Welcome. Good morning. Thanks for braving the threats that a bike race is going to close down the city and all of the rest. I'm Evelyn Atiyah. I'm a psychiatrist. I work at both Columbia University and Weill Cornell Medicine, and you're going to hear a bit of a review today about eating disorders, which is an area that I've been focused on in both clinical treatment and research for a whole bunch of decades now. There are some in the room who may know just how many, but I'm not going to say it out loud. What I understand in terms of the structure today is that when we have an opportunity for questions, we get questions both from this group and a group that's online. That's going to send me text-based messages that will ask their questions and we're going to try to accommodate both. We'll save the questions for the end, but there should be time for questions and discussions at the end of the talk. As I say, we're doing a bit of a review, hopefully with some attention to recent developments in the field of eating disorders. Just wanted to start with some disclosures. In the last year, I've received research support from NIMH and from New York State Office of Mental Health. I've gotten some royalty payments for things that I've written, an honoraria for grand rounds that I've given, and I also serve as a clinical advisor to a company called Equip Health. Here's an outline of what I hope we can do today. We'll talk about what the eating disorders are, some about who develops eating disorders, and why we need to know that. I'm going to focus on the general practitioner, not the specialist. What can the general practitioner do about eating disorders? With some attention to the treatments that we know to help. I'll say a little bit about what we know about the neurobiology of eating disorders, but I'm really going to focus on anorexia nervosa, which is the disorder that I am the most familiar with, as far as neurobiology is concerned. And again, try to say something about what's new. What's new in some of the labels we use, some of the presentations of patients, and some of the treatments that are coming down the pike, maybe. So, to start with what these disorders are, we've got a list of six in DSM-5. And some of these may be more familiar to you than others. We're going to focus today on the top four. So, anorexia nervosa, bulimia nervosa, binge eating disorder, avoidant restrictive food intake disorder, or ARFID. They're the most common, and they're the ones that we know the most about. So, again, while I understand that many of you are quite familiar with these conditions, let's just go over how they're currently defined. There are three diagnostic criteria for anorexia nervosa. Energy intake is taken in at an amount that's less than one's nutritional requirements, leading to a significantly low body weight. Significantly low body weight isn't specifically defined, but DSM-5 is insisting that it means significantly low in some way that the medical providers are defining for that individual. That can be after a weight loss. That can be failure to gain, as would be expected for a growing teen. There's also a cognitive component that's as or more important in these folks. There's a fear of becoming fat, despite that low weight, or persistent behaviors that interfere with weight gain. So, someone doesn't have to come out and say, hey, I'm petrified, but they can be engaged in a whole range of behaviors that serve to interfere with the normalization of weight, and that counts. There's a preoccupation with body shape and weight, and or, although for most it's and, a denial of the seriousness of the illness. There are two subtypes of anorexia nervosa, the restricting type and the binge eating or purging subtype. And these subtypes just describe the behaviors engaged in by the patients to restrict intake to achieve that significantly low body weight. Bulimia nervosa is a pattern of recurrent episodes of binge eating, and a binge is defined as eating an objectively large amount of food for the given circumstance, together with a feeling of loss of control. The recurrent episodes of binge eating are accompanied by recurrent episodes of compensatory behaviors to prevent weight gain, and most people colloquially call these compensatory behaviors purging. The episodes occur on average once weekly for a period of at least three months. And there's, like with anorexia nervosa, a preoccupation with body shape and weight. But in the case of bulimia nervosa, anorexia nervosa is not present. So the DSM-5 is really set up with this triage of you decide on anorexia nervosa first, and then you move on to bulimia nervosa. So by definition, these are not individuals who are significantly low in body weight. Because if they were, we would consider them to have anorexia nervosa binge purge subtype. But they've got normal and sometimes above normal weight. What about binge eating disorder? So this is a condition where there are recurrent episodes of binge eating defined identically to what I just described in bulimia nervosa. The binge eating episodes are associated with three or more of the following. Eating more rapidly than normal, eating until feeling uncomfortably full, eating large amounts of food when not feeling hungry, eating alone because of feeling embarrassed about the amount of eating, feeling disgusted with oneself, depressed or guilty after eating. In general, there's marked distress around eating and binge eating. The binge eating occurs, again, that same frequency requirement for threshold at least once weekly for at least three months. And the binge eating is not associated with the compensatory behaviors that you see in bulimia nervosa. And for binge eating disorder to be diagnosed, it's not present during an episode of anorexia nervosa or bulimia nervosa. So again, we sometimes have individuals who meet criteria for anorexia nervosa and they have some binge episodes. And they will present saying, hello, I've got binge eating disorder. But we really are asked to run through the different conditions in a bit of order. And if they are significantly underweight and have those other behaviors that were described, they would have anorexia nervosa, binge eating or purging subtype. But those who don't have that significantly low body weight would quite possibly be considered to have binge eating disorder. For some, this label, this condition is new. For others, you may have heard about it but aren't sure exactly what it is or how it's distinct from some of the rest of what we've described. So avoidant restrictive food intake disorder, which quickly adopted its acronym of