 Good afternoon. On behalf of the McLean Center for Clinical and Medical Ethics and the Center for Health and Social Sciences, Dr. David Meltzer and I welcome you to the sixth talk in our 2018-19 lecture series on improving value in the US health care system. It's my pleasure now to introduce our speaker for today, Dr. John Wasson. Dr. Wasson is an emeritus distinguished professor of medicine, community and family medicine at Dartmouth and formerly had been the Herman West Professor of Geriatrics at Dartmouth Skyzell School of Medicine. At Dartmouth, Dr. Wasson has led a number of outstanding programs, including outpatient services at the Veterans Administration, the Center for Aging, and the Dartmouth Northern New England Primary Care Research Network. He also has been national co-director of the Institute for Health Care Improvement in their Idealized Office Practice program. One of Dr. Wasson's most famous papers I show it behind me and most influential was published in JAMA in 1984. The paper, as you see, is entitled Continuity of Outpatient Medical Care in Elderly Men, a Randomized Trial. So far as I know, this paper is the first and still the only randomized trial that assesses the importance and value and outcomes of doctor-patient relationships. It's quite an extraordinary paper looking in a randomized way at over 700 patients. Dr. Wasson's current area of focus is in personal information and communication technologies. He's particularly interested in the development and testing of howesyourhealth.org, which is a family of health IT tools aimed at improving communication and health care quality by patients, doctors, and communities. Another one of his howesyourhealth.org web tools are now used nationwide to improve communication and health care quality. In 2006, Dr. Wasson received an award from the Agency for Health Care Quality and Research as a pioneer for practice-based research. Additionally, since 2014, Dr. Wasson has been working with many groups and organizations to increase personal and professional health confidence so that everyone can say, quote, I am very confident that I can manage and control most of my health problems and concerns. As you see, Dr. Wasson's talk today has the somewhat unusual title of organized crime and disorganized health care discuss. I'm looking forward to the commentary and discussion. Dr. Wasson. Thank you very much. While I'm turning on the mic here. Continuity of care. I'll just digress for a moment on this. Is a fascinatingly powerful activity in clinical practice. And so many of the interventions that we think have had positive impacts in care actually under the surface depend on continuity. We often falsely in a false positive way say, well, if you train nurse practitioners to do X, or if you do something else to do Y, look at the great results you get. For example, think of one of the earliest studies, management of heart failure. And everyone got all hung up on the methodology. But in reality, the study really was just a reputation of continuity. Because one arm got usual care. And the other arm got people to contact the patients. A form of continuity. So I just bring that up as shall we say an important point to keep in mind when people start fighting and arguing about these methods that improve care. Look under the surface. If they haven't really controlled continuity for continuity, it's more than likely the major cause of the outcome that's positive. So today's talk is not a typical headline that you're probably used to. But I'm hoping that you'll see a method to my madness by the end of this exercise. And in particular, that you'll see it's both ethical, considering the McLean Center, and practical implications for your practices, and your patients, and the population at large. So let me start in the turbulent year of 1971. This is the year I graduated from medical school. Pentagon Papers were published 500,000 people marched on Washington. And you probably don't know this, but it's the year of the longest longshoremen strike in the United States. And it was the same year that Joe Colombo was giving a talk because he felt Italians were being misrepresentatives of mafia. And he took a couple bullets in the head. By the way, in 1971, it had been several years since I'd stopped working for the mafia. So why do I have here a picture of break and bolt shipping and container ships? The answer is, since World War II, it became obvious that container shipping was a better way to go. More efficient, cost effective, et cetera. And yet, break and bolt shipping was still fighting that type of modernization and mechanization. This is an old story. Weavers Guilds fought factories for clothing. It's an old story. Sears tried to fight Amazon. The taxis in Chicago have tried to fight Uber. The point I'm trying to make is that pressure like this happens. And this particular example and the examples I just outlined constitute a term that I'm going to throw out today. And hopefully, you'll find it useful, at least for the purposes of this discussion. I'm going to call it anarchic intelligence. Anarchic intelligence implies and is based on the principles that intelligent devices and processes can reliably support and sometimes replace human effort. In addition to that, the devices and the processes need not be machine-based. But most importantly, to be anarchic, they are intentionally designed to be anti-status and usually disruptive. I wanted to put this term up because we all think of AI as something that is particularly destined to change the world. It isn't. It's a term that's been around since the 50s. And although anarchic intelligence can include AI, AI often is not anarchic. And in fact, I'll give an example in a few minutes to show how we often use AI to maintain the status quo or enhance our own incomes, et cetera, rather than try and undermine them in a positive way. So back to Joe Colombo with a couple of bullets in his head. This guy in the picture is connected to Joe because he got all his guys, his long shorman, to go to Joe's rally. And that increased this guy's prestige. And he was all about prestige. Just read about him. Lecturer at Harvard, considered for US Secretary of Labor, leader of the International Long Shormons Association, and of course, a guy with many high-pice political