Thank you so much for coming. My name's Monique Yohannan, and today I'll be talking about the top 10 medical stories of 2023. This talk is especially meaningful to me because 20 years ago is the first time I did a version of this talk in San Diego. I did it for about 13 years, and then I took a break, but was invited to come back this year, and I'm delighted to be here. I see a couple people nodding, and thanks for coming. So without further ado, let me start off with a bit of housekeeping. So disclosures. I don't have any right now, but I will go through some of my work history that I do think is relevant to disclose. So I used to be on faculty at UCSF, then worked at the VA in Stanford, and then I left academics to work in tech. So I've spent most of my career working in various health technology companies and clinical decision support. So I worked at a company called Hippocrates for many years. Then I was at Milliman, which became the Milliman Care Guidelines, and then I went to change health care, where I was the chief medical officer and oversaw the development of the InterQOL guidelines until they were purchased by United, and I left. In a week, I'm actually, I was retired briefly, and then I got a call a few weeks ago by an AI startup, and so will be joining a company called Adia Health in about a week. So with that, let me just jump into how I put this talk together. It really hasn't changed any since I first did this. I read a lot. I get news feeds in my email box every day. I get up in the morning, I read, and I read in between errands and things like that. I go through the kind of traditional sources for medical news, like Health Day and Journal Watch and things like that. But I also use sources that are, I would say, in the lay press. And I think when I consider what's important in the medical literature, I always try to see what are the stories that have made it into the New York Times, what's in the Wall Street Journal, more importantly, what's in USA Today, because those are the kinds of sources that patients are going to ask you about. So I have a couple of basic goals every time I do this talk. The first is to review the key medical literature from the previous year. I have, I think, one or two things that are more recent that I might mention. And I make my selections based on whether I consider something to be newsworthy and whether it's likely to have clinical impact. So I have three categories that I look at when I consider whether or not something's newsworthy. Is it timely? Is it important? And is it interesting? And then the next is, is this likely to change how we practice? Is this likely to have any impact in clinical practice? I mentioned a few other sources, but if I were going to only pick one that I would suggest you might look at, it's Evidence Alerts. This is a free source that comes out of the McMaster's group in Toronto. It's free. You can subscribe to it. They do an amazing job of it. But not only do I love that the source has a summary of different articles that are important, but you'll see that your colleagues, in whatever specialty you pick, will rank articles for their newsworthiness, for their clinical impact. And you'll see that different specialties will determine whether or not something is more important. And then finally, when I'm looking at the stories, I really am trying to get at 10 stories. So I'm going to go through a lot of articles. There's an extensive bibliography. The handouts are uploaded. So everything's in the bibliography there. But what I'm really looking for is, what does that literature tell you? And I think if you read a lot, you end up really understanding the stories that the literature told you in the past year. And then just in terms of structure, I like to have a way of approaching it. So when I read, I use the structure that Gary Rolfe, who's at the University of Swansea in Wales, came up with many years ago. But I really, it's a basic one, but I like it a lot. So what? I'm going to talk about the articles and the stories. So what? Why should you care about it? And now what? What should you do about it? So with no further ado, what? Well, here are the top 10 stories I'm going to be covering today. So I'll talk a little bit about COVID. I'm going to be talking some specifically about disparities, but I am going to be including a lot of really thinking about disparities and equity as I go through the articles. And I would say, just in general, in the way I read the literature, I think this becomes increasingly important that you look for issues of health equity as you're reading. Then I'm going to talk about cardiovascular disease. So some focus on hypertension, hyperlipidemia, and diabetes. I think the GLIPP ones clearly were the story of the year, so they get their own section separate from diabetes. Then a little on behavioral health, so diet and exercise. Then some on screening and a tiny bit on technology to finish up. So the so what? So why should you care about it? And I'm so delighted to see a group of psychiatrists. And frankly, the reason I started doing this talk is that myself and my colleagues in medicine and surgery, we do a horrible job of taking care of people with psychiatric conditions. We have failed this population. And it is my hope that in this talk that you'll at least, whether or not you're managing these conditions directly with your patients, that you'll feel more prepared to have these discussions and advocate for your patients when you're talking to your medical colleagues who are caring for them. So people with severe mental illness represent about 5% to 10% of the general population, but we see that they have a life expectancy that's 25 years shorter than people without these conditions. And about half of that excess risk is due to cardiovascular disease. We see a huge increase in the metabolic age of these people. And yes, there are lifestyle factors. There are medications that are part of this. But a lot of this really is just due to frank disparities and discrimination. In a study that was published many years ago that has always stuck out to me, and this is in Nova Scotia, so we don't have the excuse that it's a bad, inequitable US healthcare system. You showed up with a history of not just schizophrenia but even depression. You were having a heart attack. You were not only less likely to go to the cath lab, you were less likely to get an aspirin. And finally, the now what. What I said is that I do look at this literature with regards to what are the disparities, what are the health equity issues. When I did my MPH, what really stood out to me, my very first class, the very first sentence of that class was delivered by Bob Lawrence. And he said, every issue of public health is first and foremost an issue of social justice. So I really do think about that when I read and when I approach the literature. And with that, let me jump in. So COVID, not a lot on COVID but a little bit. I think we know that in 2023 in May, first the World Health Organization on May 5th and then the CDC on May 11th ended the public health emergency. Their declaration is this is over, so yay, it's over. But what we see in 2023 is that there's a big difference in what we have in vaccine guidance. We have the CDC and then we have the rest of the world. So the CDC recommends basically that everybody gets an initial dose and that the additional dose, then everybody's basically gonna get additional doses. So six months to four years, people with moderate to severe immunocompromised and people who are older than 65 are all gonna get additional doses. So what does the rest of the world say in 2023? The World Health Organization guidance, and this is really mirrored in the National Health Services guidance as well, is that there's a booster every six months for the following three populations. People older than 75, people to 60 to 75 with comorbidities, and this is not bread and butter hypertension, this is people with really severe immunocompromised, these kinds of comorbidities, or pregnancy. A booster once a year if you're 60 to 75, 18 to 59 with comorbidities or severe obesity, and that would be a BMI greater than 35, and healthcare workers. They don't recommend boosters for anybody else. But we see in the U.S., there's a lot of drama, I would say, in the literature in the past year. Why are the uptake rates so high? Why are the uptake rates so low? Why are they so low? And we see that the uptake rates in children and adolescents are 14%, and we see in adults older than 65, they are 42%. So we have this study that gets published in the European Journal of Surveillance in February 2023 that looks at COVID vaccination, because if we're gonna mandate something, we hope that it's gonna actually have a benefit. Now, obviously, we all learned Greek during the