ARFID, is a condition where there's avoidant and or restrictive intake due to one or more of the following, and really as we're getting to know this label. And this is a new label in DSM-5, although it describes a sample of folks that existed in DSM-4 in the pediatric section as a feeding disorder. But nobody ever used that label. Nobody was thinking about that label. The pediatricians, the adolescent medicine folks, the gastroenterologists were hoping that we could describe these individuals, but nobody quite put two and two together when we had DSM-4 and DSM-5 rose to the challenge and decided to describe this in a way that gets the attention of those of us who take care of eating problems. There really are three strands of, they're not defined as subtypes, but three ways people get into this category. There's either a real longstanding, maybe lifelong, lack of interest in eating. You know, the individuals who really don't have the appetitive drive that you would expect in a human being. So we'll hear from a family, someone always had to be reminded for a meal. It was never something we had to be encouraged, coached, you know, nobody wanted to eat, right? That's one group. Avoidance of food based on the sensory characteristics of food. So kids, and it's often something that starts during childhood, kids who avoid food saying it smells terrible, it tastes sour, they can't possibly approach it. Maybe there's a sensation inside of their mouth. The food's too thick, the food's too scratchy, and they can't consume it. Not infrequently, this subgroup of individuals have other neurocognitive challenges. Maybe they're on the autism spectrum disorder, autism, yeah, disorder spectrum, and they have that experience which gets in the way of eating normatively. Additionally, there are some who have great concern about aversive consequences of eating. So kids or adults who may have had this since they were kids who avoid eating because they're afraid that if they eat they're going to vomit, afraid if they eat they're going to choke. There's some, afraid if they eat they're going to have a full-blown allergic reaction to something that they may have been told they have an allergy toward. This is an additional category. In addition to point one, there's a persistent failure to meet nutritional and or energy needs associated with one or more of the following. There's significant weight loss or failure to meet expected weight in the growing child. There can be a nutrient deficiency that gets someone into trouble. There can be a dependence on nutritional supplements or tube feeding, or there can be an interference with psychosocial functioning. Any of these meets criteria for this second point. Some folks have more than one of these issues, but essentially it's this whole second bullet that gets you out of just worrying that someone's a picky eater and brings them into a cause for greater concern and the diagnosis of ARFID. This condition isn't due to another medical or psychiatric illness exclusively. It's not due to something that if we help with the treatment of, right, someone's not eating because they're depressed. We treat the depression. They're eating again. That's not ARFID. This is something that needs its own specific clinical attention. It's not better explained by a paucity of food or culturally sanctioned practices, and there's no presence of anorexia nervosa or bulimia nervosa. And for the most part, what that means is there's no body weight or shape concern. So this is a different category. It sometimes gets confusing as a young person is presented for failure to thrive, for low weights, for difficulty gaining weight. Often these folks were considered to have anorexia nervosa. They just weren't telling us. But we do see really two different paths for these kids. Often these are the individuals who are very relieved that there's a label for what this is, that there are strategies that might help them. Okay, what about some new presentations? And I want to say something about atypical anorexia nervosa. So DSM-5 did a beautiful job of trying to decrease the numbers of individuals who would land in the old EDNOS category, that not otherwise specified category that was heterogeneous. We didn't know what to do quite with the individuals who didn't meet full criteria for an eating disorder. There are lots of changes that DSM-5 made that helped reduce the number who wound up in this other category, but not entirely. And DSM-5 has the other specified feeding and eating disorders. It even has an unspecified feeding and eating disorder category. And in the other specified category, there are a number of specific examples. There's a subclinical bulimia nervosa, subclinical binge eating disorder. There's some night eating syndrome. There's purging disorder. There are a number of phenomena that we tend to see. There wasn't enough information quite to decide to bring those into the formal list of feeding and eating disorders, but we wanted to follow who these folks were. We wanted to learn more about them. In the subclinical anorexia nervosa category, a group that is labeled atypical anorexia nervosa, we actually see a group that's presenting for care quite frequently. So this is a group of individuals who share behavioral and psychological features of classic or typical anorexia nervosa. It includes folks who've lost a significant amount of weight, and they're underweight for their own body's needs, but they're not underweight by current medical standards. So what does any of this mean? It usually means that someone's baseline weight is in an overweight or obesity category when they get started. They may begin lots of unhealthy, food-restrictive behaviors and land somewhere. So somebody who may have been 275 pounds who reduces tremendously lands at 190 pounds. Their medical provider looks it up on a chart and says there's nothing underweight about this weight, but these individuals have gotten themselves into a lot of trouble. They fear weight gain. They fear being fat. They fear becoming fat tremendously. And as I'll describe in just a second, there are a number of medical features that really look quite identical to what we see in anorexia nervosa. And more and more frequently, these individuals are presenting to eating disorder specialty services. These individuals are presenting to their medical providers with questions about what to do because they are not functioning well, and they've had changes to their degree of functioning. So there are a number of papers out that describe how many more cases of this category of difficulty centers are seeing. In one