friends. And by the way, he was a Gambino guy. And I knew Joe. OK, so on to the next. So why am I juxtaposing these rather odd topics? Let's just look at the points I'm going to cover today. From Joe's point of view, on the left, here are the three points that always matter from an organized crime point of view. Successful parasites don't harm the host. Oh, by the way, it's Tony. Joe was the guy who took the bullet. Protect the brand. Use barriers to entry, some of which might be forceful. Adapt quickly and avoid anarchy, if at all possible, so that you keep the peace and stay out of the news. On the other hand, US health care must also try to serve the host, preferably through science. It also needs to protect its brand, preferably through value. And it should adapt quickly. And I'm going to show some examples of anarchic intelligence where it could adapt quickly. But currently, I don't know if we're adapting in the ways that we should. How are we treating our host? Not very well. Currently, we're extracting nearly 20% of the GDP from the population at large. Experts say 30% of that extraction is waste. We've created analyst market inefficiency, so that compared to comparable countries, we have twice the cost for procedures, drugs, administrative services, and physician payment. Medical education in this country is longer and much more expensive. And there are huge variations, in spite of our claim to be looking out for quality. And outsiders who look at us say, my god, they're just huge quality tasters. This picture illustrates how our parasitism has gotten totally out of hand. Yellow clearly indicates that if you ask yourself, as we moved in the 1980s to becoming corporate medicine using more and more administrators to make our small cottage industries, as it was called then, more efficient, what do we really have to show for those extraordinary expenditures in yellow? By the way, notice this timeline starts is nine years out of date. Things are much worse now. Speaking of worse, here's an interesting little analysis I did recently. I looked at the New York Times and Wall Street journals for health care corruption. And in the early period, you can see what it was like pre-conglomeration and corporatization. And you can see how it's gone up five to six fold since the 1980s. Most importantly, we're getting, although we may not becoming more efficient at adding value to our health care system, we're certainly being much better at extorting a fair amount of money. Notice how the median now of the stories is 30 fold greater than it was then. This is what the population is seeing every day. And this is not enhancing, as Joe would say, I mean, as Tony would say, our reputation. And speaking of reputation, I don't want to hurt the feelings of any lawyers, business guys, or insurers who might be in the audience. But doctors and nurses are hanging around with these folks as we become more and more conglomerate. And notice, where do we have to go on the reputation scale? Depends a lot on who we hang around with. We should be protecting ourselves. Now, any good corporate person would say, well, we have kind of a foul reputation building at this point in health care. Let's do something about it. Let's hire some consultants and come up with a damn good model. And Dartmouth has done that. Look at that. We are more powerful than medicine. Whoa, huh? Is that slightly hubris? But then again, you guys are at the forefront. OK, so let's put this in the world context. This is for 190 countries across the world. These are per capita expenditures on health care. As you all know, this is based in 2016. Currently, we're up to 11K per capita. It's gone up nicely in the last two years. You can see, as you all know, we're out costing every other country in the world in per capita health care costs. Our estimated waste alone is out costing a huge number of countries. Our administrative costs are above the median. Our estimated corruption is also above the median. And even office quality measurement, not counting hospital measurement, but the quality, the cost of measuring all that stuff, all those 500 measures and caps and this, that, and the other thing, are costing more than a lot of countries. You can see the higher ranked countries on the right. I just put positionally, Mexico and UK. And Cuba is an example below the median. So I don't know how that happened. Let me get there. OK, so sadly, this is not new news. I don't think to anyone in the audience. And I don't want everyone to get depressed at this point. But many here in the audience and elsewhere in the world have tried very much to move the gas-guzzling jalopy we call health care over the hill toward a more promised land. Unfortunately, the mountain of waste, corruption, cost, and market inefficiencies has made it really, really difficult to do. So the question is, for those of us in the audience today, what can we do? And I'm going to put forward the idea of some anarchic intelligence approaches that you can leave here today and use in your situations. But before I go on, are there any questions at this point? Any clarifications on sort of the bad news before we move on? Any? Nothing? All right. So I'm going to move to an example of anarchic intelligence that I would call a win-win. This is one that really I feel we should be testing. And in fact, I'll point out that a Medicare review committee feels we should be testing this too. So for any of you who want to start a study, this is the way to go, I think. So it works on chronic condition management. Now just to put chronic condition management in context, of the $10,000 per capita we spend each year in health care, it is estimated that about 75% of that money goes into chronic care management. So a huge percent. And any improvements we can make there puts a lot of money in play, right? 