techniques, and it raises the question, what's going to be the next fundamental change? And I wonder if it'll be using best evidence to direct us to do the same treatment for the same condition, to resolve all these clinical, or many of these clinical conditions that confront us. So this is still somewhat true, is how do we form beliefs about treatments? Well, there's an acceptance and even a promotion of divergence in opinion about treatment recommendations. We stand with pride and say we do something differently than someone across the table, across the hallway, across the city, or across the country. And to a large extent, this could be reflected in our history, which was reliant on case series. And case series can be used to support just about any treatment recommendation you might want to invoke. Now, don't get me wrong, case series has been responsible for many of the complete paradigm-shifting treatments in orthopedics. John Charnley, 1972, the total hip replacement, the AO group, internal fixation, the fractures in the 60s and 70s. Paul Harrington, instrumentation of the spine in the 50s, and in my own specialty, Ponseti treatment for clubfoot in the late 40s. But when you think about these totally paradigm-shifting treatments, many of them were more than 50 years ago. So I think of this as low-hanging fruit, where a case series, you didn't need a randomized clinical trial, but as we've increasingly picked that low-hanging fruit, we need to use another way of deciding what are the best treatments. And that enters evidence-based medicine, you're all familiar, this is the definition from David Sackett, use of current best evidence from systematic research and making decisions about the care of individual patients. And he was talking about controlled studies, hopefully prospective, ideally randomized, and systematic overviews. And clearly, our specialty is changing, this article's now 15 years old, the gold bars represent the percentage of articles in JBGS, which are case series, and the circled bars represent controlled studies, and you can see that even 15 years ago, the number of case series has dropped to less than 50% of the publications, and controlled studies are now more than 50%. So our specialty is becoming evidence-based, and my impression is foot and ankle surgery is perhaps on the forefront among the orthopedic specialties in advancing evidence. But simply giving the evidence is probably not going to be enough. I want you to find a bold and innovative way to do exactly the same way it's been done for the past 25 years. So I will suggest to you, if we're going to become evidence-based as surgeons, we need to change the way we think, and we need structures to support that change. So I'm going to start with some little audience participation. I am hands raised to, as an orthopedic surgeon, I am a rational decision-maker leading to logical opinions and correct decisions. Hands up. For those that didn't put your hands up, then you obviously are prone to bias leading to strong opinions and flawed decisions. Supposed to be a little laugh when I say that. So I'm going to take you to this well-worn example, which you've seen repeatedly, but it is really important. This is a seminal article published by a sports surgeon in Texas named Bruce Mosley, who recognized that within his city, there's quite significant variation in the use of arthroscopy for degenerative arthritis, and he took on the challenge of a randomized clinical trial. And it's seminal because, one, it was a randomized trial in orthopedics, two, it was the lead article in the New England Journal of Medicine, and three, it was placebo controlled. And his conclusion was, compared to placebo, there was no benefit. It would be an understatement to say that this was not embraced by our specialty when this paper came out, and Bruce has written and spoken extensively about the way the response to this article. It ain't so much the things we know, it's the things that get us into trouble, it's the things that we know that just ain't so. So I'm going to start with a controversial premise, and my premise is surgeons are human. So I pulled this picture off the internet. Coincidentally, all these people are drinking the same beer, and they're at a party, they all look like they're having a good time, and some of you would think, gee, I wish I was at that party. And you've now attached that beer in your mind to this party. You're a little more likely to order this beer tonight when you go for your first beer, and this is called associative advertising, and it's everywhere, and the reason it's everywhere is because it works. And why does it work? If you haven't read this book, it's written by Daniel Kahneman, who was a psychologist who won a Nobel Prize in behavioral economics. The notion that we have limited access to the workings of our mind is difficult to accept naturally. It's alien to our experience, but it is true. You know far less about yourself than you feel you do. So that association happened in your head and you didn't even acknowledge or understand, but it did occur. And the corollary is it's easier to recognize other people's mistakes