hospital, cases increased by 500% over a number of years. Other studies suggest that the prevalence in the community is as high or higher than what's seen for anorexia nervosa. And compared to individuals with the more typical pattern of anorexia nervosa, we actually see more severe eating disorder related psychological symptoms. And that makes sense because the fear of fat or becoming fat may indeed be even greater in this group. There are similar levels of depression and other co-occurring psychiatric symptoms and conditions. There are similar serious medical complications. So we are seeing the bradycardia, the orthostatic hypotension, the cold temperature, the interruption in menstrual activity. We're seeing all of those medical features in this group. There may be a greater proportion of males and non-white individuals in this group compared with the typically presenting folks, but all of this needs further study. And Tim Walsh and colleagues did a beautiful review of all of what has recently been published about atypical anorexia nervosa. And that came out just a couple of years ago. OK, so often in this audience, I get questions about, well, OK, so we've got these formal labels. We've got some other presentations that sometimes come forth. But what is disordered eating? Is that really so different from an eating disorder? I actually gave a talk a couple of years ago to the APA where the entire talk was organized around non-DSM-5 presentations of eating problems, orthorexia, diabulimia, all sorts of things that we hear about and we're trying to figure out where does that live. This was just a schematic that we had included in a book that a group of us wrote a few years ago, Eating Disorder is What Everyone Needs to Know. And it's useful sometimes, and certainly as we're talking to patients or as we're thinking about patients, to realize that normal eating includes food that's chosen mostly because of hunger, mostly because of preference. There's not guilt or shame around eating. There's not difficulty eating in social situations. Disordered eating has lots of things in it, right? Food choices are often influenced by the desire to maintain a certain weight, a certain body type. There's often a feeling of guilt around eating certain foods, ways that people label foods as good or bad, or allowed or not allowed. There's some avoidance of or difficulty with eating in social situations, someone who says, I'm going to go late to that party because they serve all the food at the start, or they serve too much alcohol and appetizers at the start. I'm going to limit what it is that I see. They can't fully engage. And an eating disorder, of course, has got food choices that are influenced by goals of restricting, losing weight, or by the behaviors of binge eating or purging that you just heard about. There's lots of guilt and shame around food most of the time. There's frequent avoidance of eating or attending social situations involving food. And these folks, not uncommonly, really develop a lot of isolation and separation from lives as they may have been for them before any of this started. So maybe this is a little bit helpful that way. Certainly, identifying the diagnosis and figuring out which category someone sits in is very useful. And I just want to let the group know about a tool that is freely available. The website is eda5.org. This site isn't its own phone app. It's a website. It works best with Google Chrome. But it runs folks through a set of questions that help really determine which DSM-5 diagnosis someone lands in. Someone doesn't have to be a psychiatrist to use this app. We have our bachelor level assistants in our group who've got familiarity with the diagnoses. One needs to have some familiarity as you're asking these questions. Conduct this assessment. And that's really how we know who may be eligible for different of our studies. It's a very easy to use tool. OK, so we said that we would next talk some about who gets these disorders. And that falls into the epidemiology category. So let's say a little bit about that. So in the US, a recent and well-done study by Udo and Grillo identified the following. And I'll just note that the publication was in 2018, which means this happened before the pandemic. And there have been changes since the pandemic. But the 12-month prevalence rate, so the percentage of individuals who, in a given year, are expected to have the various diagnoses, are 0.05% for AN, 0.14% for BN, 0.44% for BED. So that predicts, given our US population, the following numbers who are affected at a given time. And that's a slice. Clearly, lifetime prevalence is very different. But here's what we're kind of looking at, trying to take care of now. There's always a gender imbalance when we're talking about eating disorders, with females being affected more frequently than males. But the ratio is somewhat different for the different disorders. So for anorexia nervosa, we've got the highest imbalance. For bulimia nervosa, a little bit less so. For binge eating disorder, we've got the closest to some gender balance. And actually, while I don't present information on ARFID, because less is known epidemiologically about how many people are affected, in terms of gender balance, that one also is much more balanced, something like 3 to 2 in terms of who's presenting. And it's usually to their medical providers initially with this difficulty. So that awareness sort of helps us know how often we're expected to encounter some of this and who may be affected. Clearly, when we're talking about subthreshold eating problems, the numbers are greater. I said there's insufficient data about ARFID. The anorexia nervosa rates have been stable for a very long time. I think there are a number of folks who think that that is growing, growing, growing every time we turn around. It's actually remarkably stable over much of the latter part of the 20th century into what we're seeing now in the last 20-something years. For bulimia nervosa, there's actually been a decrease in prevalence rates. These are expensive illnesses to society. Eating disorders account for, according to one large study, calculated a very particular type of way, nearly $65 billion in healthcare costs in a recent year. There's a whole much larger number of billions of dollars in additional costs due to diminished wellbeing. And that's a calculation based on years of life lost for those who die from these conditions together with years spent in disability. So that daily calculation, disability adjusted life years, and all of this is calculated with a dollar value on each of our heads in our