last few years, so when we were looking at Omicron, they looked at this and they looked at the benefits for people who were vaccinated, and what they found is that for people who are older than 65 there was a benefit in terms of decreasing the risk of hospitalization or severe outcomes. They didn't find any other benefits for younger age groups. Next, I wanna talk about a study that came out in June of 2023 from the American Journal of Preventative Medicine, and this was centered in the Bay Area, and it's got a framing that when I think about COVID, but I also think about applicability and a way to approach social determinants of health, I think this becomes a really important study. So we know that nationally that there was about a 64% uptake in COVID vaccination rates in 2021, but what we saw in the Bay Area where we had rates that were much higher than this, and I'll show you what the rates are in just a second, we saw significant increases in morbidity and mortality in people who were in historically disenfranchised groups. So what the authors of this study did is they created a vaccine equity metric, and I would say if you're gonna think about social determinants of health in an easy-to-remember way, look at people's zip codes, because zip codes really track pretty well with a lot of the social determinants of health issues that drive disparities. So they looked at economic resources, social resources, education, transportation, neighborhoods, housing, clean environment, and healthcare access, and in the Bay Area, this really did fall, unfortunately, pretty neatly into zip codes. The green line at the top are the people who were in the most privileged zip codes. So in the Bay Area, we started out in July 2021 with a vaccine rate of 86%. Those vaccines rates went up to 92%, but part of why you see that there were differences in outcomes is that in the lowest-rate groups, which were still better than the national average, you saw vaccine rates that were about 10% lower. So they started at 68, they did go up to 83%, but really, when you look at this, it tracks very significantly with the poor outcomes we saw in people who were in historically marginalized groups. So what's the harm in being more conservative in taking the U.S. approach that everybody should get a vaccination? I think we see the potential harm in this study that's in the American Journal of Transplant Medicine that comes out in September 2023. Now, lots of transplant centers are requiring that people have active vaccination for all vaccines as a condition of maintaining active status on the transplant list. So in this study, they looked at people who were removed from wait lists due to a lack of COVID vaccination proof. And what did they find? That people who were in the lowest quartile of this vaccine equity metric that had been developed earlier in the year, these were the people who were most likely to be inactivated. These are not people who would have been inactivated if they lived in Sweden, if they lived in the UK, but they were inactivated in the United States. So the what of COVID-19, I think we see the big story is that the public health emergency declarations are allowed to expire. And we also see that vaccine guidance varies. The so what I would say is that we have right now a lack of evidence for hospitalization mortality basis and mortality benefits for people in younger age groups. Vaccine mandates can have unintended consequences. And I would add that not just about the transplant list, but the thing I think we have lost most in keeping this mandate is we actually have a great vaccine that probably would be much more helpful for many people, which is the vaccine against RSV. I do think there's a certain amount of vaccine fatigue that unfortunately, as low as the rates of uptake have been for the COVID vaccine, they've been even lower for RSV. And actually the evidence for RSV is pretty good in preventing harms. So the now what I would say, it is important to do equity evaluations in all healthcare delivery. And I do think considering maybe a more conservative approach to vaccines and then having personalized decision-making and shared decision-making when we look to expand it and tighten recommendations for other groups. Next, I want to talk a little bit about some specific literature on disparities. We see an evolving literature that first really started appearing about five, six years ago, the missing Americans. So I think people are aware that the 1930s to the 1960s, the United States had a mortality rate that was lower than peer nations. And then starting in the 1980s, the death rates in the United States grew. This was really disproportionately borne by underserved populations. And half of the people who died early, these are people who died before age 65. We saw in 20 and 21, 2.1 million fewer people than we would have otherwise expected. And from 1980 to 2019, so predating the pandemic, 335 million fewer years of life. Not surprisingly, COVID-19 and HIV AIDS lead the pack. And drug and alcohol mortality and suicide, not a huge surprise. And even though cardiometabolic disease is only 4% of this excess mortality, we still see that this represents a disproportionate burden that really, when we look at every one of these top issues, I would say these are disproportionately being borne by marginalized populations. So then we see in JAMA in March of 2023, this study on cardiovascular risk prevalence in treatment in adults who are 20 to 24. So they looked at the NHANES study, probably the best study of nutrition and health that we have in terms of a database in the US, about 13,000 adults, so people 20 to 44, about half of them women. And we looked at changes in risk factor prevalence during this time and what we see. Big change in the rates of obesity. And generally speaking, not much of a change in the rates of diabetes and hypertension. But, and I'm gonna use the terminology that the authors use, this is not necessarily the descriptors I would use, but we see very significant increases among black and African-American people from 16% up to 20%. Mexican-American members of this cohort, 6.5 up to 9.5%. And Hispanic-Americans from 4.5 up to 10.5%. We didn't see any real changes though, even though we had these higher rates. We didn't really do any better job in treating these people. We didn't bring their hypertension under control and we didn't actually treat them more often for diabetes. Interestingly though, hyperlipidemia did drop, which is kind of random. The authors posit that this is due to trans fatty acid regulations. I'm not sure that I'd buy that and I don't think we have a great descriptor for that. Finally, we look at this study that was in the Annals of Internal Medicine in August 2023 and we see that cardiovascular disease mortality is about 30% higher in black versus white patients. So these authors looked at a prospective database, about 50,000 adults, and they looked at metabolic risk factors and behaviors. And then they also looked at social determinants of health and the same sort of factors that were in the vaccine equity index that I mentioned earlier and what they see. Well, when you look at the 10-year, they followed these people for 10-year mean follow-up and they saw that the annual cardiovascular disease mortality was about a hazard ratio of 1.54. When you adjusted for behavioral and metabolic risk factors, it really didn't change much. When you adjusted for social determinants of health, the differences fell off completely. So this is really what is driving these disparities. And so when we think about this, we see that cardiovascular risks and other risks are unevenly distributed. Morbidity and mortality that we have seen growing in the United States are not evenly distributed. And I would say more than anything, I think that there's a chronic low-grade stress that just goes along with being poor, that goes along with being in these zip codes that are bearing the highest burdens that we see in this equity index. And I guess I think that much, if not most of this increased risk is really related to social determinants of health. The next thing I would say is, what can you do about this? Well, I do think, again, we should be looking at this. So thinking about people's zip codes and addressing them directly. And this is not technically in the guidelines yet. It is coming, I would say, in the next couple of years. And what I would say is that this should be a consideration in treatment decisions. So for example, at a point when I worked for a Medi-Cal plan and Medi-Cal is the Medicaid for California, when they were picking the threshold for hypertension treatment, they used a tighter threshold. And I am gonna get into their multiple guidelines because they were making the case that this is a higher risk population