 I now would like to introduce the final speaker on this panel, Dr. Albert Wang. Dr. Wang is professor of medicine, the director of the Center for Chronic Disease Research and Policy and an associate director of the Chicago Center for Diabetes Translational Research here at the university. Previously, Dr. Wang served as a senior advisor in the office of the Assistant Secretary for Planning and Evaluation in the Department of Health and Human Services. Dr. Wang's research focuses on issues at the intersection of aging, diabetes, and health economics. He's received numerous awards for his work, including the research paper of the year from the Society of General Internal Medicine and an elected membership in the American Society for Clinical Investigation. Today, Dr. Albert Wang will present a talk entitled, as you see behind me, the Science and Ethics of Medicine De-Intensification. Please join me in welcoming Dr. Albert Wang. Can you hear me? Is that loud enough? Thank you again for having me at this amazing annual conference. It's quite a network of, quite a social network you've created, Mark. So I hope this is coherent. So I'll start with a question. How did we get here? And I'll just explain what this graphic shows. This is an international study looking at the proportion of older people over 65 who are taking more than five medications. That's the definition of polypharmacy. And on the x-axis is the years and the y is the proportion of people. So just to put a point on it. And these are from individual studies. So if you look at the study from Kraftmann, from Sweden, the rate of polypharmacy, as they define it, has risen from 22% to now 60% from 1980s to 2005. And the pattern you'll see across all countries is that the number of people taking more than five medications is a uniform across, it's an international phenomenon across all countries. I think actually the H is a study from our own National Social Health and Aging Project. I think it's a study from Dima Kato. That's the H that also shows the rise from 2005 to 2010. So why is this happening? Why are more older people taking more medications? There are at least two major reasons that have been proposed. One is that there are more indication, there are more drugs to prescribe and there are more indications to prescribe those drugs. And part of what fuels this also is that to this day, the different societies for different medical disciplines have actually splintered into even more subgroups and they each write their own clinical practice guidelines. Everyone's practicing their own kind of medicine in their own small discipline. It's easier to control your world if you just treat one condition. And so no one thinks about the consequence of all these guidelines and all these recommendations and new drugs. The consequence is polypharmacy. So who's monitoring polypharmacy? It's the dwindling number of primary care doctors like myself who actually don't have much time to even reconcile all the meds that are being prescribed. So another second reason that this may be happening is that we may be actually just being more successful in keeping people alive for longer. So we are doing a really good job and maybe it's due to some of these medications. So part of it is also we're a greater success in maintaining longevity in our populations is leading to this overall higher rate of polypharmacy. So why would we mess with a good thing like all these medications? So on the left are just some ideas that I think that many people would share, which is that one reason to reduce prescribing, de-intensify or to lessen the number of medications that a person's taking is to reduce the harms of those medications, potentially reduce the risk of drug-drug interactions in particular. Another possibility, which is more controversial is that maybe some of these medications, while they may not cause harm, they're not doing any good either. So some medications that we've prescribed in some populations are just there, we've maintained them, but they may not be producing any clear benefit to the patient. A third reason to consider de-intensification is that life might be better. We might be able to improve quality life by reducing the number of medications that a person takes. I did not have time to include pictures, but there's these classic pictures of an older person sitting in front of there at the dining table with 20 medications. It is a real burden to take all those medications to manage the sequence of when they have to be taken. And actually, I've had discomfort taking a single ibuprofen. So imagine taking three or four at a time. So the burden of taking medications is real. And David Meltzer thinks, in the audience, he and I actually quantified the overall burden of taking multiple medications for diabetes, and it's real. And another idea is that if we took less medications, it would reduce the cost of medications to the patient, but also to the system. So these are all reasons to de-intensify. In the right-sided graphic is a figure that's depicting the proximity of all the effects of polypharmacy from the most proximal to the most distal. So the most proximal effects of polypharmacy are drug-drug and drug disease interactions, adverse drug reactions. Outside of that are the events or consequences of adverse drug events, such as falls. And then outside of that are loss of physical and cognitive function, frailty and sarcopenia that could result from a fall due to medications. And finally, of course, on the outer ring of the most distal effects would be, of course, hospitalization and death. In, for those of you who are geriatricians, one of the classic risk factors for falls is actually just taking more than four or more medications a day. I did not believe this association, even