society. And I believe that for this study, it was 200 and something thousand dollars per individual. That was part of the calculation for this very, very large number. But it's notable, we're losing a lot of opportunity, potential in lives as a result of these conditions. So I don't think this group needs to hear, but it's useful to remind all of us that eating disorders don't discriminate. People from all different backgrounds can develop an eating disorder. This wasn't always the assumption, and it wasn't always as clearly worked out and identified, but it certainly is something that we better recognize now. In terms of ethnic and racial minority populations in a community sample in the US, 12-month prevalence rates for bulimia nervosa was higher in Hispanic and black groups compared with non-Hispanic whites. Most of this difference was carried by the higher rates of males in Hispanic and black groups that endorsed bulimic symptoms, and probably that contributed to this difference. In anorexia nervosa, we weren't seeing higher rates in the racial and ethnic minority groups. When we look at any binge eating, lifetime prevalence for any binge eating is significantly higher in Hispanic, black, and Asian groups compared with non-Hispanic whites. What about sexual minority groups, transgender and gender diverse groups? Sexual minority groups defined as individuals, this is a large sample, individuals who do not endorse heterosexual orientation, so there's a range of who lands in that category. The likelihood of an eating disorder is nearly three times higher than in majority groups. In transgender and gender diverse groups, and there was a very nice meta-analysis by Kesky-Raconan, 20 to 50% report disordered eating, 30% endorse eating disorder symptoms, and two to 12% have received eating disorder diagnoses, formal diagnoses. In another sample that's looking at medical records in youth receiving gender-affirming care, and again, a large sample, 5.6% of the 12 to 15-year-olds in that group carried a diagnosis of an eating disorder listed in their medical record. We're concerned about the high rates in these samples. There is some evidence that there may be added severity to having both factors in one's history. Those who are transgender with an eating disorder appear to have higher rates of non-suicidal self-injury, suicidal ideation, and suicide attempts than transgender groups without an eating disorder or individuals with eating disorders who don't identify as transgender. So again, when we're thinking about risk, we're thinking about what we ask, who. Of course, we should have a bit of a universal approach as we're thinking about patients, but it's worth knowing that we may have more to worry about. These aren't illnesses that only affect individuals in North America and Europe. The Global Burden of Disease Study, which is published at intervals in The Lancet, started to include eating disorders or see an uptick in eating disorders in global populations two surveys ago. And this is from the most recent 2019 survey. Essentially, the different subgroups of individuals with eating disorders and the female-to-male ratios that we see in North America and Europe and Australia are also present in other parts of the world. And this appeared in the Global Burden of Disease Study. What happened during the COVID pandemic? You know, we know that a youth mental health crisis ensued. We know that there were many areas that affected young people that really surged during and just after the pandemic. We certainly saw this in eating disorders. There's some reason to think that eating disorders saw a greater increase than what was seen for some other psychiatric conditions. Hospitalization rates went up dramatically in different samples, different parts of the world. You're seeing a chart that describes what happened at Boston Children's, what happened in Australia, what happened in a center in France that essentially upon lockdown. And there was some questions that followed some of these reports that said how much of that was a decrease in available treatment services that may not have been hospital-based. Is that part of what we saw? But there are now, you know, in a pretty significant meta-analysis that tried to look at lots of publications about COVID, there are a large number of studies that are describing increased severity of symptoms, increased numbers quite purely of individuals with these disorders who are in various regions. So we think it's real and we worry about it. And it makes some sense when we think about the screen time, the time that especially youth were trying to fill with new ideas for exercise, new ideas for diet, ways of managing their anxiety and mood, time spent comparing their image to either other images of themselves or other images of peers. And so it does make sense that some of these symptoms developed or worsened. All right, what about risk factors? What do we know? So, you know, biological and environmental factors seem to come together in the development of these conditions. Biological factors aren't only one's gender, but the fact that we're seeing over, you know, all of the years of knowing about these conditions, that females are affected, that it tends to begin right at the time of adolescence, used to be thought late adolescence. We're now seeing this at the start of adolescence, is giving us a clue that biology is playing a role. Also, environments that emphasize the pursuit of thinness, which is becoming more and more and more of our world, is playing a role. And as these come together, we're imagining that there are real epigenetic processes where the environment is helping to turn on certain genes and having the illnesses manifest. Large-scale twin studies support heritability for all eating disorders, especially anorexia and bulimia, and especially anorexia, right? So we really know the most about there being genetic influences for the development of anorexia nervosa. And in most considerations, we see a heritability that sits for anorexia nervosa somewhat lower than what we'll see for bipolar illness and schizophrenia, but higher heritability than what we'll see for unipolar depression and anxiety disorders. The group at UNC, led by Cindy Bulick and Patrick Sullivan, and now a large and international team, has been very interested in doing genome-wide analyses for individuals with anorexia nervosa, really trying to identify whether there are patterns, whether there are areas of gene difference in this population. They conducted a large study that got a lot of attention in 2019. They're now on to an even larger, more parts-of-the-world