7,500 to 10% improvement is a big amount of money. So you're not saying that we spend 75% of health care trying to think about management of chronic conditions as a discrete result of treating catastrophic health But you're saying 75% of health care strategies is in people who have chronic conditions? Yes. Yeah. Sorry. Can everyone hear that? 75% is in people with the chronic condition, of which a fair amount has to be managed. Now our current method for managing this is what I'd call the top-down model. It starts during medical school where we get a whole bunch of cognitive stuff, a little technology, some supervision out in the practices. And out of all this stuff, we're supposed to figure out and apply well without a lot of variation among us management of patients with chronic conditions. And the way we can do that is through perhaps something like a model that says, oh, well, this would be the best practice if you took all this stuff that's floating around in your head and apply it through, for example, the chronic care model, a miracle will happen. And below that is indeed the chronic care model. So in summary, if you take our smarts and put it through the chronic care model as one example, in the real world, we would expect that costs would be reduced. In other words, in the jargon of today, there would be a bend in the cost curve. And most importantly, the patient outcomes would be improved. There'd be greater value. Unfortunately, the evidence isn't very strong that this has worked at all. And the question is, the comment I would make is that these results shouldn't be surprising if you just think about the fact that at the top, we really have a model. We have people with many different views on how to apply things, putting their inputs in. We have high variation, in other words, at the top, and deliver it in many different ways. So the top-down model, in essence, is a top-down model that results in some real mush at the outcome set. Now, although we can't prove easily that any of this stuff is working well, what we can see quite easily are the unintended consequences. It has inherent variation at the inputs. It's grafted onto a wasteful, increasingly corrupt and very costly health care. It's, in essence, a Band-Aid on our usual care. It stimulates large-scale top-down reimbursement schemes. Instead of front-line innovation, you hear it all the time, oh, if we change insurance this way or if the employers do that, something great's going to happen. We get these top-down dictates. It catalyzes quantifrenia and the measurement industry. It undermines smaller primary care practices that have been generally very cost-effective, and it fosters more corporatization and conglomeration and additional categories of workers as we keep adding Band-Aids to a broken system. We have so many different types of workers now, all of whom are just adding to the bottom line. This likes to jump. Whoa. Worst of all, I think, is in addition to the adverse impacts of all of this on society at large in the budget, the real victims, the ultimate victims, are at the front line. They're the clinicians, many of you here in the audience, and your patients. We often hear, and I hate this term, that we are burning out. Burnout sounds like we lack resourcefulness. We can't figure things out. We're burning out because it just, I can't find it. That's not an issue. I think we're pretty resourceful. The problem is we're experiencing what some have called moral injury, just like World War I, when it was first described, and I'll just quote it, from continuously being unable to provide high quality care and healing in the context of our healthcare mess. You get up in the morning and you just don't feel you can do the right job, and the number of doctors I sit next to and regale me with this type of story is depressingly increasing each year. However, the good news is, compared to the Longshoremen in 1971, we have two advantages, I think. Advantage number one, we still have some public trust, and number two, we have a science ability behind us that we should use more and more, rather than less and less. To use the good old jargon of the mafia, we should use our science instead of being stooges for powers like Tony, who manipulate us into doing things, and we just sit back and say, oh, poor me. So, back to the issue of the top-down model. Failure can be a useful stimulus for improvement, as we all know. Given the difficulty of top-down model model of chronic condition management, all of us who helped create the chronic care model, and I hold myself partially at fault, knew that the best evidence we had converged at one easy to understand, one easy to understand point. And this was a fundamental truth that applies to all humans in any situation, probably all animals. That is, when you're all on the same page working together by golly, you can do good things. Unfortunately, at the time, we called that a productive interaction, which was jargon that not particularly useful. Later on, it became engagement, which is also not particularly useful because neither one of those terms we can measure. A sensible, measurable solution came up from the term health confidence. And here we see across many nations, what happens when you compare patients' ratings of health confidence with their ranking of their engagement by their clinicians, engagement being defined by getting good education, people listening to them, good explanations, et cetera. And what you see is a strong, strong correlation. The point being that health confidence is a useful proxy for engagement and productive interaction. And this represents a form of anarchic intelligence, if you think about it, because it can turn the entire chronic care model on its head. Instead of directing a top-down approach by asking patients how confident are you, we can begin to drive changes in health care from the front line up. And that's rather different than the way we're doing business right now. Whether we customarily articulate the term or not, it applies across almost all of us. Whether we are the five to 10%, the five to 10% who manage a health problem every day, the larger group in green who must manage their illness less than every day, perhaps, but chronically, and even all of us who have risks to our own health, bad habits, et cetera, is a pretty easy to apply term. How about health confidence here in Chicago? How's it going? Well, here you see in red, the non-poor, and here the blue, the poor. And based on the number of conditions, you can see how health confidence drops rather dramatically. In other words, there's a lot of room for improvement. But health confidence is not just bad for your health. It's also bad for the bottom line. And here we see patients use our patients reporting that their emergency