and we need to actually treat to a greater intensity. And with that, I will go on to hypertension. And in the next section, I'll be talking about cardiovascular risk factors specifically. So blood pressure targets are sort of all over the place. I feel like sometimes my colleagues in psychiatry are like, oh, well, in medicine, everything is so stable and you have all these guidelines, and we debate them all the time. And people go back and forth about this. In 2017, the SPRINT trial gets published. And SPRINT was published in the New England Journal of Medicine about a year after the JNC9 guidelines came out. Now, those guidelines were kind of notorious at the time because they really liberalized the hypertension treatment. And they basically had people not getting treated until they had a blood pressure of 150 over 90. Those were in JAMA. And at the time, I said, well, the only thing that's going to change this is if somebody publishes this something in New England Journal. And the next year, that's what happened. So SPRINT gets published. And they found that having a really aggressive blood pressure target for people who are at a very high risk of developing a cardiovascular event in 10 years, so a greater than 20% risk, there was a dramatic decrease in the risk of stroke. So they did this. They published early. And the next year, the American College of Cardiology comes out with a set of guidelines. And they say anybody with a risk factor of 10% likelihood of having a cardiovascular event in 10 years, and they use the ASCVD calculator, you should aim for a blood pressure of less than 130 over 80. So in 2023, we get a new set of guidelines from the American Association of Family Practice. And they are much more liberal, not quite as liberal as the JNC-9 guidelines, but definitely tighter. So they had a blood pressure threshold of 140 over 90. They did not find benefits in terms of decreasing the risk of myocardial infarction or stroke, but they did find a mortality benefit. However, when they tighten the threshold a little bit to 135 over 85, there was a decrease in the risk of myocardial infarction. They did start to see a trade-off in adverse events. And that's mostly hypotensive events. And so I would say for many people, 135 over 85 is what people were looking for. At the end of the year, to complicate things more, in December, the European Society of Hypertension comes out with its own guidelines. It's a little harder to remember these because you start to have to do age cohorts. But they basically say 18 to 64, you use the guidelines from American College of Cardiology, but you don't actually risk stratify. So it's less than 130 over 80. 65 to 79, less than 140 over 80. And 80, basically 140 to 150. There are lots of different cardiovascular risk calculators. I would say the appropriate calculator to use is gonna vary based on age and sex. There are biases that go into all of these calculators. So I would encourage you to take a look at the calculators when you're looking at the individual person who's in front of you. So again, what we saw in 2023 is a couple of different guidelines. The AAFP of 140 over 90 and the European Society of Hypertension of 140 over 90. But they basically say, well, an optimal blood pressure is actually less than 120 over 80. So with that in mind, you have a goal. So how are you gonna get there? And so there are a few studies that come out that I think are important in terms of treatment. One is from JAMA in April, 2023. And this is in Sweden. So 258 people, small study, but really well-designed. So an entirely white European population. These were people who were otherwise at low risk for a cardiovascular event. Mean blood pressure about 155 over 90 when they started. And what they found is that blood pressure response varied considerably between individuals, between treatments and within individual participants. And what I mean by that is individual people responded to different drugs. They responded to the time of day they got them. As an aside, if I were gonna pick a time of day, if you have somebody who you know is gonna take their medication at night and always remember it, it's probably a little better to take the blood pressure medication at night in general. A lot of people forget if they don't take their medications right the first thing in the morning. But in general, there was some literature that supported that this year. But when they looked at this, they looked at all these different variabilities, they found that when you personalized your treatment, that you looked at the individual person in front of you, that you got an additional blood pressure benefit of about four millimeters of mercury. And why is that number important? Because that's the benefit you get from any monotherapy. So basically, you'll get basically four points. And what this means is if you personalize treatment, instead of somebody having to be on two drugs, they can be on one drug. Three drugs, it drops down to two drugs. So really, I think compliance drops off exponentially when we engage in polypharmacy with people. So I think it's a really important thing to consider. The next study I wanna mention is from Allhat. And Allhat is an incredibly seminal, important article and study in the way that blood pressure is managed. And it's so critical because Allhat is the longitudinal study. It goes out from 1994. And it tells us that treating for decades actually does work, is safe. So what this study looked at was a variety of different types of medication to see which one was really the best. So they looked at diuretics, they looked at ACE inhibitors, and they looked at calcium channel blockers. So lots of people on thiazide diuretics, some on calcium channel blockers. These are the calcium channel blockers, the non-inotropic calcium channel blockers that don't change heart rate. This is the ones that we're talking about, drugs like amlodipine, falodipine. And then ACE inhibitors. People were in the trial for four to eight years, and then they had 23 years of passive follow-up. I do wanna make a little asterisk about thiazide diuretics. These are actually great drugs in combination. Nobody gets to therapy goal on a single thiazide diuretic. But they are amazing as a synergistic drug. So if you start somebody on a thiazide and they don't get to goal, don't worry about that. You have probably, if you're gonna start a thiazide, you should start them on something else as well. So the primary endpoint in this study was mortality due to cardiovascular disease. And then the kind of typical secondary outcomes like stroke and heart failure and end-stage renal disease. So the cardiovascular disease mortality was similar in all three groups. The thing that was kind of scary about this one was ACE inhibitors increased the risk of stroke by 11% versus diuretics. This is an effect that persisted well beyond trial period and definitely, I would say, has got a lot of people nervous about ACE inhibitors. And people were rethinking that as a first-line therapy. The final thing I wanna mention in terms of hypertension is not about a drug, but in the way that we measure hypertension. And among the things that I think are important in a practical sense in clinical practice, this one I think might actually show up in your practices. So if you measure arm circumference, so basically from up here at the olecranon down to the elbow, and you take the arm circumference, it can be small, it can be regular, it can be large or extra large. But most people only use a single-size regular cuff when they're measuring blood pressure. So if you use that regular cuff and you've got somebody really skinny, you've got somebody who's larger, you've got somebody who's, say, has a BMI of greater than 40, you might expect there'd be some differences, but this study really jumped out at me. If you use a cuff on somebody who's really skinny, you're gonna overestimate and you're gonna actually miss blood pressure in those people. You're gonna see a blood pressure that's four millimeters of mercury too low. If you use a regular cuff and their arm is large, you're gonna get a reading that's five millimeters too high. If somebody is morbidly obese and you use a regular cuff, you're gonna overestimate their blood pressure by 20 points. So it's not that I don't think that being overweight is not a risk for cardiovascular events, but when we talk about the morbidity of overweight and obesity, and we say, well, there are all these comorbidities that are associated, part of what is happening is we are giving people who are morbidly obese diagnoses of hypertension that they don't have. Many people will get those diagnoses after they get in their 50s, as we all become more susceptible to things when we start to get in our 50s and older. But I will say that part