study to add to their numbers of samples. They identified eight loci that consistently seem to have differences in the anorexia nervosa group. And interestingly, or at least they were making a big point about it being interesting, that some of these loci are areas known and genes known to control elements of metabolism, and really wondering whether that is describing something about who develops anorexia nervosa. Is there something about some individuals, not more than the some, but the some individuals who really can develop and then sustain these frighteningly low weights, and that that's a piece of who it is who is most vulnerable. Social media, as we said before, there have been a number of studies now really trying to examine exposure to social media and how that impacts the emergence of eating disorders. So lots of things that we worry about. So what about illness course if this does develop? I think I've got a couple of things to say about that. One is that it certainly varies, right? So we'll all have experience with or hear about individuals who presented with an eating disorder, received an effective and evidence-based intervention. That was the episode. They got better, and it didn't come back. But then we'll hear lots of other stories. Folks who come in, there's relapse after an attempt at treatment. Maybe they present for treatment a number of times. Maybe those individuals go on and do well, but it's after a number of attempts. And there are some who become more chronically affected. I think it's important to focus on this. It's not the only long-term study. It's a long-term study that I found quite inspiring for work that I do, and it's a long-term study that came out of the Mass General Group that had taken a sample of close to 400 teens who had presented for anorexia nervosa or bulimia nervosa for outpatient treatment of those disorders a lot of years ago now. And this group continued to follow those who were willing to participate for many years, and they've published at a number of intervals. Most recently, they published, on average, at 22 years after presentation. Those varied from 20 to 25 years. Prior to that, they had published a nine-year follow-up study. Here's what they found about the percent recovered after initial eating disorder presentation. So look at the anorexia nervosa numbers, right? At 22 years, they've got close to two-thirds of patients in full recovery category. And what they note, because those were not the numbers at nine years, is that half of those with anorexia nervosa who were not recovered at nine years post-presentation progressed to recovery status at the 22-year mark. That is really important information for all of us who worry that our patients, after a few attempts at treatment, are really in a chronic category. For material and literature that's out there that suggests that large numbers of patients with anorexia nervosa are treatment-resistant, they don't respond to any of the things we've got out there. That's the end of that story. It doesn't seem to be the end of the story. And certainly, some of the publications you may have seen certainly got to the lay press about the question of whether there's a terminal form of anorexia nervosa, a really treatment-resistant, might-as-well-give-up-now form of anorexia nervosa. Some of that literature is suggesting that once someone is after 30 and they've had this illness for more than seven years, we really should be wondering if there's much to do. If it takes 20 years to start to get into that better outcome group, 30-year-olds haven't even had the illness for that period of time. So be aware. It's an important study. And it's not the only longitudinal study out there that is letting us know that sometimes remission and recovery take a very long time. For bulimia nervosa at nine years, the 2 3rds doing all better was already identified and the numbers didn't change at 22 years. All right, let's talk a little bit about treatments and treatments that work. So there are many treatments that have been considered for individuals with eating disorders, treatments that are delivered on outpatient basis, treatments that make their way into the more structured treatment programs. I think it's useful to think about what's shared across a number of these treatments because many of us are a little eclectic in what we deliver. And patients certainly have preference for which style of approach maybe feels like it fits best with them. And if there are some features that really seem to be present across the board, let's think about them. These are behavioral disturbances. And treatments that emphasize behavioral change are most commonly in the groups that are found to be helpful and effective. So we want to think about behavioral change. We want to ask our patients to work with us toward behavioral change. We want to measure our outcomes in terms of behavioral change, whatever flavor of treatment we wind up delivering. And if there is no behavioral change, then likely a treatment is not successful for that individual. These are treatments that are not so different from all the other things that we do, but just to say need an empathic stance, need a therapeutic alliance. For many of these conditions, certainly for anorexia nervosa, baked into the diagnosis, we've got ambivalence about treatment goals. Individuals are afraid to take those next. They may not be happy with the life that they're in at that moment, but they're very reluctant about taking the steps needed to really make significant changes in the ways they eat and what they weigh. They need to work with us. They need to experience us as working with them, moving things forward. And elements of motivational interviewing and things that really can help them identify motivational elements is going to be important. As I said before, several episodes of care are not uncommon, and we want to educate our patients about that if they've gone through a treatment with us and are still struggling. The illness course may be long. Recovery is possible. That's something that I certainly communicate to patients. Remission and recovery rates are best in younger patients. So this really suggests the importance of early identification and early intervention. So I think sometimes there's a sense that a teen with this disorder is the same as an adult with this disorder, and they'll really engage when they're ready. If we can move a younger person into identification and effective treatment, their chance of good outcome is much higher. And so I really do push when it comes to a younger patient. I push everyone, but that's