or hospital admission was possibly avoidable. On the left-hand side are those who are not very confident versus those on the right who are very confident. This translates into a lot of moolah. Okay, so let me stop here for a minute. Is there anything I've said to this point? I've built the story of health confidence as a form of anarchic intelligence because it moves potentially from the patient up. Instead of us having all these check boxes saying, does your system provide, you know, an electronic medical record and all these process measures that are supposed to trickle down to benefit the patients. So imagine your system now asking this of everyone and responding to it. Any questions, comments? Yes. To me, the confidence is gonna be based on perception which can be based on, not necessarily based on any sort of reality. And so, stop in the notion of confidence. I mean, they often do that with medical education. And I'm often most worried about doctors with the most confidence because they often don't know what they don't know. And I think the same, Chris, is the whole implications of that. Yeah, how many agree with that? You agree? That's a common statement. If you think about it, I'll give you 100 diabetics, all right? And you ask 100 of them and you say, how confident are you, you can manage and control your diabetes? Half of them say they are and the other half say they are. Which group's gonna do better? There you go. Are there gonna be some false positives? Yeah. Are there gonna be some false negatives? Yeah. You gotta go where you can go. Okay, that's the argument for this. That's been shown again and again in large small numbers, et cetera. And a lot of it depends also, and we had this, is Joyce around some, yeah, Joyce and I had this talk because she's doing this. Joyce, do you wanna comment on that? Discharge process and trying to figure out the best way to do that. But it's been really interesting to try and identify barriers to people's health confidence and what issues we need to address with the discharge. And some of them, she felt misled her. And again, that can be a function of how you ask, hey, are you health confident? You know, I mean, there's a lot of that goes on and framing, et cetera, et cetera as well. But on balance, just think of a hundred diabetics and whoa, I think it'll answer your question. I'm gonna get into that a little more in terms of false positives and false negatives. Yes. I think they're gonna get this. This is useful. I mean, particularly, presumably the reason this is important is, it's one of two stories. I mean, it's important to know that these people are for some reason, but probably more importantly, you're gonna try to promote their health in some way. Exactly. And when you try to promote something, that's not always the same thing as having that thing arise spontaneously. It can arise for different reasons, right? And so, the concern that Dan mentioned, you know, of sort of false confidence, you can imagine that an intervention would increase false confidence and yet confidence itself could be very positive. So it's really gonna come down in the end to interventions that presumably increase confidence but ultimately matter as they change outcomes. I assume you were gonna tell us that. So we'll keep moving along, but thank you for the comments. And obviously, I wanna try and get done before one and the clinicians have to run out so I can get more questions. Okay, so we observed this. We had some of the misgivings others have mentioned. But nevertheless, we began designing and testing, building on health confidence and asking, what are the common issues that impact health confidence or as stand-alones we would want to know in either case, in any case. These five items combined into a form of anarchic intelligence, we ended up calling the what matters index. Each one is efficient, easy, totally actionable and most importantly, they correlate when summed very, very strongly with a patient's quality of life. And I wanna just set the stage with that. You could ask a patient, hey, how's your quality of life on a scale of one to five? And they can come up with a number and you can sit there and say, oh, I see you're not having a high quality of life. And then you gotta figure out, what does that mean? What am I gonna do about it? Whereas these five items, in essence, equal quality of life. The worse, the more of those, the worse the quality of life. And each one of these is immediately interpretable and actionable. And that becomes pretty powerful. We've done a number of controlled trials and basically the point is, you can take your score, you can immediately respond based on a few questions at the beginning, health confidence or the WMI, you can immediately give helpful information and then you can use it to stratify patients based on their response to, for example, a follow up question. Here we see thousands of responses to patients who felt they were not health confident in terms of what it would take for them to become more health confidence. So the standard thing, gee, Mrs. Jones, I notice that you're not health confident. What would it take for you to improve your health confidence? And they'll bring up issues. For example, notice the more other issues they have, the more they want the professionals to respond in certain ways. And each one of these are verbatim responses. So the patients telling you, I really think I need more info about diabetes, for example. Or personal changes, I really ought to quit throwing out my pills and not taking them. Or in some cases, social support they need in the community. This is just, I lumped them down. But on our How's Your Health website, this is done automatically and the report is immediately given to the doctor in verbatim terms. The basic approach is you identify someone who doesn't have health confidence and you say what would it take to make it better. How can people say I have to hate them? Sometimes they do. But it turned out you can see it's up to about 80%. Again, you gotta start somewhere. Okay, now, I haven't gotten the real attack that should be coming right now because you as a medical center and every other medical center in the country doesn't buy into starting with the patient and building up. You start with predictive analytics that