of what we see in the literature on obesity, and when we say, well, there's so much more hypertension, some of that's true and some of it's just we're doing a really bad job of measuring this. So the what in hypertension, so we have, I think, a few different guidelines. They're not entirely consistent. We have a lot of studies on drug treatment and we have some guidance on how to measure blood pressure. Why about this? I think risk stratification really is important and many psych patients, I would say, when you think about risk stratification, they are going to be at higher risk. So as a bread and butter threshold, in general, I would not aim for 140 over 90 if somebody has severe mental illness. I think it's unlikely that that's gonna be the right measure for them. I think aiming for a goal of 130 over 80, maybe even a little tighter is gonna be a better goal when I think generically about people, especially those with severe mental illness. And we are probably over-diagnosing people with morbid obesity and we're under-diagnosing, frankly, people who are lower weight. Then now what I would say is, again, I would have a goal of at least 130 over 80, give or take, maybe even tighter. I do think it's really important when you're assessing risk to think about social determinants of health. You can use all these fancy calculators, but if you're not looking at people's zip codes, if you're not looking at those risks, you're probably gonna be under-treating them. The meds, frankly, it's a dealer's choice, but personalized treatment is going to cut the total number of meds that people need to be on and use the right cuff. If you're gonna do a measurement, just use the right cuff on the person. Otherwise, it's a waste of time to do the measurement. Only a couple of studies on hyperlipidemia this year, but I think a couple things that are practical to know about. So when we think about statins, and this study came out in JAMA in April 2023, there's been this feeling for a lot of years that we should have statins in the water, and we should just push them to as high as they can, because really, the higher the statin treatment, the better, and that we should aim for at least a 50% reduction in LDL. There's an alternative approach, though, and that is that we know that there are some harms with these medications. So we know that people who have prediabetes, many of them will tip into full-blown diabetes when we start these medications. We know that for some people who have diabetes, it makes their disease harder to manage. So the other approach is to have a lower-dose statin and to titrate to specific LDL goals. The other thing is, frankly, not every person's going to be somebody who wants to come in or can come in multiple times a year to get lab work done. So if you're coming up with a strategy, also, again, we wanna meet our patients where they are at. So in this study, it's a randomized multicenter trial, and this was in South Korea. These are people who had known coronary artery disease, and there were 4,400 patients, about a quarter of them women, and what they find, that it really didn't matter which strategy you took. So if you treated to target that LDL of 50 to 70, or if you had high-intensity treatment, I will say this one was strange to me, and the authors didn't comment on it. A 12-statin 40 milligrams isn't really as high as you can go. It maxes out at double that dose, but basically, high-intensity statin treatment, and really, when you look at this, again, no difference between either strategy. The next study I wanna mention is looking at dietary supplements, and so this was looking at a low-dose rosuvastatin, which is Crestor, I think, is the brand name of it. I think it might be generic now, I hope so, versus dietary supplements. So a lot of people take supplements, and I do think it's something worth looking at. So these are people who did not have any history of coronary artery disease, but they were on a risk-scoring system, the ASCVD, found to be at increased risk of having a 10-year event, and they looked at people, and they were either assigned to get rosuvastatin at five milligrams, or an alternative, and that could be placebo or a variety of different supplements. I will tell you, I have taken fish oil supplements at times. You cannot do a placebo of a fish oil supplement. What they found is when you gave somebody a five-milligram dose of rosuvastatin, really a whisper of it, the LDL dropped about 35%, and for the alternatives, there was no change in the LDL cholesterol. So when I think of high cholesterol, really not a lot of new literature this year, a lot of it really did focus on that risk of diabetes, but I do think we have this high-intensity statins that get often recommended, and we also know that a lot of people take supplements. So in terms of risks, I would say the risks of harm are real, including the progression of diabetes, and with regards to supplements, the benefits are known, and again, it's hard to randomize. People do know when they're taking fish oil. So now what? I would say high-intensity treatment. I would think about this, again, if we're thinking about an equity lens. People who have less access to care, people who, for whatever reason, can't come into the clinic multiple times a year to get labs, high-intensity treatment might be a more effective strategy, and then treating to target could be a better strategy for somebody with more reliable access or somebody who has a higher risk of developing diabetes. Even modest doses of statins show dramatic improvements. Now, full disclosure, I do take some supplements, but I take them in gummy form, so it turns out if I am wrong, I get to start my day with candy, which I think is a good thing. Next I wanna talk a little bit about diabetes. So in this study, we look at the effects of lifestyle interventions. This one I do think in the next couple of years really is gonna change practice. It's not quite there yet, but I think it's incredibly important. So this was published in Diabetes Care in November 2023. This is a systematic review of four trials, about 3,000 participants. And they looked at people who had, they had prediabetes, and then they had a continuous glucose monitor placed. And so they looked, and they basically found three patterns. There are some people who had isolated impaired fasting glucose. Wake up in the morning, your blood sugar's high. Then they had isolated impaired glucose tolerance. I eat a meal, my blood sugar goes up. And then they had people who had both. And they looked at, did a lifestyle intervention make any difference in people progressing to diabetes? And what did they find? They followed people for it with a lifestyle intervention for two and a half years. For people who had impaired fasting glucose, no difference if they had a lifestyle intervention. These are people who are watched pretty closely for two and a half years. And actually you did see a difference for people who had high blood sugar after they ate. And you did see a difference if people had high blood sugar after they ate, and they had impaired glucose tolerance in the morning. So when I look at this, it really does mean that we have to look at this as a heterogeneous disease. Right now we are not recommending routine continuous glucose monitoring for all people we are diagnosing with prediabetes. I think this study really starts to lay the groundwork. And my guess is we are here where we were, I would say five or six years back, with ambulatory blood pressure monitoring, where we realized we do need to get a 24-hour reading of people's blood pressure to get a better idea of what their real blood pressure is. I do think if we're going to make an appropriate recommendation, we do need to understand what kind of impaired glucose tolerance people have. The next article, and I would say, if you're gonna pull one article to have as a reference, this is a meaty one, but I think it's got some really great tables and really good comparison charts. This is from the British Medical Journal, and it came out in April 2023, and it looks at the benefits and harms of type 2 diabetes drug treatment. So there are a couple of drugs that got added this year. One was Finarone, which is a nonsteroidal mineralocorticoid NSMC receptor antagonist, and then Terzapatide, which is a GIP receptor, GLP-1 receptor agonist. I hate saying these words, so I'm just gonna say GIP and GLP-1. So this was a systematic review and meta-analysis. There were 816 trials, almost 500,000 patients, and 13 drug classes, and what did they find? The only two classes of drugs that had a high certainty of a mortality benefit were the SGLT drugs, the Flozans. I'm gonna mention a brand name just because these are difficult to pronounce, and sometimes people think of them that way. So Farxiga