just me. While eating disorders include shared features, the illnesses are distinct and respond to different treatments. So let's hear a little bit about those differences. So for anorexia nervosa, what do we know? So we know the most about approaches that are successful for younger patients, for teens, for youth. And some of you are familiar with these resources. The Treatment Manual for Anorexia Nervosa by Jim Locke and colleagues, I think we're up to the second, at least, edition at this point. It's terrific. It's user-friendly. It actually is usable, even by somebody who's not a specialist, because very specifically case material is presented and how to think about the stance. This is a treatment that really is saying, OK, we've got a condition that's as life-threatening as so many other medical conditions. Parents get involved and help assure treatment adherence for those medical problems. Why aren't parents engaging with and really feeling empowered to do whatever is necessary to move their child through the steps of treatment onto health? And so it shakes it up. It asks everybody to be involved. It gives the parents a lot of power at the start with a clear message that the goal is to move that back to the adolescent, as would be developmentally appropriate. But in any case, it's a good read. And the other book is a good parent handbook that really parents find very useful that explains this approach and works in tandem with the treatment. The bar graph over here is from the publication in the archives that this group, Jim Locke and colleagues, did that really helped put FBT on the map as an effective strategy. I want to say a couple of things about that. So first of all, the dark bars at the bottom are the individuals who had complete recovery at the end of treatment or at whatever the assessment points. I'm sorry, end of treatment, six-month follow-up, 12-month follow-up. Complete recovery as measured by medical features, psychological features, and weight. The grayer bars are partial remission, so individuals who did much better but maybe didn't achieve the full recovery status. And then the white part at the top are the individuals who remained affected. So just without needing to know any specific numbers, we're seeing the vast majority of kids doing much, much better after a relatively short intervention. It's an outpatient intervention. They haven't been taken out of their lives. And six months into that treatment, many of them are doing better. That's a much better whole image than what we see when we look at adult trials. And so that is a take-home. This is a treatment that's helpful. But I want you to look at the comparison treatment. It's listed here as AFT. That's an adolescent-focused treatment. Actually, this group has done a number of comparisons and looked at different kinds of strategies. Maybe the FBT did better. And actually, statistically, it didn't do all that much better at the end of treatment point. It did do statistically better at six months and 12 months. But just looking at the black and gray bars, that comparison treatment did pretty well, too. And the take-home there is kids do well. Getting them into treatment, getting this focused on, kids respond at much higher rates than what we see in adults. Just get them into treatment. But yes, FBT is considered first line. What about for adults? What about other levels of care? The graphs are not for you to worry about right now. And I'll just say that adults often try outpatient treatment first, but not uncommonly need more supports to move their weight and their eating behaviors into where they need to be. Just like kids need their parents and a lot of supervision to move this forward, adults, not uncommonly, need something. And since they aren't living with parents, they're the ones who frequently land in the structured treatment programs, whether it's the partial hospital programs, the residential programs, the hospital-based programs. The setting may seem different. The approaches, if they're effective, are all quite the same. Steps toward healthier eating, the very hard job of getting in more than what feels possible when one is staring at a plate of food all by themselves, and making sure that the calories make the changes necessary to assure that rate of weight gain so that they're not spending five years making the weight change, but they're actually efficiently moving things where they need to be and getting the supports needed. So structured treatment has a place. And while theoretically an outpatient arrangement could get some of that done, sometimes our patients need something more than what outpatient can deliver. Back to these charts. This is just an example. It's actually from a teen group. This study was done at Stanford. Hospitals used to be very, very tentative about getting people started on their calories. This program had started with 1,200 calories a day and moved things very, very slowly toward the kinds of caloric prescriptions that allowed patients to gain weight at a good clip. And Andrea Garber and her team said, why? We've not had any medical complications. We're not seeing refeeding syndrome because we're following these kids so carefully. Let's see if we can. And there are other programs that were ahead of them at starting patients at a higher caloric rate. They actually randomly assigned these individuals to either starting at 1,200 or starting at 1,800 calories and moving up from there as would be needed to achieve the two to four pound per week weight gain that is commonly seen in hospital stays. And essentially, what you get to see is successful increase in BMI, shorter hospital length of stay. These studies demonstrated no worse medical outcome, nothing that we're worrying about. And when they did follow up, no difference to those who were fed a little faster and a little slower. So just to kind of embolden us, it's a hard job to help people refeed. Structured programs do it quite effectively. It's wonderful when they're collecting outcome data so we know what works. And if they're enhancing what it is that they do, all the better. At this point, patients do come in and commonly, in many of our programs, receive something like 1,800 calories to start, but often well over 3,000 calories a day needed to achieve their recommended weight range. What do we know about what predicts good outcome? So two comments on that. This is a study that was done in France, Rigo and colleagues, that really demonstrated that the BMI achieved during hospital stay very much predicted risk of relapse. Those with the lowest BMI at the end of hospitalization, the top