than it is our own. So this is what I'm talking about. You have best practice. Here's some of the characteristics on the right-hand side, but how do we get surgeons to adopt best evidence when it becomes available? So I'm going to take you through two studies that show why this is a complicated issue. Now, this is a study we did many years ago looking at actually total knee replacement, and we surveyed every orthopedic surgeon in the province of Ontario, and we asked them, holding every other thing constant, how do the factors along the rows affect your decision making? And somewhat to our surprise and a little disappointment, we found huge variations. So for a patient more than 80 years of age, a third of surgeons said it makes them less likely to perform a knee replacement. A third of surgeons said it doesn't affect my decision, and a third surgeon said it makes me more likely. We couldn't find any characteristic that could explain this. It wasn't young versus old. It wasn't who did arthroplasty versus who did relatively little. It wasn't who had an arthroplasty fellowship. So we went back more than 10 years later, and we did the same survey again. And a little less variation, but still disappointingly, when we asked them about a patient more than 80 years of age, 25% of surgeons said it made them more likely to perform a knee replacement. 50% said it had no effect on the decision, and 25% said it made them less likely. But this time we did something different. We resurveyed the same surgeons or a subset of surgeons six weeks apart. And what we found is there was a group of factors where surgeons amongst themselves and surgeons were consistent in their opinions six weeks apart. We found a group of factors, including patients more than 80 years of age, where most of the variation could be explained by variation amongst surgeons. We found a number of factors where it could be explained by variation amongst surgeons, but also surgeons changing their opinions. And there was a group of factors for which most of the variation was due to surgeons changing their opinions six weeks apart. In fact, 50% of the variation was due to surgeons changing their own opinions. Now, I recognize this is just a survey, but it does raise the concern that depending on when you see your surgeon, you might get a different opinion. But in fact, that's not the most interesting thing. We also asked surgeons, are you uncertain about your opinions? And for the patient more than 80 years of age, only one out of 100 surgeons indicated they were uncertain. So despite obvious variation amongst surgeons and even surgeons themselves changing their six weeks apart, only one surgeon was willing to admit that they were uncertain how this factor affected or should affect their decision making. So you've seen this quote, not always right, but never in doubt. So this first study, what does it say about surgeons? Well, not accepting uncertainty makes us unwilling to accept the wide divergence in opinion. Presuming we can all be right, it makes us less likely to accept new information. If you can't even accept that you don't know the right answer when confronted with new information, it's hard to believe that you're open to changing your opinion. This is another study we did, and it was driven by understanding what we call unmet need. That is people who would benefit from orthopedic procedures but weren't receiving it. So what we did is we surveyed 50,000 people in two counties in Ontario, and we asked them, did they have any complaints in their hips or knees? For those that responded yes, we then administered something called the WOMAC scale, which is a commonly used scale in hip and knee arthroplasty, and we used a very conservative threshold of a level of pain and disability which would make them an ideal candidate for hip and knee arthroplasty. We then x-rayed the patients, we had them physically examined to ensure that the complaints were coming from the arthritic joints and they had arthritis. We excluded all those with comorbidities and we underwent a decision aid to have them decide, would you be willing to consider joint arthroplasty? And we only looked at those that were very willing to consider arthroplasty, and none of these patients was on a waiting list for hip and knee replacement. So this study made me extremely popular with my orthopedic colleagues. They all wanted to have the names of these patients so they could put them on their waiting list. But what was interesting is when we looked at men versus women, and this is despite the fact that women receive more arthroplasty than men do, in fact there was three times the unmet need in women compared to men. So despite getting more, they're still not getting anywhere near the amount that they need. And remember, these were patients who were ideal candidates who indicated they were very willing to consider joint replacement. So we wondered, could this be coming from the surgeons? Well, in fact, when we go back to the survey, we found this is actually one of the factors for which surgeons were completely, almost completely unanimous, that patient sex had no effect on their decision making. But