and Jardians are examples of brand names of these, but Flozan, when you look at the generic, and then the GLP-1s, which semaglutide and loraglutide, and I will mention that loraglutide is coming out as generic next month. So yay, we actually will have a generic of one of these. We did also see a mortality benefit with the NSMC antagonists as well. moderate benefit, but I think it's real. They also looked at a couple of other issues. So they also saw that for these two classes of drugs, there were decreases in the rates of cardiovascular death in non-fatal MI, in hospital admission for heart failure, and in end-stage renal disease. The non-steroidal mineralocorticoids decreased hospital admissions for heart failure as well as hospital admissions for end-stage renal disease. Finally, the GLP-1 agonists decreased non-fatal stroke, and the SGLT2 inhibitors were better than any other drugs for end-stage renal disease. In terms of quality of life, the GLP-1s and trazepatide, which is the dual agonist, the GLP-1 GIP agonist, and probably the SGL2 inhibitors improved quality of life as well. In terms of harms, the SGL2 ones cause genitourinary infections, so you get more urinary tract infections, some genital tract infections. The GLP-1s, you see severe GI effects that for some people lead to them not being able to tolerate them. And then the phenerone side effects, sometimes you have severe hyperkalemia, severe enough that actually people can get hospitalized with this. You also, I think not surprisingly, and we'll be talking separately about the GLP-1s, see the weight reduction. And I do think this becomes part of why people comply with them, and part of why they work is that people stay on them, and they stay on them because they lose weight. About a 19 pound average weight loss with trazepatide, and basal insulin, you see about a five pound weight gain. And again, about a six pound weight gain with this, I can never pronounce these, sorry. So I think next going on is looking at the effectiveness of the GLP-1 receptors. So we have this big systematic review in BMJ, about a half million patients, and we're really only seeing mortality benefits with the SGL-2s and the GLP-1s. So how do these drugs work in terms of, if you have somebody with type two diabetes, do they affect other cardiovascular risk factors? So in this study of the GLP-1 agonists, they looked, as a meta-analysis, there are 76 trials in about 40,000 adults. People were followed for about 12 weeks. And they found that in addition to, you have this HbA1c drop, so long-term glucose control, better fasting glucose, and about exact same drop in body weight, about a 19 pound drop in body weight. Semaglutide, cut those numbers in half, but exactly the same direction, decreases in A1c and glucose and body weight. So when we look at this then, I think one of the criticisms we have with these drugs is they are really expensive. So I like this study that came out in the Diabetes Research and Clinical Practice Journal in April, that looks at the GLP-1s and long-acting insulins and what they cost. So this was a study in Taiwan, looking at a national healthcare database and people who were getting treatments over 10 years. So it was about $6,000 a year per quality adjusted life year. But what I want you to really pay attention to is that for specific populations, they actually were pretty cost-effective. So if somebody had known cardiovascular disease, it was about $673. People who you knew definitely didn't have cardiovascular disease, though, that went up to about $9,000. People who had known kidney disease, again, significantly more cost-effective, and again, significantly less cost-effective if people had no renal impairment. So when you look at diabetes, I think we have, again, we often will recommend lifestyle as a first line for prediabetes. And we have a lot of medication options, and some of them are really, really expensive. So what I would take away from this, if somebody comes into your office, they have prediabetes, and it turns out they only have impairment in their fasting glucose, lifestyle alone is not going to help their progression to diabetes. Doesn't mean they're not good reasons to recommend exercise and diet, and you can do all that, but write the script for metformin before they leave the office. The next thing I would say, the SGLTs and the GLP1s, these are the drugs that have the best evidence for mortality benefit with diabetes. And the cost-effectiveness of them, though, I would say is really limited to known cardiovascular disease and kidney disease. So at this time, I would say, strongly consider continuous glucose monitoring to personalize treatment in prediabetes. The SGLTs and GLP1s, they are expensive, but they work, and the most cost-effective ones, again, are gonna be for people with known cardiovascular and kidney disease. Next, I wanna transition over to the GLP1s, and this is a picture of Svetlana Mazjev. So she's a researcher who, until not long ago, worked at Mass General. When she did her post-doc, which is at Rockefeller, she studied the GLP1s, and she stayed on a few years after her post-doc work to finish this research, and she went to Mass General. This is completely done work, and Mass General loved this work, and so did the men who worked there, and they loved it so much that they took credit for her work. And lots of other organizations, including the Government of Canada last year, gave the men who... When somebody's lying, I like to use the word disingenuous if I wanna be polite, and I don't know a polite word for stolen, but that's what they did, and that's... And she basically had to call out the Canadian government and say, you know, actually, I'm the person who discovered this, so I'm just putting that out there as an aside. Anyway, with that in mind, the GLP1 receptor agonists. One of my pet peeves with these drugs is people say, remember fen-phen? I don't wanna do this. This is too early. So I do wanna make a couple of comments about this. The research on the incretin peptides started in the 1970s. 19 years ago, the first GLP1 was released to the market. This was Exanatide, which is Bayetta. In 2023, the reason we are all hearing about this is really twofold. One is that there are long-acting formulations that people could take on a weekly basis, and so people started seeing that there was weight loss, a weight loss benefit that extended to people without diabetes, and we also, though, I think in many ways more importantly, started to see a literature for cardiovascular benefit that was both in people with and without diabetes. As I mentioned earlier, in a month, we're gonna see loraglitide, which has been on the market for 10 years. We anticipate a generic's gonna be released. So there are many reasons to criticize and critique the GLP1s and the mania, frankly, that surrounds them. To say that they are brand new, I would say, is not a critique I would be in agreement with. Another thing I hear is that people say, well, we just don't understand how these drugs work. Well, yes, we do understand how they work, and this is in the slides, and we can go through this. Let me go to the next slide, though, to just kind of talk through some of the ways they work, because they work in lots of different ways. In the liver, they basically decrease the gluconeogenesis, and so you get lower blood sugar. They also decrease steatotosis, and so this is part of why they work for nonalcoholic, for NASH, nonalcoholic steatototic hepatitis. In the brain, they decrease food and water intake, and the hormones that are associated with this. They decrease inflammation, and they increase learning and memory. There's pretty solid literature that they decrease the risk of dementia. In the pancreas, they increase insulin, and they decrease glucagon. In the heart, they increase A and P, which leads to an increase in cardiac contractility and cardiac output. In the kidneys, they increase diuresis and natriuresis, and in the stomach, they decrease gastric emptying, and this one I really like. They convert white fat, the bad fat in our body, into tan and brown fat, which the body can easily use for energy. Among the side effects that people will talk about, I did want to make an aside about this one. A lot of the side effects we see with these drugs are the side effects that tell you they work really well for rapid weight loss. If somebody loses weight fast, the liver will secrete extra cholesterol into bile, and it'll form gallstones. The same side effect profile you see with these is the side effects profile you will see if somebody gets gastric bypass surgery or gastric sleeve surgery. These are not inherently unique to the GLPP-1s, especially the gallstone side effects. These really are much more about rapid weight loss that we see in many people. So