 It's a pleasure now to introduce Dr. Gretchen Schworzy. Gretchen's an associate professor of surgery at the University of Wisconsin. She is the holder of the Morgridge Professorship in Vascular Surgery. Gretchen's nationally recognized expert in surgical decision making with a focus on improving communication between older patients and their surgeons. And she has a particular interest in aiding patients and doctors in making medical decisions which support patient preferences, values, and goals. About 10 days ago, Gretchen gave the very prestigious John Jay Conley lecture in ethics, I'm sorry, in philosophy and medicine at the American College of Surgeons Clinical Congress. And it was outstanding. And I think anyone who is there knows what a wonderful speaker she is. So it's a real pleasure to welcome Gretchen back. Thanks, Peter. OK, hopefully the acoustics are all right here. Any two seconds? All right, so wow, this is a tough group. That was a really extraordinary presentation by Alex. I haven't seen anybody tease out the realities of the collaboration that happens in the operating room. So it was brilliant. I'd like to both acknowledge my funders and express my frustration with my husband who has invested tons of our money in this meslite business with absolutely no return. But somehow it needs to be on my disclosure slide. And in the Shorzi lab, we have three rules. And the first rule is study what pisses you off. The second rule is follow the data. And the third rule is don't piss off the surgeons. So today I'm going to talk about some work that actually began here many, many years ago after a question that Rachelle Benackie asked at the end of one of my talks. And it really goes along the lines of this issue of follow the data. And Rachelle stood up at the end of a talk. And I honestly don't even remember what I was talking about at this point, but she said Gretchen, why is it so hard for surgeons to withdraw life sustaining treatments on their post-operative patients? Is it just because of their outcomes and their mortality statistics, or is there something else going on? I was like, wow, that's a really interesting question because it seems like it has an obvious answer, but it doesn't really ring true. And while I'm not going to talk about surgical buy-in and all the other stuff that followed afterwards, I am going to talk about what we did subsequent to our exploration of surgical buy-in, which is that we started audio recording conversations between surgeons and patients because surgeons had told me that they had made a contract with patients. That when we talked to patients about surgery, it wasn't just about surgery, but all the other stuff that came afterwards and they had agreed to it. And I was like, huh, I'm a surgeon. I don't really do that. What are we doing? And so when we started putting audio recorders in the surgical clinic, trying to listen to these preoperative conversations, we came up with some really interesting data. And indeed, surgeons don't really make this contractual plan with patients, but on top of that, after we listened to these conversations, we interviewed patients both before and after their operation. We found that patients had very different impressions of what was going on in these conversations as well. And when we talked to them, they said things like this. When I went and talked to the surgeon, I felt that surgery had to be done. There really was no choice. We did it by and large because he said, well, it needed to be done. And there weren't any other options. They also noted that recovery was much harder than they had anticipated. And now, even though the surgeon had disclosed multiple disarticulated risks of the operation, they were pretty shocked that the recovery of surgery was so challenging. And then finally, they were very clear that there was never an opportunity to talk about what would happen if things went poorly. That especially in cases where we did interview family members for patients who had very severe complications, including death, they really didn't know what to do after surgery because it'd never been addressed before surgery. And this is what the state looks like. This is a neurosurgeon in one of our earliest studies, and he's talking to a patient about tumor resection. And very appropriately, he goes through all the risks of surgery. And I would say he does everything that we've been taught. What questions do you have for me? And the patient would say, well, washing my hair. And the doctor very appropriately says, well, you can use shampoo, but don't scrub the stitches. And the family's like, oh, are you gonna use stitches and staples? And what we noticed when we saw all these questions that patients were asking is that they wanted to ask questions, they weren't quite sure what to ask. And in my lab, this gets coded as logistic and technical concerns, that while these questions may seem really important in the moment, they actually don't help patients figure out what are the options, what's it like to have surgery, and what's the recovery gonna be like, and what would happen if things went poorly? How might you know my wishes? So we sat down with a group of patients and families over a year went through all of these