we didn't stop there. What we did is we took a man and a woman with moderate knee arthritis. They went through a standardized patient program. That is, they were trained to present themselves as standardized patients. And they were scripted to be identical in every way other than the fact one was a man and one was a woman. They did not know what the purpose of the study was. And the controversial part is we sent them blind into orthopedic surgeons' offices. And this is where I lost all the friends I had from the previous study. So they walked into an orthopedic surgeon's office, regular appointment. The patients didn't know what the purpose of the study was. The surgeons didn't know they were study patients. Now, we needed to be absolutely sure that the patients were presenting themselves as they were scripted. So we had them go to a couple of, we had the mock patients go to some mock family doctors, and we videotaped them with a camera in the Kleenex box. And we confirmed that they were presenting themselves exactly as they had been trained. They were identical in every way other than one was a man and one was a woman. And what we found is that orthopedic surgeons were 22 times more likely to recommend a knee replacement for a man than they were for a woman. Now, the question that I always get is, well, did that differ whether the orthopedic surgeon was a female or a male? And obviously, at the time we did the study, we didn't have enough to, uh, of female orthopedic surgeons to answer that question, but it didn't look like there was a difference. Number's very small. So what does this study say about surgeons? Well, surgeons may have unconscious biases that could play a role in explaining disparity, uh, of joint arthroplasty and may need to be addressed if we're gonna, uh, ensure equal access for all. And this book is a great book written by a journalist called Daniel Gardner. Uh, once we formed a view and he talks about this thing called confirmation bias, which I'm sure you're familiar with. Once we formed a view, we embrace information that supports that view while ignoring, rejecting, or harshly scrutinizing information that cast doubts on it. And as you expect, I'm gonna show a picture of a politician. But the reason I show this is this is a pernicious bias that is present in all forms of, uh, human interaction and opinion. That which confirms what we believe we accept and that which refutes it, we harshly deny it and it's everywhere. So I'll take you back to Bruce Mosley's study because there were legitimate concerns about the patient selection that was used for this study. So it was a follow-up study done by, uh, a woman, uh, orthopedic surgeon from, um, University of Western Ontario named Sandy Kirkley who tragically died in a plane accident before this was, uh, published in a selected group of patients and found that arthroscopic surgery had no benefit compared to, um, medical therapy. There was a concern that perhaps it was, uh, for those with a meniscal tear. So there was a subsequent par- um, uh, trial done by Jeff Katz and although there was some crossover, they found no benefit of arthroscopic partial, uh, meniscectomy compared to, uh, usual, uh, treatment. There was a concern that maybe it was a degenerative menis- meniscal tear in which there might be a subset. And in a subsequent, um, uh, placebo controlled trial, again, no benefit of arthroscopic debridement for, uh, patients with arthritis and degenerative meniscal tears. So it seems like this question has finally been answered that, um, arthroscopy, uh, for degenerative, um, knee arthritis is not an effective, uh, procedure. Why is it so hard to accept that arthroscopy may not be useful? Well, patient presents an orthopedic surgeon with pain and MR shows a meniscal tear. So you see a patient with pain, you see a mechanical problem. You might do an operation going into a, uh, partial arthroscopic resection and that's very satisfying. You've got a mechanical problem, you've done an operation and the patient may seem to get better. Now why is that? I'm going to talk later today about this phenomenon called regression to the mean, which is if any of you have had back pain, you recognize it can be terrible and then without too much, it goes away with time. Placebo effect, I'm not going to talk about this, but surgery might be a very effective placebo for reducing pain. And we have to accept that there's a certain number of patients who want to please their surgeons so they may minimize their symptoms upon return. So these are all the reasons why it's so hard to accept that maybe arthroscopy, uh, wasn't successful, um, as a treatment. So to take you back to Daniel Kahneman, he calls system one this automatic, uh, phenomenon that goes on inside our heads without us thinking about it. And he calls system one provides the impressions that turn into beliefs and it's the source of impulses that often become our choices and actions. And this is where it gets really important. The best we can do is compromise, learn to recognize those situations in which mistakes are likely and try harder to avoid significant mistakes when the stakes are