curve is 15.5 to 16.5, their risk of relapse was significantly higher. This was a prediction of relapse within two years following discharge. Now, of course, you've got to ask who are those individuals who get to only a BMI of 15.5 at the end of hospitalization. There may be other features that really are getting in the way of their commitment to health, but notable anyway. And there are other studies that are out there getting patients as much of the way toward healthy weight in those structured treatments. Very, very useful in setting them on a course that's going to have a good outcome. Another little feature that we did at our center and is worth noting. So we've got a program that, because it's a research unit and because it's a voluntary unit, actually can get the patients who choose to come to us all the way to healthy weight range before they're discharged. We're supported by the state of New York, and because of that, patients aren't cut off at an insurance point before they've reached that healthy weight. And if they come to us, the vast majority get fully to a BMI of 20 or so before they leave. And we have very brilliant research nutritionists who said, OK, well, at the end of that stay, patients are choosing their menu items more of the time, and they're going out on some passes that allow them to make food choices. Can we take a look at what they're choosing and whether that's going to tell us anything about how they do down the road? And remember, all these individuals got to a BMI of 20, and all these individuals are eating the caloric requirements to maintain that weight range. So there aren't a lot of differences there. But when we examined their food records, four days of food records before they were discharged, two things really stood out. The individuals who had brought in more energy density, which means higher dietary fat, lower amounts of water and foods that include water, so less lettuce, more cheese. Those who had more energy density in their own choices, same total caloric intake for the day, but that's what was present, as well as those who had more dietary variety, a higher number of unique items in their diet in any given day. They weren't eating the identical thing at breakfast and at lunch, or just four things that felt comfortable, or not even just, it wasn't even per day. It was across those four days. So it wasn't the exact same lunch each of the four days. Those two factors predicted who was doing well a year out and who was not. So across the same weight, across the same number of calories, but with these differences in what people did in their individual choices, seemed to really predict who was in a good clinical outcome group and who was not a year after discharge. So this is a replication. I just showed a picture from this study. We did this twice to be sure that it was true. It's true. So there are some features that we should be looking for and encouraging and underscoring as patients undergo treatment. All right, psychotherapy for anorexia nervosa. There's several manualized treatments. CBT, specialized supportive clinical management, which actually was a treatment that was developed as a placebo arm in somebody's study in Australia. And it turned out to be just as effective as the CBT that was being studied. Interpersonal therapy, another combination of behavioral measures that was done in the UK by Ulrike Schmitz group called Mantra. These different treatments have been manualized. They help. But no one therapeutic approach is superior to the others. And that's a little disappointing. But it's also notable as we just think about, OK, what are those common principles? And can we at least deliver those? And we want to build on it. We want to improve. And I'll say something about that a little bit later in the talk. Being in an eating disorder focus therapy is recommended. It's associated with modest weight gain. And the recent publication of the APA Practice Guidelines for Eating Disorders says so. That's what's more important for adults with AN than any one particular approach. What about medications? This is actually the first medication study I ever did. It was a long time ago. And the details and being able to see the graphs well is not what's as important as the gestalt here. But this was a small group of 33 women with anorexia nervosa who came to our eating disorders treatment program. They were randomly assigned to receive fluoxetine or placebo, fluoxetine at 60 milligrams. And you don't need to know by legend which is medication and which is placebo on these curves. It turns out the dark blue is medication and the turquoise is placebo. They're identical. So the rate of weight gain and the way that weight changes in our treatment program, so there are other things going on here, but still no added benefit of fluoxetine. We also looked at Beck Depression Inventory. Mood does improve, is known to improve as nutrition improves. There was absolutely no added benefit associated with fluoxetine. Many studies of medication are out there. They all say something similar to this, that surprisingly, in a group of medications that seem to help everybody else that we take care of in psychiatry, anorexia nervosa remains this challenging condition where we don't see benefit to the core features of the illness. We don't seem to see benefit in the mood, anxiety, and obsessionality that we're seeing as symptoms along the way. And that is a surprise, but that is a consistent finding. This was from an outpatient study. So we got all of our patients. This was a group of 90-something patients across two sites. All the individuals got weight restored before they entered into the study. So everybody is starting at the top of this survival curve at a BMI of 19.5 or 20. And then they're all receiving CBT for outpatient treatment for a year following discharge. And the group is randomized to fluoxetine, 60 milligrams, or placebo. Relapse rates are significant. Patients fall out of that trial either because their weight's too low or their mood's too low or something is going on to make it no longer wise to keep them in an outpatient treatment. But there's no benefit to being on fluoxetine in that experience. This is quite a beast, this condition. And we find ourselves quite challenged about how best to really help people get where they need to go. The one medication trial where there was a positive finding was a study of olanzapine versus placebo. This was a five-site study, 152 subjects participated. We saw some significant, modest, but significant BMI change. Those receiving olanzapine changed BMI at a rate of about a quarter