tell you X number of patients are high cost. And you need to concentrate on this group who are high cost to save money. And you're going to give them high touch and the others you're gonna do something different. Now, you're gonna stratify on cost risk. You're then gonna have, you're gonna send the list to the clinician. The clinician's gonna sit there and say, ah, I wonder what I'm supposed to do with this list. And they're gonna, depending on the clinician, the day, whatever, they're gonna ask some questions. But each clinician will ask different questions. They'll come up maybe with some sort of solution. And then they're gonna try and engage the patient and say, hey, would you like to get on our new program called whatever it is that we have for patients like you who have multiple needs? And this is pretty cool, right? You're focusing your resources on a very few who are gonna cost your money. The only trouble is it hasn't worked very well at all. And I can overload you with the increasing number of controlled trials and other studies that show this, but in reality, it shouldn't be surprising. It's because first and foremost, the predictive models give nice likelihood ratios or nice odds ratios. They sound really cool. Now, three to one, four to one, odds of patients in this group will cost more. But when you translated into true positives and false positives, which you should with any diagnostic test, you find that noise and signal are way out of step, way out of step. So here, for example, I took a proprietary, what I call instead of a predictive model because that sounds so predictive when it isn't. I call it a computerized risk model or CRM. Here we took, for interest's sake, and be able to show you on the slide, a proprietary predictive model or computerized model and came up with 25% who were at the high risk group. And you can see they were more likely to be hospitalized than the other patients in the low risk group. This is based on about 10,000 Medicaid patients. We also totaled up in those patients their what matters index, those five questions. And if you look at three or more as an example, 31% of them would be hospitalized. It turns out, you don't need to buy the computerized model, which everyone's buying, just ask the patients. And most importantly, what you see here in the red boxes are the following. Yes, in the high risk patients, you've also got a lot of patients who have problems that are predictive of future utilization. But you've got an even larger group you chose not to deal with who are gonna use a hospital and emergency room. You're in essence rationing care from a larger group of patients who deserve your attention. From a McLean center point of view, that's unethical. That's unethical. If you have such a lousy diagnostic test that it puts more false negatives out than true positives, you better ask yourself, what test am I using and why am I paying Hopkins, for example, for huge amounts of predictive analytics that are not predictive? So, in summary, if you wanted to look at this as a form of artificial, anarchic intelligence, the brief WMI deserves an anarchic intelligence seal of approval by Golly. Because it's quality of life focus, it helps clinicians by being specific instead of just giving you some sort of categorization that you gotta figure out your 10 minute visit. It immediately serves the patient's needs because they're explicitly put out, gee, Mrs. Jones, I see you think your medicines are making you sick. Huh, maybe we might wanna talk about that. It can be used to stratify just as well or as badly as current computerized resist models and it doesn't cost you a squat. Okay, and finally, I might just add that two months ago, this organization, the Physician Focus Payment Model Technical Advisory Committee that serves Medicare, suggested to the Secretary of Health and Human Services that everything I've told you should be tested. So, there is some high level visibility going in this way. Why? Because all those other things I laid out earlier aren't worked, so Medicare is concerned. I hope some of you here, and I asked Joyce, get on it. Get on the call CMMI and get some money coming into you here. I just wanna share, you were there on Thursday. We were there on Friday. That's right. And we both got recommended, but I'm assuming nothing's happened with yours on it. It's very unlikely anything will happen. I don't know, it's two months, right? Yeah, I think you probably got the same email as last week from the air, right? No, oh yes, the AMA. Yeah, yeah, I was asking if you had anything to do with it. Yeah, yeah. The AMA is covering its own, yeah. All right, so what does all this mean for education? Our educational system, as you know, was set up the same time Sears was becoming a retail giant back in the beginning of the 20th century. That's wonderful. How are we adapting to all of this? Well, I would say the current educational system, which is, you know, has some cognitive stuff, some technology stuff and tech-nay or craft stuff, does its thing, it dumps it all in our heads. We take the boards at the end of the second year and forget it. Or we could start asking ourselves, should we build a healthcare system based on what matters to patients very simply, tailor information built on that, build our workforce about it, add more behavioral stuff and motivational and all the other tools we need and keep testing and testing and testing. That's an implication. There's a lot of jobs potentially lost in medical schools if this were adopted. Speaking of loss in medical school, let's talk about that. So I'm in the next few minutes. Let's see if I'm giving enough time for the clinicians. Yeah, I'll be done before the clinicians have to run. So I'm gonna use dermatology as an example. Here you see AI, anarchic intelligence, unmasking educational and service issues. In this case, it's dermatology. Now other countries have said for public health reasons, we wanna start using a device that can read worrisome cancer, worrisome skin lesions as well as dermatologists. You've probably read this in a number of these devices. But the important thing here in the Netherlands is they're looking at it as public health. It doesn't have to go through a clinical setting. Let's get out there so we find people before because there