higher. And what could be higher than surgical treatment for patients? So I'm not going to go into this in detail, but this has become a really, uh, interesting area of endeavor, um, that you'll increasingly hear about and it falls under the category of something called implementation science. And some of the, uh, concepts are the involvement of all stakeholders in evaluating the evidence with a big push to get patients involved. We clearly need the involvement of professional societies such as the Foot and Ankle Society to get engaged in how we take evidence and get physicians to change their practice. And the whole idea of engaging patients and doctors through social media is an untapped, uh, area that we need to, uh, use, uh, more deliberately than perhaps we have in the past. So I thank you very much for listening. Doing the right thing for patients may require us as orthopedic surgeons to undergo what might seem like some difficult changes. We don't like changing. I would suggest to you that changing surgeons' behavior is a complex task and I think that it's almost substantially influenced by bias and heuristic. Having said that, real change can only come from surgeons and from professional associations. And finally, I would conclude that surgeons may be less human than other humans, but we are still human. Thank you very much for your, uh, listening. We'll have some questions. Oh, yeah? Okay. Thank you. Great talk. John Ketzer, are you here? Uh, John is, uh, vice chair of the, uh, research committee and, uh, we'll have an opportunity for, uh, some questions for our speaker. Um, if you feel free to step up, step up to the microphone. Um, I'll, I'll start off, um, we have a record number of attendees here at the meeting, 1,300 or something like that, over 1,300, many, uh, young surgeons and fantastic, uh, exhibitors, new techniques being introduced in the symposia, new techniques, new implants. How, how do you advise, uh, a young surgeon who's looking at a way to improve their practice when they're introduced to new techniques and devices that have great results initially, but we don't have evidence, long-term results and evidence that you spoke of? How do you advise them? So, so thank you. Um, it, it, um, is an introduction to the talk I'm going to give a little bit later on today because evidence-based practice means all of us doing the same thing based on the best practice. And yet there's a real need for innovation, which is doing something different. So how do you reconcile those two, um, views? I call it the innovation cycle, and I'm afraid not all of us can be innovators. Um, there was a time where, uh, the fields were so ill-formed that I think there was a need for many of us to innovate. But I think innovation needs to reside with a relatively small number of people who, um, start with a pathway that may be a case series, uh, which is a proof of concept, then leading to, um, perhaps a few centers, uh, coming together to do prospective comparisons. Um, there was a recent article in JBGS from the Canadian Foot and Ankle, uh, Association that, uh, looked at a retrospective. They, the information was collected prospectively, but they then looked back retrospectively. And then for something that's really important in terms of the cost or the number of patients or the health impact, I think you then need to go on specialty-wide randomized clinical trials. So the short answer to your question is I don't think we can all be innovators, and I think people, while it's very tempting to see something really cool and new, I think we, um, largely have to, um, accept that there's only going to be a few innovators, and I think in many cases those reside in the academic centers, and I know that's a bit controversial. Rick, thanks for a great talk. This might be a little redundant for the last question, but you were able to successfully randomize people into fake surgery and not fake surgery, which is a dramatic thing to do and very, very difficult to do. And as someone who's failed to randomize patients in a few trials, I wonder about your comments on prospectively non-randomized cohort data that's built into the EMR as an alternative. What volume of patients do you have to get to to minimize the confounders that you're not controlling in a randomized trial? So I absolutely believe that there is a spectrum of evidence, and I, well, much of my career has been based on, as a proponent of randomized clinical trials, and that was partially because it needed to shift the profession. But clearly all forms of evidence should be informing orthopedic surgeons in their decision-making, and I personally believe that even retrospective good data can give very useful information for clinical decision-making. And there's a lot more sophistication in which prospective studies where bias was a real problem. To answer your question, placebo-controlled trials are very narrow in where they can be used, but there's a recent trial looking at shoulder decompression surgery in the UK that was placebo-controlled that, again, showed a surgical procedure which is very common in the UK, at least, did not help patients. So there are some very