transcripts looking at the patient-doctor conversation. And what we came up with was an intervention that's called a question prompt list. It's pretty simple, it's a brochure, it's a little trifle brochure, and you can see that the brochure has three elements. The first element is really this issue of options. And it's not just what are my options, but what is surgery gonna do for me? In your opinion, do you think surgery will help me live longer? In your opinion, do you think surgery would help me feel better? The kinds of questions you might wanna ask your surgeon in order to figure out if surgery is right for you. The middle panel is really about what to expect. These are expectations when things go as well as we could hope for. Am I gonna need lines or drains? Is this gonna change my other health problems? What kind of care am I gonna need after surgery? And then the last panel is really about advanced care planning. How will you know my wishes if I can't speak for myself after surgery? So we applied to the government and they gave us a ridiculous amount of money to study a paper brochure. And this is our conceptual model of what we were trying to do here. The idea of our intervention is that prior to surgical visitation outpatient, we would have the surgeon send the patient and their family a letter. And the first is a letter from the surgeon saying, surgery's hard. And when you come to see me, I want you to ask questions. And here's a list of questions you might ask. And we included the brochure. And the idea in the literature is really this idea of patient activation. When you prepare patients for decision making, they may have a better time. So our primary outcome of our study was what questions did they ask? Do we change the questions from these ideas of logistic and technical concerns about stitches and staples and shampoo to questions about should I have surgery? What should I expect if everything goes well? We also had some other outcomes, including surveys of patients and family members after they met with a surgeon, which are really sort of how self-efficacious do you feel in your patient-doctor relationship. And then finally, we used a measure of shared decision making, which is observer measured, meaning that we audio recorded all of the conversations from patients in our study and then used this observer measure of shared decision making to calculate whether we improve shared decision making. So we did a multi-site randomized clinical trial using something called a step wedge design. And if anybody wants to talk about a step wedge design, I'm more than happy to do it with several vodkas in hand. It's really a very painful way to run a study, but the idea is that you randomize surgeons over time to when they're gonna take on the intervention. So every surgeon is in a pre-post study, but their pre-section is gonna be a little bit different than their post-section dependent on your randomization pattern. So we did this at five sites across the country, Oregon, San Francisco, Wisconsin, Boston, and Rutgers in New Jersey. And we had 40 patients, or sorry, 40 surgeons enroll in our study, and they all performed high-risk surgery, so neurosurgeons, cardiac surgeons, vascular surgeons, cancer surgeons, colorectal surgeons, GYN oncology, and urology. And the patients we enrolled in our study had to be at least 60 years of age and have one or more comorbidities. And I'm just gonna give you a little spoiler alert. It's a totally negative study, which now I need a support group for people doing interventional studies. But it's really sort of very interesting to look at the data. So when we looked at these questions that people were asking, we did see some movement in the questions with our intervention. So I'm gonna go just talk about this first outcome of question asking. So when we looked at options questions, you might note the light gray bar is the intervention group, the dark gray bar is the control group, and that we were able to very gently move the numbers of questions asked, reducing the zeros and increasing the two plus. You'll also note that some of the p-values I'm gonna show you look statistically significant, but they're not. And again, this death by math thing is really very frustrating when you run a study. If you study multiple outcomes, you have to use a Bonferroni correction, which reduces your p-value to the point where we did not see statistical significance. But I think what's most interesting is I run you through these questions that people ask. These are expectation questions. Is how many patients didn't ask any questions about what to expect after surgery, or what options they might have besides surgery? And here's the issue about risks. Very few ask questions about risks at all. So even though this was a really frustrating study, one of the best things about the study is that we audio recorded all of these preoperative conversations, and we have real information about what occurs since we enrolled 450 patients into our study. This is our issue of shared decision-making. So this option five score is a score invented by Glenn Elwin at Dartmouth, and it's a nice way to try to discern whether shared decision-making is occurring. We did see some movement in this