in this study that I'm gonna talk about that came in the Journal of Diabetes and Obesity and Metabolism in 2023 in December, it's a meta-analysis of seven randomized controlled trials, and it looked at trizepatide and its effect on blood pressure and lipids, and I apologize, I had started talking about this in the last section, so it's now in this section. Kind of went back and forth about where to put the GLPP-1s and diabetes are here. So what they found is independent of the benefits you see with these drugs with diabetes, and the diabetes, the markers like HbA1c and glucose. We also saw that there was a drop in systolic blood pressure of four points, and again, think about that four points is the four points that you would see with monotherapy. So people who are on these drugs, oftentimes they can either go off of blood pressure medication or they can be on fewer medications. There's also a drop in cholesterol that we see. We see an increase in the good cholesterol, HDL, and we see a decrease in the bad cholesterol, so LDL and a decrease in triglycerides. The next study, again, when we look at this with cardiovascular outcomes, this becomes important because this is in people who don't have diabetes. So this gets published in November in the New England Journal of Medicine, about 18,000 patients. These are people with prediabetes, the mean HbA1c is 5.8, and these are people who had had a history of cardiovascular disease, and we wanna see what are their outcomes, they don't have diabetes. So this is a 40-month follow-up, and what did they see? They see a drop in myocardial infarction and stroke in cardiovascular death, and this is statistically significant. It drops from 8% down to 6.5%. I wanna get back, again, to cost on this one, and they do comment on this in this study. So when you look at this, these drugs cost about $17,000 a year, and to prevent one major cardiovascular event, you need to treat 70 patients for three years. Now that's not a huge number of patients. It is a really expensive treatment from a system level. So next, we look at semaglutide and its effect on heart failure, and again, when we think about formularies having worked in a Medicaid plan, we look at the cost of drugs. So this was its use in heart failure with preserved ejection fraction and obesity. This is in August 2023, again, a New England Journal, and these are people who have preserved ejection fractions, but a known history of heart disease, and they get a weekly injection of semaglutide, and this is after a year. They have a 13% loss in body weight, 3% with placebo, and their symptoms are improved. Their physical limitations are eased. Now I would contrast this even with the most concerted efforts. When we are thinking about weight loss before the GLPT-1s, we would say, if you got somebody to lose 5% body weight in a year, that would be amazing. This 13% body weight, this is not the way things have been before, so some of these benefits are just weight loss, but some of it is, again, these drugs increase ANP, they increase cardiac contractility. Next, though, I think this is sobering in terms of what do these drugs mean in terms of are they gonna work long-term? So this is a study of trizepatide, and this is in Lancet, and it's published in August 2023, and this is the Surmount trial, and this is the sort of long-term follow-up trial. So there are 900 adults, and we have an HbA1c of seven to 10. These are people with diabetes, 72-week follow-up, and we see that 13% to 15% of their body weight gets lost if they have diabetes. Now, here's the part of the study, because they did a sub-analysis of the people who didn't have diabetes. The thing we really don't know yet about these drugs, they work better for weight loss for people who don't have diabetes. They work for people who have diabetes, but for people who don't have diabetes, they are better. More than 50% of the people in this study of trizepatide lost at least 20% of their baseline weight, and this is compared to 3% for placebo. And they did look at the long-term maintenance, so in that one, but I think it's even more pronounced in this next study I'm gonna mention, which gets published in JAMA. Same cohort, but sort of an extension of this, and this is in January of 2024. So 783 participants on a pretty high dose of trizepatide, and as a reminder, trizepatide is that dual agonist, the GLP-1-GIP agonist. There's 670 participants, and they get randomized one-to-one for 52 weeks. And then after this, they continue to either receive trizepatide, or they get switched to placebo. So the primary endpoint they're looking to see, did these people lose weight, and did they keep it off if they got switched to placebo? So about 70% of the people in this study are women, and what'd they find? So the mean weight reduction in the 36-week lead-in period was about 21%. We have never had numbers like this, and these are the numbers long-term that we see with gastric sleeve surgery. Nothing other than surgery comes anywhere close to this. What they found is that if they stayed on trizepatide, they continued to lose weight. So they went from that 21% loss, they lost another 5 1⁄2% after that. This is compared to placebo, where after they were switched to placebo, they gained 14.1, they gained 14% weight. So when they looked at this long-term, so at 88 weeks, 80% of people maintained their initial weight loss overall. So trizepatide, it was 90%, placebo, 16%. And at 88 weeks, the people with weight reduction, and again, the trizepatide total, because they continued to lose weight after that initial weight loss, was about 25%. These are phenomenal numbers. But again, I do wanna go back to cost, because really this is one of the limiting factors. So this comes from a white paper that was published by Prime Magellan, which is a PBM, a pharmacy benefits management company, and they looked at the cost of these drugs. And so what they found is kind of sobering on this. So they found that among people who are on the GLP-1s, compared to a matched control group, the cost for these people went up almost $8,000 a year. Persistency was poor, and they thought, well, maybe this is because of access. But among the people who were actually adherent to the drugs, their costs were even higher. They were double that of the prior year. So when I think about the GLP-1 receptor agonists, arguably the story of the year, I would say there's convincing evidence for the role of GLP-1s in weight loss in people without diabetes, even stronger than for those with diabetes. And there's also really a significant cardiovascular benefit that we are seeing. Again, this is both in people who have diabetes and who don't. So why should you care about it? Well, these long-term results approximate bariatric surgery. But not everybody's gonna tolerate them, and they are incredibly expensive. With the GLP-1s, I think what we see with this literature, lifetime maintenance is gonna likely be needed. So these are not great drugs to start if you don't think people are gonna be able to stay on them, whether their insurance is gonna cover them, or whether they're gonna have the ability to private pay. If they can't stay on them, these are tough drugs to keep the weight off if they don't stay on them. There are huge issues with cost and with access. Next, I wanna transition to a couple of studies on screening and this is Gray Muir, who's one of the authors of the Oxford Textbook of Medicine. And he has a quote that I like a lot. All screening does harm, some screening does more harm than good. So in 2021, the USPTF and the American Cancer Society recommend screening for colon cancer starting at age 45. In 2023, the American College of Physicians issues a guidance statement saying that there is not a benefit to justify, there's not really a literature to justify screening earlier than age 50. And as a compliment to this, we see an article, a study that gets published in the Annals of Internal Medicine in August, 2023. And what they did is they went back to the micro simulation models that USPTF used and they found that they think that they overestimated the net benefit. They do say that screening definitely should stop by age 75 or life expectancy less than 10 years. A couple of things that I think are worth mentioning, it really doesn't matter which kind of screening in terms of FIT, the fecal cult testing, a colonoscopy or a flex sig plus FIT, but they recommend against using stool DNA testing. so I include the brand name only because I don't think there is another alternative option. These are incredibly expensive, they have a very high false positive rate, and right now we don't have morbidity and mortality benefit data for them. That might come at some point in the future, but it's not here now, and so in terms of doing this as a cancer