narrow indications. So I think there is definitely a place, and we can't all do randomized clinical trials. Not every question can be answered for clinical trials, and I think there's a real place to move the profession along with prospective non-randomized trials. But you need to have methodologists, you need to have biostatisticians, you need to get a team involved just like you do a randomized trial. So a big place. Yeah, Bill Granberry from Houston. And Bruce is a partner of mine, and I was actually a resident while he was doing that study. And there was a patient, I remember, who came back into the clinic who had been randomized into the placebo procedure. I mean, he had the trocars put into his knee and then a cortisone shot. And he did so well from that, he wanted it on his other knee. And we tried to talk him into a scope because, you know. But the point of that is that I was at the AAOS meeting, and there were several articles that are coming out on these patient-reported outcomes. And there were several on bunionectomy. And the interesting thing was that there was very little correlation between what we think of with our AOFAS scores or our bunion scores or whatever and the patient-reported outcomes. And, you know, we talk about moving forward about, you know, making the patient happy. Just, I don't know, some comments about that because I think we haven't figured that out or correlated it. So there's two comments there. One is, how do we determine the treatments we're giving are actually effective? And as surgeons, I think that's our primary question. But of course, a lot of these PROMs and PREMs, as they're called, are being used for other reasons. Driving physician compensation, driving payments for accountable care organizations. So I don't think the PROM and PREM issue is solved. And of course, a PROM and PREM is, on average, every patient has their own issue. So I would say to you, it's a complex issue and I don't have an answer to you other than we need to continue to try and define how we decide that treatments actually make patients better. On your first one, it's a fascinating observation. And the most compelling example I can give is one of the first placebo-controlled trials was in the 50s. And it was related to internal mammary artery transplantation into the heart. So what they did is they cracked the chest, they took the internal mammary artery and they just sutured it into the myocardium. That was before bypass was even possible. And they did a placebo-controlled trial where they cracked the chest, they opened it up, and they closed it back up again. I mean, it's just astounding that that, even in the 50s, that that happened. And what they found was that the procedure was ineffective. But that wasn't what was interesting. What was interesting to me is in the placebo arms, six months after the procedure, people had sustained improvement in their angina. So I would say to you that surgery is a very effective placebo, and that gets really complicated. Because in general, you know, simply putting a trocar in and giving someone a cortisone shot, it gets really ethically complicated. But I think we have to accept that there's a big placebo component to what we're doing, which is putting a patient to sleep and making an incision. We have a quick question in the back. Sure. Hi. Tim Daniels from Toronto. Jim, you know the knee arthroscopy saga quite well, particularly in Ontario. We've known for 12 years that it doesn't work. But implementation, there's been a lot of financial consequences. I mean, there's a lot of orthopedic surgeons that actually make a living off of doing knee arthroscopies. And the government, it seemed to me, was worried about alienating the orthopedic community by coming down in a draconian way and saying we're not doing this anymore. Have you thought about how we get around the financial conflicts with this knowledge? Yeah. So I think we have to accept that this does have financial implications. And that's where professional societies, I believe, I'm quite proud of being a member of the AOS, who's come down on several procedures which they don't feel are evidence-based. But clearly, there is a lot of financial gain. Interestingly, Tim, you just recently, the week before last, what was called the Approprious Working Group, which is a bilateral group of the government and the Ontario Medical Association, has said that the government should no longer be paying for knee arthroscopy with degenerative knee arthritis. So my personal belief is we shouldn't be coming down with these hammers that come from fee schedules and governments. I think that this is our responsibility as a professional association to stand up and actually implement when we think we're not helping patients with procedures. And there's so many things that we do that are so effective, and I've taken some grief for this, but I don't believe we should be doing things which are of no benefit or harmful to our patients. Maybe I'll end there. Thanks. Thanks, Harold. Thank you very much.

orthopedic surgeons

evidence-based

neurosurgery

life-changing

cost-effective

controlled studies

surgical decision-making