score. I'm not sure how great six points is. Again, the Bonferroni correction makes this non-significant, but it's interesting this idea of thinking about what are the domains of shared decision-making and how might we measure it? Because when you survey patients and clinicians after encounter about shared decision-making, you get very different results than you do when you actually count up the domains of shared decision-making and see if there's a real difference in the moment. So I'm gonna talk just a little bit more about the question prompt list and then talk about the exciting data we found with all these audio recordings. So is this enough to consider routine use of the QPL? So it's pretty low risk, it's very low cost, and we found that it didn't change the visit length. It's really just a piece of paper. But I think the most important thing that we found is that questions about options, expectations, and risks are not often asked, and you might note that I didn't even show you the advanced care planning data, and that's because we saw only 13 questions about advanced care planning in an entire study of 450 old people considering major surgery with multiple comorbidities. So that is a huge problem to not think about advanced care planning at all in these conversations. I'm not sure QPL's gonna solve that problem, but clearly it's a big issue. The surgeons in our study, we sent out 6,000 of these QPLs prior to the clinical visit, so we didn't know who would be enrolled, so we sent it to all of the surgeon's patients, and the surgeons really liked it. They really wanted to continue using it even after the study, and maybe we made some movement on shared decision-making. So what else did we find? We found that there was huge variability in the use of shared decision-making in our cohort. In fact, there are really some high performers in the surgical cohort by surgeons who really do actually spend quite a bit of their effort in the pre-operative conversation doing shared decision-making, and then there's a very clear group of low performers. Remarkably, the surgeons, the high-performing surgeons have great variability in their use of shared decision-making, and they seem to employ it selectively, either when there's real clinical equipoise for something like a uroostomy versus an ilial loop conduit. There's clear notions about patient preferences around those kinds of operations, or when patients were very, very high risk, and they were worried that the outcomes of surgery were suboptimal. The other thing we found is that the patterns of pre-operative communication are incredibly robust, and I think all of us do this incredibly consistently, that we start with this explanatory phase where we talk to patients about their disease and we talk about their treatment. We say things like, here, this is you, slice-like bread, this is you in a CAT scan, and this is your liver, and this white spot in your liver is a tumor, and then we say, and I'm gonna take it out. And I think that the way we've been taught to describe surgery and its consequences and to get informed consent very much revolves around this notion of generating patient understanding of their disease and the treatment, and the problem with that is that we use a lot of fix-it language about what the problem is and how we might repair it. As a vascular surgeon, you might imagine I say, here, here's your blockage, and I'm gonna do a bypass and go around it. We use this language again and again and again, and then we step back in the second half of the conversation and we say things like, this might not be the right problem to fix, or maybe it's too broken to fix, or I can fix it, but you're never gonna be the same because once your stomach is turned into your esophagus, you're never gonna be able to lie flat. And I wonder if we're not setting people up for real harms because it's very challenging to think through that deliberative part of the conversation. Once I've told you this is the problem and this is the operation I have to fix it. On top of it, the other thing we saw in our data is something that looks much like this, and I'm just gonna put it out there that we had a competition to find the worst case of this and maybe this one, one. But this is actually pretty typical of our data, is that the surgeon does this fairly long monologue, trying to describe to the patient what is gonna happen with surgery. And it gets a little bit to this point about transparency and what do people need. But this is a surgeon who is working really, really hard to describe a Whipple operation, which is a removal of a pancreas cancer. And he's going on and on and on, talking about how this is just like plumbing, and now I'm going to describe the plumbing to you, what I'm gonna sew back together and where's the gastro-judgenostomy. And at the end, he says, you know the worst thing that can happen is you can die. I am not sure how patients can use this type of information in order to figure out whether surgery is right for them. It's as if the plumber came to your house to fix your toilet and the plumber spent tons of time explaining all the little gizmos and the things in the back of it and what were broken and what he was gonna do in order to fix your