though it's a long, it's been studied many, many times in the 80s and 90s. We recapitulated the finding in the 2000s. It's a real association. The more meds you take, the more likely you are to fall. So one tricky part of this field as it's an area that is growing is that there actually are alternative definitions of what de-intensification means to patients, to doctors, and even to reach investigators, if you look at the growing literature of de-intensification trials, the definitions used to define de-intensification vary. They vary in terms of actually what happens to the change in the regimen, and also they vary in terms of their intended or unintended effect on the actual biomarker. So I'm talking really in the realm of chronic disease management. So thinking about de-intensification of blood pressure drugs or diabetes medications and their effects on glucose. So in terms of change in regimen, the simplest and the least dramatic change would be, of course, a reduction in dose. The next level of intensity of change in regimen would be something where we just make a switch or a swap out a drug that is not favorable in terms of its delivery route, such as an injectable drug for an alternative, like an oral agent. So just lessening the intensity of the regimen itself. And the most drastic de-intensification move is, of course, just to stop a medication. And I'll show you some examples of recent trials of attempts to just stop medications. The consequence for biomarkers, for many people, de-intensification is really different if it means that there's actually no effect on the biomarker. If I take less medicines and I'm achieving the same biomarker levels for many people, that is actually a complete win, that is a win. But in many cases, our recommendations for de-intensification are actually to allow the biomarker to rise. That is more controversial in the minds of many patients. So I'm going to, these are the next two slides show you results from really, now in the field of de-intensification is gonna be one of the first classic, large-scale multi-center randomized controlled trials of an attempt to de-intensify. This comes from the field of, this comes from the Palliative Care Research Network. This is a trial led by Gene Kuttner out of Colorado. And they started with one of the simplest ideas of de-intensification. What if we just stopped the drug that has no, in this case, they stopped statins. They took a population of people who were in Palliative Care and who were already taking statin and said, what happens if we just ask, if we have them randomized to stopping the statin or not? And these curves show you the rate of more, so what they were concerned, what people are very concerned about is if you stop a statin which has effects on the cardiovascular system and prevents heart attacks, if we stop statins, are people gonna start dying at a higher rate? Are we gonna have more cardiovascular mortality? This curve shows you overall mortality and you can see the statin arm and the discontinuation arm and there was actually no statistical difference between. So there was no increase in deaths if we stopped statins. And this is a group of people in Palliative Care with a fairly short life expectancy under one year. This is what happened to quality of life. If the dot and the cough intervals is to the left of zero, that means it favors the intensification arm. So this is to show you all the different ways in which quality of life improved with the intensification in this population. So from this trial, we would conclude that this continuation of statins in a Palliative Care population is beneficial, if anything, and with no harm. But just this last year, again, this is a single trial I don't know of many others, but this is an observational study from France where they took data about people who were over 75 already taking a statin in France. And for some reason, some have stopped statins and others have continued. And this is an observational study, so the reasons for why people are stopping statins may be tied to the result that we see. But in this case, what happens, they looked at rates of hospitalization for cardiac events. And the group that stopped taking the statin had a higher hazard of ending up in the hospital with a cardiac event. So at least from this data, this observational study of de-intensification is not a good idea. So these two results are conflicting. So I'm going to, now, what is gonna be challenging is that the idea about de-intensification is gonna be different from condition to condition and from decision to decision. So that was the simple, in many ways, discontinuation of statins is a very simple idea because there is no, because you just stop a single pill. Now I'm gonna shift to the field of diabetes, which I work in, in a more complicated de-intensification in this case could lead to poor control and actual symptoms of the actual underlying disease. So we can't simply stop medicines in this case. We have to maintain some medicines in order to control the disease, but there are maybe cases where we can de-intensify. This is a summary of guidelines that have been published in the last 10 years. The ADA guideline, which you see in the middle, is one that I helped to shape. And in this, these are guidelines for diabetes and people over the age of 65 and there's general consensus that we need to individualize diabetes care, at least the intensity of care. For those of you who don't know, diabetes is a lot like long-term planning for retirement. You have to put in your money, you have to save your money early on, you have to invest in, and actually early on you should invest in the stock market and be as aggressive as possible. But later in, as you approach retirement, it's time to start