are access issues. Now and in particular, if you go even deeper, ask yourself, what are we doing in dermatology, anyhow, from a medical education point of view. And the craziest example and anachronistic example is a Mohs surgeon. A Mohs surgeon, as you know, as I know personally, chops out little skin lesions, looks at them under the microscope, makes sure he or she got it all and that's it. How much anatomy, physiology, pharmacology, et cetera do you need? Years ago, I was involved in training medics. These were people who were high school educated. We gave them 18 months of training and then in one or two years, they were in the field precepting. One or two of ours went into dermatology and were so good that they in essence trained most of the dermatologic residents who went through Dartmouth. Compare the educational requirement and cause to society of sending a Mohs surgeon out with a scalpel and a microscope versus sending Tom at 1.3 years. You can begin to multiply this by many others. I don't have the reference here, but the same thing has been asked as we see more and more of these technologies coming along for radiology. We've seen it also for pathology. It's gonna keep nibbling and it's gonna raise a deeper question about what are we doing in medical education? Why are we spending so much time for an output that adds very little marginal value compared to a medics? Now, one other example of anarchic intelligence is a study we did actually following up on the continuity study where we substituted phone calls for seeing patients in face-to-face visits in the office. Same clinicians, same nurse practitioners, same PAs. We randomly assigned the patients both ways and we found indeed that seeing patients was actually more expensive and no better than having patients followed up by phone. And these were all chronic disease patients. If you think back on the period of that study, which is 1992, the technologies we had available were laughably archaic compared to what we have now. And in red, difficult to see, I see, is the winner of the Qualcomm XPRIZE that is now coming to market. It's capable, and I emphasize this, of diagnosing a set of 13 medical conditions independent of healthcare professional or facility. This is kind of scary stuff as you begin to think about it because we have, as I showed, an office visit for phone call mismatch and you're beginning to see, for example, across states, they're now making it much easier to offer telephone care by non-physicians and non-nurses for that matter. You can see the growth in telephone use here at Kaiser Permanente and you can see a reaction, for example, by David Pugash, I think is how you pronounce his name. You really could see an elimination of physician element from the practice environment and his hypothesis is that would be to the detriment of patients. Well, I'm not sure it's always gonna be to the detriment but in some situations there may be some trade-offs. I don't know, but given what I told you in the beginning of the story, we haven't been earning a lot of trust lately. Speaking of trust, let's go back to anarchic intelligence in general. Five years ago or so, a medallion in Chicago went for $500,000. That was routine. This summer, there was an auction. There were no takers, zero. Sears you all know about here in Chicago, they never had a chance to respond before Amazon ate their lunch and gave the coup de gras. How are we doing in healthcare? Are we also giving people exactly what they want and need, exactly when and how they want and need it? Well, here's an answer in Chicago. You can see among the poor, less than 20% feel they're getting exactly what they want and need, exactly when and how they want and need it. And for the non-poor, it's only about 40%. I didn't slide there. See, you told us earlier that we need to be careful and we need to hear the interventions that are the interventions that really have conflict. So, let me just push you on what matters. Is it what matters, is it any questions or is it a conversation involved? I think you already know the answer. The answer is, of course, it's the conversation that follows up, the interaction, the communication. So, I know this may not be the moment for this question, but is it what matters or is it having a, why that approach to a conversation versus the person who walks into your office who you don't know and they just look different with their tones, maybe the answer is, you don't know them, we've gone past them all. So, you better ask these questions, do you have no idea who's the person? Well, it gets back to an earlier point. Take 100 patients, seeing 100 clinicians, all saying, I'm not health confidence or whatever you want. How many are gonna get the same queries and the same actions? The answer is none, all right? It's just, you know, we're talking quality here. If you're talking quality, you're talking replicative. You know, you're gonna get, and by narrowing the questions to begin with, see, if you get a response like this from your patients saying, no, I'm not getting exactly what I want needed. You gotta do deeper digging, right? Yeah, why'd you say that? Well, your breath stinks. I don't know, something like that, right? Whereas if you had the five questions, you've got more specificity. Let me build on all this. So, we all know their faults out there and it's the changes we fear. So if we get that jalopy at the forefront of healthcare up over the hill, we don't really know, do we? What's gonna happen on the other side? What's gonna happen to jobs and education? What about all those measuring folks who are out there measuring everything quantifrenically? What about all the sunk costs we've got in the way we do business? This is pretty scary stuff. I would argue though, let's not undervalue science. Let's ask ourselves, do we really wanna stick with a status quo that's expensive and not working or should we begin to ask ourselves the more fundamental questions I'm trying to throw in front of you today. I realize it's a bit of a fire hole as talk, but when in doubt, go back to the Godfather to ask. So after 1971, I learned that the boss was still alive. This is probably a decade later. So I went up to him and it was like a scene from the Godfather, I kid you not. I might add, both his legs had been broken. I