screening, ACP does recommend against it. I do want to mention then this study that comes out in JAMA in November 2023, and I mention this keeping in mind that I know the breast cancer screening guidelines just changed a week ago, but this looked at screening tests that get promoted as increasing longevity, but people don't really know if people are going to live longer with common cancer screening tests. So they looked at mammography for breast cancer, colonoscopy, stigmoidoscopy, and fecal occult blood testing for colon cancer, CT for lung cancers in smokers and former smokers, and PSA for prostate cancer, and what they find. Looked at 2 million people, and they followed them for 10, 13, and 15 years respectively for lung cancer, breast cancer, and colon cancer. The only test that had a lifetime gain was stigmoidoscopy, 110 days. Colonoscopy on a population level, zero days, PSA, 37 days, colonoscopy, 37 days, fecal occult blood testing, 37 days, and lung CT, 107 days. We all know individuals who have had these screening tests, and again, I would say that these are population level numbers, but to say there's a population level benefit for these things, that is, I think, cherry picking the literature. This is a consistent finding that we see that screening, as much as we would like it to on a population level to show a benefit, we really don't have a published literature that would support that. So for screening, I would say the big things I take away from this year, we have updated colon cancer screening guidance, and the value of many cancer screens has been questioned. So why do we care about this? So population screening studies probably have limited applicability, and that they are important though, because from a public health and insurance standpoint, we know that coverage determinations are going to drive access. In terms of now what, well, colon cancer screening really probably is worth doing, the other ones on a generic recommendation level, I would say the others really do have to involve a discussion with individual patients and having risk benefit discussions at an individual patient level and having those kinds of risk benefit discussions. Next I want to talk about diet and exercise to finish up before I go into just one or two things on tech. We looked at the effect of diet on blood pressure, and this is a study that came out in JAMA that I do think is actually important and relevant and actually pretty easy to put into practice. This is a randomized controlled trial looking at blood pressure. This is a study out of University of Chicago, and so a non-white population as a majority. So this was just one week of doing this trial. This is actually pretty easy to do in practice. So high sodium diet versus low sodium, and lest you think high sodium was extra and really hard to do, that's an extra teaspoon of sodium. Doesn't take a lot of extra to get a really high sodium diet. Low sodium, nobody could ever be on 500 milligrams. You might aim for this. Nobody actually gets to that. So what they find is a usual diet is actually about four and a half grams of sodium a day, and then you have this high sodium diet. You end up with people who are actually even a little higher than that. As I said, they were aiming for 500 milligrams. The best they could get with this low sodium was like 1.3 grams. What they found though is reducing intake by about a teaspoon a day, and again, this is after a week that you could see this benefit. It dropped blood pressure by about six points, and remember I said four points is what you get from a drug. And how likely is this to work? For about three quarters of people, dropping your sodium by about a teaspoon a day is going to drop the blood pressure about as much as taking another pill. So again, when you think about the things that we can start to take away medications, a low sodium diet for many people is going to drop their blood pressure. Now paradoxically, for about 25% of people, the blood pressure actually goes up on a low sodium diet. We don't know exactly why that is, but I do think this is why, you know, if you're going to do this, you could try it for a week and then have them come back in and measure it. Next study I want to mention is about plant-based diets and do they protect against cardiometabolic risk. So this is looking, a systematic review and meta-analysis that was in JAMA. People go all sorts of different diets. I will mention a couple of different diets. So you know, here's the literature for a vegan diet this year. What they looked at and they found is that mostly people lost about seven and a half pounds if they followed a vegan diet for about six months. HBA went down, HBA went down kind of on the margins, like a quarter point, and the LDL went down. It was significant, but by about six points. No real change in blood pressure. Another study on vegetarian or vegan diets, I think each year we sort of pick the ones. A few years ago it would have been Mediterranean. We see a lot more literature for vegetarian, vegan this year. So this is another systematic review, 30 trials, and they compared this with omnivores, plant-based diets, and what they found is drops in total cholesterol, drops in LDL, and drops in ApoB. And the other trial type of diet we saw a lot of literature on this year was time-restricted eating. Now here's what I want you to take away from time-restricted eating, and let me just jump over to this. So this is, again, a study out of University of Chicago. People had a mean HBA1C of eight and a mean BMI when they started of 39. So this is eight-hour time-restricted eating. They only ate from 12 to 8 p.m. They did not count calories, and they compared this to caloric restriction, about a 25% caloric restriction a day. And they looked at changes in body weight as well as some secondary outcomes. The thing to take away from time-restricted eating diets is they aren't magic. People eat fewer calories. That's why they work. So you don't have to try on these diets. That's why they work for a lot of people. If they get in the habit of this, they will end up eating about 313 fewer calories a day, and this is even versus if they're counting calories. So again, no specific magic to them, but if people time-restrict their eating, in general, they're going to eat a couple hundred fewer calories a day than they would otherwise. So people end up losing about eight pounds on time-restricted eating, and that's versus a caloric restriction. Now, the caloric restriction group doesn't reach statistical significance, but it's close. So I'd say time-restricted eating, if somebody can stick with it, that's fine. Didn't see any differences, though, in HbA1c or blood sugar range or blood pressure or cholesterol. I do want to mention this one. This is from a few months ago. Whenever the New York Times and Fox News agree on something, you probably should know about it. They both published lots on this. So is intermittent fasting bad for your heart? So I just said there was no change in any of the cardiovascular risk factors with this study, but people with the intermittent fasting did lose weight. This is a poster that got presented at the American Heart Association. Everybody got bothered about this 91% increase in the risk of cardiovascular death if you have time-restricted eating. 20,000 adults, and what do they do with this? So an absolute increase in the death rate of 4%. A couple things about this poster. It's a poster. It's not been peer-reviewed at this point. The groups were very different. The time-restricted eating group had 40% heart disease to start with versus 15% in the control. They didn't include anything about family history or other cardiovascular risk factors in the poster. They also were different. There were a lot more smokers. They were more overweight. They were just sort of a different group, and they also based their predictions for eight years on two food recall studies that were done at the beginning of an observation period. I don't know. I can't look back to eight years ago and remember what I was eating then, but that's what they based the 91% on. So this is one I would say, you know, in the popular press, people will talk about this sort of thing. You know, we don't really know. The study has not been published yet. We don't even know if it's going to get published, so we'll see. I do want to mention every year the British Medical Journal comes out with their Christmas Journal article, and they always have some fun articles that come out, and this one was on chocolate brownies and caloric restriction to see if offering brownies on the wards would change people's perception of intermittent fasting. They didn't, but they liked the brownies, so I would put it out there that they also include the brownie recipe in the article. It's a really good