pulling the money out and maybe move to safer investments. That's the case in diabetes care. Most of the trial data shows that it's very beneficial to intensely control sugars in people in their 40s, 50s, and maybe early 60s, but the trials of people in their late 60s, 70s, and 80s have been very mixed. In fact, you may remember a landmark randomized controlled trial of diabetes care in 2008 called Accord, had to be stopped because increased risk of death in the intensive treatment arm. So the reason for the three tiers that you'll see, I've listed, is basically we're saying that the longer you have to live, you should be pursuing lower targets and aggressive treatments, but the shorter your life expectancy is, we should be using less medications and allowing the sugar to rise. This is controversial. This is not fully proven in a randomized controlled trial, but it is based on this concept of remaining life expectancy. And actually you'll notice that this is actually, you've heard Dr. Tol, this is a research question or a clinical problem in the space between chronic disease management and end of life care. So we're in this space in between. So we recommended different targets for different patients. And many of these, there were early recommendations of this nature in 2005, and then we published these ADA guidelines in 2012. And this is what we see in data five years later in the country. Basically no one's listening to us. And so this is an observational study from NHANES looking at the divides up the older patients into the three tiers I described to you. And this is a histogram of the achieved glucose levels of healthy, complex, and older patients in poor health. There is no difference in the rates of glucose control across these, treatments essentially the same. Achieved glucose levels are essentially the same. And what's actually even more striking is rates of use of very aggressive medications like insulin and sulfireuria are also nearly identical across every class of older person. And actually for those of you who practice general medicine or primary care or endocrinology, it's actually easy to see how this happens. We don't keep track of who the, we don't know that we don't sort the older person in clinical practice into these different buckets. And we've practiced a lot of whack-a-mole medicine. We're just trying to deal with problems as they arise. We don't think about the long-term trajectory of a disease. And then this is a great study from the VA that is a similar follow-up. What they asked was, in the VA, they actually remind the doctors of the health status of the patient and they have different tiers of glucose control recommendations for these different patients. And in the VA system, even though they have this automated system, there is essentially no de-intensification. The height of this bar is the rate of de-intensification for blood pressure and for glucose for different patients of different levels of health status. So the overall rates of de-intensification are very low and they're no different across health status. So why doesn't de-intensification happen in clinical practice? It may be that we just don't have enough time to waste the risks and benefits. It's cognitively challenging. It may be difficult to communicate with patients about this idea of taking less. And it may be a legacy effect of maybe about 20 years ago, we had multiple public health campaigns driving into people's brains, different glucose targets, and it's hard for people to forget that. And the other challenge is that care is highly segregated. So if I wanna change diabetes care and how it's delivered to older people, I have to not only convince people in outpatient practices, but I have to convince the hospice. I have to convince the long-term care nurses who actually decide on how much medication to prove to give for demented patients. So the system's so segregated with so many different players, you have to somehow reach all those parties and convince all of them that care needs to be changed. It creates, this idea of de-intensification creates obviously new dilemmas. And this is a phenomenal paper that comes from Alexi Torqui, one of the McLean Center Fellows, where she did focus groups or maybe one-on-one interviews with older people about the idea of stopping screening for breast cancer or colon cancer. And some of the quotes from this paper are just amazing. So many of the older patients, these are all, I think, older adults that came from Indianapolis, is that right, Alexi? And they said things like, I think stopping would be the same as me taking my life and that's a sin. So even the idea of stopping a mammogram or no longer doing colonoscopy had a lot of meaning attached to it, more than we think. And so physicians' recommendations to stop something may threaten patient trust and most patients had limited discussions with, in this study, most patients had not spent a lot of time talking to their doctors about whether or not to continue screening. It was just assumed that we would, it was the default that was to just continue doing what we're doing. And this is a more recent study about diabetes students of de-intensification. This is a national study of older adults where these investigators from Hopkins actually asked patients about what was important for them in terms of adding medications or de-intensifying medications. And they lined up what's in the guidelines with what patients thought was important. And basically patients' ideas are completely, are apart from adverse side effects of drugs, there's discordance on every other idea. So older patients, for example, when they were asked, well, if a person had a longer life expectancy or a shorter life expectancy, who should have more medications? The patient