don't know why, but in any case, the boss met me, came up, gave a hug. John, how you doing? Great, we caught up, talked about things. And then when he got to Tony, he really got animated. And the conversation went something like this. And Tony, he always said, keep a reputation and don't badly damage the host, right? Smart guy. You're a doctor, right? Doctors especially should think about Tony, guilty on all counts, up for 20 years. He didn't do much stuff. He only had tax evasion, extortion and bribery. I mean, nothing big. And with a little help from his friends, he walked in 39 months. So the question for all of us here is, are we gonna be like Tony and depend on our friends to get us out of trouble? Or are we gonna take our own initiative to begin really anarchically challenge what we're about? So let me be specific about next steps that are positive. I think, again, we should start embracing anarchic intelligence thinking. It's out there. Every one of us is capable of that type of thinking. What if, what if, but we're so bludgeoned, right? We're being stooges. We're bludgeoned by the system, I just, the conversation we had last night at dinner. There were some stooge-like comments, no offense, but you guys have been bludgeoned. We've all been bludgeoned. But the question is, can you fight it with science? And you're all capable in this audience. So when it comes to this, this is the Joyce heroin slide. She's already jumped to saying, I don't know, this guy Watson may be a nutcase, right? But there's something appealing there. I'm gonna play with it. And that's cool. That's cool, because the next day she could go on to a controlled trial or all of you and you can begin to see, is this hogwash or real? I think we have kind of an idea that it's better to build from patience up. Somehow that sounds more Hippocratic than Hippocratic, which is what we're doing right now from the top down. Next, protects the brand. I really think we ought to be snitches to use a good old mafia term. And we ought to lose the scandals. Be a snitch. Look at these headlines. They're nauseating and they're just this year. This one just blows my mind, right? And the whole idea of thinking about the fact, just try to envision, nobody knew about this guy being on 15 different boards. Come on. Or this guy, how can you as a dermatologist run up 26 million? I mean, just think of that number. Somebody's gotta start snitching. So I hope there's some snitches I'm getting going here in the audience. And these are good websites for future snitches. And then finally, I had a very off-record, we can talk about it, interesting, horrible situation. That's the whole, you're perfect for, I think you're setting me up for good responses because that's the whole thing. We're all so used to these flaws and we all benefit to some degree by the craziness of healthcare, either our salaries, our jobs, et cetera. We have real trouble seeing the faults. And yet the people out there are seeing these faults more and more. And if we don't do anything, we're guilty. That's the point. And we're not gonna have friends like Tony who are gonna be able to say, hey, you can walk after 39 months. So I think it's worrisome. My big one is this $3 billion one. You can do this tomorrow. You can go up to the head of this hospital and say, how much are you spending for PR to come up with at the forefront? Come on. We're waiting on the forefront. Oh, I know, and Dartmouth's better than you because we're better than medicine. Yeah. So I would argue that if tomorrow we took $3 billion, we could pay the tuition of every medical student in the United States and 20,000 PAs or NPs. We could cover everything. And then they wouldn't have to go out into practice into super specialties to bring in huge amounts of money, perhaps unnecessarily. And they certainly would stay away from crazy billing practices that run up 29 million of dermatology stupidity. They wouldn't be so pressured. And the hospital, meanwhile, can say, our motto is helping you and our community be health confident. And they could tomorrow, you could see in the Chicago Tribune headlock, this year, University of Chicago patients, 87% are health confident compared to the national average of only filling the blank, 12%. So in summary, I've taken you through a rather crazy romp. I'm right where I wanted to be. Here are the summary points. I think clearly the McLean Fellows should look at this from the ethical components of what I've talked about. There's a huge amount of unethical stuff going on out there. And we can talk about it or maybe we should do something. Here's some ideas for doing it. And as the boss would say, if you don't, things happen. Thanks very much. So if one person expressing confidence is an educated PhD with an IQ of 130 and they're 50 years old and they're not demented and whatever. And somebody else expressing confidence has an IQ of 80 and not a high school education and maybe dementia is thrown into boot. What does confidence mean coming from the vast array of the distribution of the population that's capable of expressing it? Well, first of all, I think you threw in dementia as one of the characteristics of someone with an IQ of 80. Did you say that? Dementia is not a characteristic of somebody with an IQ of 80. Dementia is not a characteristic of someone with an IQ of 80. Yeah, the whole. I just created two opposites factor of n's and say, each of those says they're confident or they're not confident or whatever. What does it mean? Well, what it means is whatever it means to the patient and it's a follow up point that I think we pointed out that would be the key thing. So if both of them said they're fully confident, I'm gonna take them at face value at least at that point. But if they say they're not confident, then I'm gonna ask, why is it that you've responded that you're not confident? And the one with an IQ of whatever could very well say, because I'm caregiving for my wife and it's just killing me, right? And the guy who is high school educated or something like that could say, I don't have any money. And some of that we can deal with as I showed you on those bars, some of them are very socially oriented, but a lot of them, 60% or so for the people who are like the ones you care for who have multiple issues are gonna