toilet. And never told you how well your toilet was gonna work afterwards, how long it was gonna last, and what it was gonna cost you. Because at the end of the day, as much as I can imagine, some people wanna understand how their toilet works, not everybody wants to know how their toilet works, but everybody needs to know how much is it gonna cost, how long is it gonna last, and how well is it gonna work. And if we spend most of the preoperative conversation doing stuff like this and forget to do that other stuff, we've really misled patients. And so when we went through our data and we looked at how often surgeons talked about the goals of surgery, what was the aim of the targeted objective of the operation, we found that over 50% of the time in these preoperative conversations, the only goal mentioned was to fix the problem. And the problem with this is that if you tell patients that this surgery is gonna fix your problem, then they will attribute whatever they want to the fix of that problem. Many years ago, they did this brilliant study in the UK, which I guess they could do because people sat around waiting for their carotid endarterectomy, but they interviewed patients before carotid endarterectomy, and the only thing that operation does is prevent stroke. And they interviewed patients and they were very good at describing the risks of the operation. And then when they asked patients what they were gonna get out of the operation, they said things like, my memory is gonna get better and the ringing in my ear is gonna go away and I'll be able to see better and my headaches will be better. The problem with using this fix it notion of what we are doing when we're doing surgery is that people will attribute whatever they want to the fix without understanding the real reasons for pursuing surgery. I suspect many of you have seen this video, but if you haven't seen the video, I'm not gonna play it. It is hilarious. So if you Google orthopedics versus anesthesia, you will get this video and the little blue bear with the blue stripy shirt is the orthopedic surgeon and the red and white stripy shirt is the anesthesiologist. And the orthopedic surgeon says, I have a fracture. I need to fix it. And the anesthesiologist says, where's the fracture? And the orthopedic surgeon says, it's in the emergency room. And it goes on and on and on and on. And the reason this is funny is because it smacks of truth. And it's not just because it's so easy to make fun of the orthopedic surgeons. It's because we all do this. We're working really hard to generate understanding of disease and treatment. And I worry that we have taken this way too seriously and misled patients and don't actually supply them with the kind of information they need. So the next step in my lab is to try and revise this pre-operative conversation, but we're not there yet. Anyway, that's all I have to say. I wanna thank my amazing lab. This takes a ton of people to get this kind of work done and I'd be more than happy to take any questions. Ah, Giuliano. Hi, Giuliano Testa from Bailor d'Alesson. This is a great thing. I have a couple of little problems. One is probably, there is a lot of male mentality in surgery. And that's when I talk to my wife, she has a problem, and she says, I'm gonna fix it. I don't want you to fix it, I want to listen to me. Which already is a problem in itself. After so many years of doing surgery, and what I'm gonna ask you is, my impression is that no matter how simple you put it, no matter how many data you put on the table, no matter how long you explain, the patients very often don't wanna know many things. They really won't have the problem fixed. I mean, that's at least has been my experience as a surgeon. What would you propose with a different way of approaching this? You're not gonna approach a problem of the toilet telling somebody it's gonna cost you a lot of money and not telling whether they really need the toilet fixed, which is the bottom line of it. So I'm just gonna put it out there that Giuliano is a liver transplant surgeon and I am a vascular surgeon. Giuliano fixes things in a way that I just rearranged the deck chairs on the Titanic. So I actually think that there are things that we could say to patients that any patient could understand and that actually really they need to hear in order to figure out whether to have surgery. And so it's two things. One is talking about outcomes. Surgery does four things. Maybe live longer, feel better, make a diagnosis or prevent disability. If we don't mention one of those things in our pre-operative conversation, we've really failed people. And the other thing that I think people can completely understand is this idea of a trade-off. That surgery might help you live longer but there's a real trade-off. Even in this Whipple conversation, there is a real trade-off and some of that trade-off is a risk that you might die but another part of that trade-off is that you are never gonna eat the same again and that you may have gastrointestinal distress in a way that makes it a little bit harder to live longer. And I think the problem is that we haven't distilled this conversation into this space of what are we trying to accomplish and what are you willing to trade-off