recipe, so even if you're not going to read the rest of the article, I would recommend the recipe. It's really good. So the what in diet this year. There's always going to be some studies on diets and which one's the best and which one's the worst. Getting an idea, though, of what people are eating and how much really is probably healthy, and diet does impact cardiovascular disease, but I would say it's caloric restriction that matters, and it probably, frankly, doesn't matter what diet. Every year, people talk about which diet is best, and it's the diet somebody can stick to and cut calories. So if you have a person who can stick to keto, have them do keto. If they do it with intermittent fasting, that's fine, too. If they can do it with Mediterranean, the diets that somebody can stick to to limit calorie intake is probably going to give them the best cardiovascular benefit. As for salt restriction, for about three quarters of people, it is going to work, but there is that quarter of people who will paradoxically have an increase. So it is something, very quickly, you're going to know if it works. So if you're going to have somebody do this, just have them come back. The other thing I would say, though, just for salt restriction is it's really hard to be on a very salt-restricted diet and eat a lot of ultra-processed foods. So this may, in some ways, be a proxy for ultra-processed foods and may be part of why we see this paradox. I do want to mention a few studies on exercise that came out this year as well. So this is a study looking at the addition of exercise to a low-calorie diet on people with type 2 diabetes to see did they get any extra cardiovascular benefits. And what they saw, so systematic review and meta-analysis, and people did walking or jogging, stationary bike, or they played soccer. This has to include the number of weeks to remind you that meta-analyses are interesting. So two-week studies and 52-week studies were all included in this meta-analysis. I'm not sure that that's really a cohesive group, but take that for what it's worth. They saw improvements in cardiopulmonary fitness, but they didn't see increases, benefits in terms of weight, in terms of blood sugar, in terms of body fat or muscle. This next study looking at exercise training and blood pressure, we do see some benefits here and I do want to mention some of the ways that they did this. So this is a meta-analysis, 270 randomized controlled trials, about 16,000 people. And they found a variety of different benefits. So aerobic exercise, again, four points is the magic number for can I get somebody off a drug. If somebody is doing aerobic exercise five times a week, four points drop. About dynamic resistance, so some sort of weight training, four point drop. Doing a combination, you get about a four and a half. So you get kind of a hybrid, so you get a six point drop. High intentional interval training, again, you go back to that four point drop. An isometric exercise, you get an eight point drop. I would put in there that a lot of isometric exercise is going to include things like yoga. So to the extent that this is a proxy for mindfulness and maybe adding some mindfulness to the others would have given additional benefit, that's not clear. But I would put that out there as something they didn't discuss, but might be confounding the results. Next I want to mention this study on aerobic resistance or combined training in people and their cardiovascular risk profiles. And this is looking at people who are overweight or obese. And they randomized these people to resistance training, aerobic, and resistance or aerobic. And they looked at some different cardiovascular risk factor parameters. What they saw is people who did just aerobic exercise or they did aerobics plus a resistance training had improvements in a composite score. When you looked at, you lumped it all together. When you look though at individual elements, body fat was actually improved in all the groups, but they didn't see individual benefits for systolic blood pressure. They didn't see individual benefits for LDL and they didn't see individual benefits for fasting glucose. So when I take all of this literature and put it together, so what do I take away from it? I think the caloric restriction really is key, that exercise might not give you additional benefits in terms of blood sugar and weight. For hypertension, you're looking to see, you'll probably see exercise benefits in the range of four to eight millimeter drops in mercury and blood pressure. And that overall cardiovascular risk, maybe not so much of a benefit, but you will see some decreases in body fat. So I think in terms of diabetes, you need to set the right expectation. There are a lot of good reasons to exercise, but lifestyle benefits are going to vary in people with diabetes. And going back to the study I mentioned earlier, the type of impaired fasting glucose that people have is going to impact their response. So many of these studies that have had variable benefits, the thing I would put as an asterisk next to them is that we don't know if these were mostly people who had fasting glucose or these were people who had more of a postprandial glucose elevation. But when we think about the benefits, I would say there is likely to be more of a benefit in people with that postprandial elevation, so it's kind of looking at the continuous glucose monitor. Exercise benefits, again, looking at four to eight millimeters, that's really potentially one or two drugs you can take somebody off. Do you think there's a potential confounder for isometric exercise with mindfulness? Probably not a bad idea to add some sort of meditation if people are going to do any kind of exercise. And then resistance training alone in terms of overall cardiovascular risk probably is not sufficient. So if you want to live longer, what do I take away? You cut calories. If you want to live better, you should exercise, and if you want to do both, you should do both. Let me just finish up with technology, and there are a couple of studies on tech. So let me say a bad thing, since I'm going to work for an AI company, I'm going to say something bad about AI to start. So the Annals of Internal Medicine, September 2023, and they looked at adding AI to colonoscopy detection of advanced neoplasia. There were a few different studies on AI with neoplasia, and a couple were good, a couple were bad, and this one was one of the bad ones. So basically, they found that the addition of AI to colon cancer screening, no difference than the results that you got for not having an addition. It doesn't mean I don't think AI is going to have an important benefit in terms of technology, in terms of its role, especially in cancer screening in the next couple of years, but I do think that right now, we don't have great evidence for it. I do want to finish up, though, with this study, which I found a little sobering. So this was a study, I think we all know that CHAT-GPT, this was the very first year, 2023, the first year we had a full year of CHAT-GPT-3. And so these authors, in May, decided, well, let's take a look at the ability of an AI to generate a medical article in the literature. So the authors posed questions. This was, it was a psych article, so for what it's worth, the goal was to create a completely fabricated article. And so they looked at this, and they had reviewers from neurosurgery, psychiatry, and experts in statistics, and they compared this to similar existing articles. After one hour with no special training, CHAT-GPT-3 had an AI-generated article that included a standard introduction, materials, results, methods, discussion, and data sheets. It was 2,000 words, had 17 citations. There were a couple of errors in the references, but most of the reviewers were fooled by it. So as to leave you today, I want to give you this word on how can you find a CHAT-GPT study. And with this, I will say Medium in December 2023 came up with CHAT-GPT's favorite words. So words like dive, and explore, and aesthetic, and seamlessly, and realm, and world, and illustrious, and wavering, and additionally. But by far and away, the word that CHAT-GPT loves the most is delve. So with that, I'm going to, any guesses to what happened with the, if you do a PubMed search of the word delve in 2023? This is small, but I'm going to, I hope you'll go with me on this. So this is 2023 and the appearance of the word delve in the medical literature. So when you're looking at this, I think, just food for thought. Anyway, with that, thanks so much. I really appreciate everybody staying.

medical advancements

health equity

COVID-19

vaccine guidance

health disparities

hypertension guidelines

GLP-1 receptor agonists

cancer screening

dietary approaches

AI in healthcare

cardiovascular health