said the person with shorter life expectancy should get more medicines. And then if you ask them about what if the person has complications of diabetes already or no complications, they say the person with more complications should get more medications for diabetes. So completely discordant with what we know about epidemiology and treatments of diabetes. So it's going to be very different, challenging to implement this in practice. So one idea we've been trying is to use shared decision-making with patients, could share decision-making help. We've tried two different trials. The most recent trial is this one, My Diabetes Goal. We've basically had patients interact with a survey and told them what the ADA recommends for them. And just as a teaser, this has not formally been, but we know from the anecdote from talking to the nurses that have been doing these discussions that almost every patient, if they're given a target, they'll pick something always lower than the target. So it's another challenge that we're gonna face in communication about this idea. So many unanswered questions. When is the right time to de-intensify in a patient's life? What are the right clinical outcomes for studying de-intensification? If we need to do it, how do we do it? Thank you. She's great. Dr. Segal. Oh, she was first. Oh, let her go first. She got to the microphone first, please. I jumped right up. Yeah, she did. That was really interesting. Thank you. I just was wondering back to your discussion with the statin case. Now, I don't know this for sure, but I have this intuition that the study populations really seemed to me to be quite different. Physiologically, not just with respect to their proximity to death, say, but it would seem to me a person who's a candidate for palliative care would be physiologically much different, even with respect to, you could show that on a lab works. And I was wondering if that was taken to account. And the reason I'm wondering that is the group who are in the 75 year old group it seems to me just intuitively that if a physician were to say to that group, like we'd like to take you off the statins, but we'd like to get you physiologically sort of tuned up so that you wouldn't experience adverse effects from that. So diet and exercise and stuff like that. Would it be the case that those two groups would look more similar than they did the way that you presented? So I think your observation is, I think you're absolutely right. That's likely the difference between the French study and the palliative care study. They're probably different populations. And that gets to this question is when is the right time, when is the right, what's the right population to de-intensify in? The palliative care population had a life especially under one year. They were very, very sick. And so if you're over 75 in France, who knows, you may be living for another decade. So there's a very different populations. I don't know about the nuances of statin removal. But yes, people frequently try to encourage more behavior change in other parts of life when they stop a statin. And the willingness of patients not to engage in those kinds of lifestyle changes that could actually favorably affect them could be added to your list of why just in practice the de-escalation doesn't occur. Sure. And the French study I saw went for four years. Yes. In contrast to the palliative care. One year. One year, yeah. Yeah, I think that's a very good point because it seems to me that what you're really showing us is just the way we introduce some market drugs coming to kick us in the butt at the back end. Because what we do is we'll say that a drug reduces stroke risk, say by 25%. So it starts at, that means it takes it from 4% to 3%, which means our number needed to treat is 400. And we do that on large population-based trials. And then we take these older folks, smaller samples run them for shorter times and we say, well, we don't see a benefit. If we were honestly selling these drugs at the front end, what we'd be telling patients is that this drug, if you take this in a group of 400 people, one person is gonna have a stroke prevented, but we don't start the drugs that way and it's no surprise when we take them away, we don't see an effect. Yeah, that's actually, that's an excellent question. So why do we have the de-indezification question when we could actually potentially prevent this whole thing by doing a better job at the start? When we introduce drugs to people being more forthcoming about, and actually there are tools designed to show people the modest gains with individual drugs. So you're absolutely right that this, we're trying to solve a problem that's been created by other practices in medicine. Yes, please. Got a follow-up on this. It's not actually that easy to find that kind of information versus going to a company and out of the cell. Oh yeah. So I'll tell you a humbling story, a humbling experience we have. So the personal DC trial that we did was actually a tool that had a simulation model embedded within it that calculated the person's risk of different complications at different glucose levels. We thought it would, in actually in many cases, the differences were minuscule. So you can guess what happened to the intervention arm versus the control arm. The intervention patients routinely picked a lower target than their original target even though the numbers showed no difference. And it's because just the idea of having risk displayed to them made them scared and picked something. And there's a similar pattern happening in the ongoing trial. So there's something about numeracy, the idea of risk. So even talking about it can lead to things you don't expect. It's a fairly complicated area. Thank you. Elbert, thanks so much.