say, doc, here's an issue I have and it may be I can't use a hemoglobin A1C thing or it may be much deeper. I don't know what hemoglobin A1C means or I don't give a damn about hemoglobin A1C. Does that answer the question sort of? I mean, what is the point you were driving at? If we're going to assess how well we're doing by the degree of confidence that patients express, I'm not confident yet anyway that the degree of confidence that the patient expresses necessarily tells me a whole lot about how we do it. I think I'd agree with that wholeheartedly. However, if I had the five questions and the patient says, for you, all five are negative, I'm doing great. And for him, all five are doing badly across comparable patients. Let's say you start, I'd be very concerned about that, right? So the more you don't wanna get, we chose five for obvious reasons. Human memory can't really go much beyond five, number one. And number two, I was gonna show to the group afterwards we're having a briefing that the ROC curves, the receiver operation characteristic curves, it shows when you go from one question like health confidence, that's not so hot, right? It's not gonna be perfect, true positive or false positive. You go to five, you're really getting there. You go to 20, marginal, right? I mean, it's a standard game. So that's behind the five. So I'd agree with you on one question, I'm not gonna say anything, but on five, if all your patients over time stay at fives and all the his go up to zeroes, I know where I'm gonna go. Yeah. Ask about where it's even going, ask about smoking, ask about everything. And when people look at it all the time required to do those and exceed the amount of time that it would take. So I'm assuming that when you think about either the health and confidence question or what matters in this question, you're primarily not imagining a clinician literally building this into everything that you would ask me. You're thinking there's a form that's filled that however somehow this gets hidden. I don't know the answer, I'm just guessing, but I wonder if you can talk about how you imagine this being most effectively used, particularly in a longitudinal relationship. Because there are alternatives, which is the 10 second pause after how are you doing? Are you actually just a person? I agree, but that gets us back to your previous comment, that how I pause and listen versus someone else. In answer to the question, how do we fit this stuff into practice? That's the other side of my head which has been for the last 30 years longer than Google. We've been putting on the internet free of charge things like howesyourhealth.org that's been picked up by the Academy of Family Practice, that's been picked up by the American College of Physicians that we had for Medicare, we're using it statewide. The answer is we can't ask all these questions, but before an office visit, a week or two ahead, patients of all educational levels do a darn good job. If they know, A, they're getting feedback and it does it automatically, and B, if they know, you're gonna take it seriously. So if an email goes out to a patient or something like that, in fact, they answer the questions, there's some score that comes back, it may be a high level or low level or whatever or change or something like that, and then that goes back to the clinician and so when you presented your office, you see that this has changed? Yeah, so bottom line with the technologies, it can be anything, but the way it works almost 99% of the time is clinicians who use something like How's Your Health say to the patients once every year or so, would you mind having a complete health checkup? The patients immediately get information tailored to their responses. They then can send their action form back to the clinician and the action form says to the clinician, pay attention to this, this, and this. It doesn't give these little color codes and everything, you gotta figure out, it's very verbatim, and in this case, if they say I'm not health confident, you get verbatim why they said they weren't health confident. So it saved you all that 10 seconds of silence. You can immediately say, wow, I didn't know you were not health confident and you think you need, eh, let's talk about it. So that's how it works. It's pretty, it's not rocket science, but it takes the work out of the practice. Yeah, I mean, one of those things we have in our clinics, for example, just literally a tree house nurse asking, you know, why are you coming in today? Yeah. You know, and sometimes that's not that useful, but often it is. Yeah, it can be, sure. That brings us back to Alvin Feinstein. He always used to call that the iotropic episode. Yes. Yeah, I think you've convinced me that in a big population that confidence is useful, but I think that, you know, his question and what Dave is pushing it also gets to the idea that for each patient, you really have to unravel what that confidence means and why they're not confident or why they are confident. And for some patients, it will mean that they're able to take care of things better and for some it won't and untangling that. But getting back to this whole notion of anarchic intelligence, I mean, it's based on the word anarchy, right? And so the idea of using the same term every time you see a patient seems to me to be sort of a step away from a real anarchistic way of approaching care in that it's not, it's another rule about how to, you know, what you should do at this moment instead of actually responding to the person in front of you. Yeah, I agree on the second point that the anarchic implied the current hierarchy of healthcare, anarchic against that. You're stretching it further and I'm not sure. It's sort of like productive interactions. It's one of those terms that you just say, oh God, but it's the best I could do for today. In terms of your prior point, all I can say is that you should go sign up tomorrow for How's Your Health and try it on your next 20 patients and see if you like it or not. It's been built on 33 decades of clinicians who are just as skeptical as you are and out in the real world salt mining and not happy. So, you know, it works well enough and the price is right. Well, thank you all very much for putting up with a rather odd presentation. Thank you. Thank you.