to get there? And I don't think talking about the risk benefits and alternatives is doing that for us but I think we could have a different conversation that would get them there that everyone could understand that is not about fixing problems but it is about achieving certain goals that people are willing to actually expend a fair amount of time, physical distress in order to achieve. I don't know whether it is because I did the fellowship or because I've always been this way but I think I approach the conversation with my patient exactly as you say but so you think it's a widespread common behavior from surgeons to ignore the second part? So this is the best, no, so I think surgeons do the second part but this is the best thing about showing up with data I have 450 conversations that look like that. Okay, Peronner, Chicago, soon to be Kalamazoo. Gretchen, this is amazing work. I think you're really revolutionizing the whole concept of informed consent, taking it away from sort of the mechanical procedural aspect which many people tend to fixate on into what is really being offered and the real benefits and risks. I think the data you presented of that turn that the surgeon take is incredible that somebody could talk so long without having any feedback from the patient and it really points to the fact that people really aren't having conversations. They're having monologues and I think one of the basics things is people will need to learn is how to have a conversation in which there's turn taking and which there's a back and forth. Yeah, I mean I think we're working really, I mean so first of all it's easy to pick on the surgeons A because I have data and B we probably bring it on ourselves a little bit but I think other people who have audio recorded conversations between clinicians and patients very much see this sort of same notion of this is your problem and this is what we're gonna do to manage it. I think it happens in the ICU for sure it happens in oncology clinics and I think that oftentimes when I'm giving this talk I say things like I blame the bioethesis and I don't mean the people in this room I mean sort of bioethics as a group that came up with this brilliant idea of informed consent in the 50s and 60s and then never did the empirical work to see how it was playing out in clinical practice in order to sort of revise and innovate and make it better because I do think that what we're doing is exactly what we've been taught to do generate understanding of disease and treatment and that I think is very flawed. I don't have a problem with the second part of the conversation I just think that it's what are under the bridge at that point. Well it also brings up the question of I mean the people have got very technical and I think a lot of surgeons consider them are very you know they're good technicians and they know how to do the operation but questions about you know what the values of the patient and will they actually benefit and what ways will they benefit those kinds of questions very often the surgeon don't even ask themselves so I think to ask them to have that with the patient I think is difficult. So Daniel point about what is the benefit of this operation is a really important one and I'll often start an operation saying to the med student or the resident why are we doing this operation and they'll say something like there's a blockage it meets criteria the ABI is low and oftentimes I will have to ask that question three or four or five times to get to the point of make the patient feel better help the patient live longer prevent limb loss you're a spot on about that. Thank you. Hi I'm Neil Fine I'm a plastic surgeon at Northwestern and I enjoyed your talk. I think though I didn't hear you say specifically some ideas about people understanding probability because when my patients were mostly women with breast cancer are asking questions like when what's when will I be recovered or how will I do? I find that I have to try to tell them well I can tell you what's most likely are common but I will have a hard time telling you exactly what's going to happen to you. I do this operation hundreds of times and I will get a very diverse answer from my patients about how they're recovering and how they're doing. And so the whole idea that someone wants to know for instance am I going to live will I die in this year? And you say yeah you're gonna die this year but I didn't I live for 10 more years so you lied to me versus you can live for 10 more years but I died next month. Yeah so I'm gonna direct things. Yeah so I mean I think this issue of uncertainty and how we describe that with patients is really interesting and very important and I think we shouldn't hide behind it as a reason not to do things. I'm gonna send a little Elizabeth Warren like here but I have a plan for that. That is my surgeon mediated intervention and also equally challenging to study but I have a different intervention for that. Well I talked to them about what I tried to tell them is you're gonna ask me this question and I'm gonna tell you what's most common and I'm gonna try to tell you what the ranges are. Yeah so I'm gonna put a plug in for my other intervention it's called best case worst case. Thanks.