of a BMI point per month. Those receiving placebo did not. We also, of course, were looking at a number of psychological features. And we didn't find much change in psychological experience for the individuals. We had individuals, anecdotally, who were letting us know that their anxiety changed or that their preoccupations changed. But we did not find that in a statistically significant way. We did find some somatic improvements where patients were saying they were sleeping better, they were able to sit more comfortably, they were able to focus better during their days, they had less irritability. A number of features that did reach statistical significance, but our primary outcomes of really hoping that this would decrease anxiety, improve eating symptoms, and decrease obsessionality, those were not found. Now, to those who may say, well, of course it helped them gain weight, it has everybody gaining weight. Remember that many medications that are associated with robust weight gain in healthy controls have been studied in anorexia nervosa. So the tricyclic antidepressants and a number of meds that are thought to increase appetite and do a number of things that cause weight increases in healthy populations have been tried in anorexia nervosa with no weight change in those groups. So this was something, and this was important. These individuals did not develop any metabolic difficulties. Their hemoglobin A1c did not change in a problematic way. Their liver function tests and measures of triglyceride did not increase in ways that we might have worried. And there may be something, again, very different biologically as they're starting at these very low weights in terms of what we might expect on this medication. It's a hard one to convince patients to take, but we were able successfully to get patients into this study. And some of you may have experience with using these medications in some of your patients. OK, what's coming down the pike for anorexia nervosa? And I was at a talk yesterday where Graham Redgrave from Hopkins said, if you have a pulse, you're aware that psilocybin is of interest in whichever diagnosis we're talking about today. So certainly there are some who are curious about whether psilocybin might be helpful for anorexia nervosa. The thought is that that's a certain urgic activity that's associated with this agent, as well as the question of cognitive flexibility and rewiring rigid networks. Is there a way that that could be useful for these patients? Here's what we know so far. Not a lot yet. So there's been one publication. It was in Nature Medicine. It was by Walter Kay and the group at UCSD. It's 10 individuals in a pilot study who participated in this open series. So we don't have any comparison group. This was sponsored by Compass Pathways that makes a synthetic psilocybin. Individuals tolerated the intervention. It was tried at 25 milligrams. They came. They were eager to participate. There was some improvement on EDE that he described in a subgroup. So that's the eating disorders examination. So people who say how much they feel like they need to restrain what they're taking in or worry about their body shape or weight. There were some improvements. And I think it's interesting on follow-up, a three-month follow-up, there certainly are individuals who say that they believe that their quality of life has improved, that their connection to spirituality, that their willingness to try this again, that there were a number of positives that people did experience. Compass is now sponsoring a randomized controlled trial across a number of sites in the US and the UK. And we'll see what happens when they complete their work. But that's what's happening so far. So bulimia nervosa, what can we say about treatment? The story for bulimia nervosa is a little bit different. Oh, I'm going slow here. A little bit different in that we've got therapies and we've got medications that do work. And it's a more straightforward story. Cognitive behavioral therapy is effective for bulimia nervosa. This is a picture of the Chris Fairburn book, the most recent edition. That's a well-read, well-received book that describes CBT as an approach for lots of eating problems that include binge eating, including bulimia nervosa. It's a focused treatment. It's 10 to 20 sessions over six months. It's not supposed to go on and on beyond that. It's a treatment that has, over many years, been associated with a 50% or so remission rate. There's a question of whether the remission rates may be higher for adolescents. There's also an interest that Daniel Lagrange and the Jim Locke group have entered into where they have begun to demonstrate that FBT is a helpful way, either to deliver some of the CBT elements, but to use the family in the treatment of bulimia nervosa when it affects a young patient, possibly. It can be delivered as self-help. So that's a book that really has a self-help tradition, whether that's received by the individual or whether it's guided by a therapist. There's recently support for a transdiagnostic version of CBT. It's called CBTE. E stands for Enhanced. I always think it's for everybody, CBT for everybody, but that's not right. Anyway, and that's the latest version coming out of Chris Fairburn's group at Oxford. There's new interest in a shorter course of CBT. It's called CBTT. It's a 10-session CBT. Patients seem to respond. And needless to say, Glenn Waller had developed that, and the National Health Service is delighted that there may be a shorter treatment, a cheaper treatment, that is as effective. Interpersonal therapy is also useful for bulimia nervosa. Medications. OK. Again, another bar graph where you don't have to read all of the details to get the take-home here. This is a bar graph I've shown for many years. The dark blue bars are the active medications that have been tried, and there have been even more that have been tried. The turquoise bars are placebo. You're seeing with the length of the bar just how much reduction in binge frequency a medication is associated with, and you see that the dark blue bars are longer than the turquoise bars. Medication is more effective than placebo across all sorts of classes of medication. And almost always, this is a statistically significant difference. All right? For fluoxetine, which is our only antidepressant that has gotten to the FDA's threshold for a separate indication for the treatment of bulimia nervosa, it actually wasn't 20 milligrams that was helpful for bulimia nervosa. It was 60 milligrams. 20 milligrams was no different than placebo and its effect on binge frequency, but 60 milligrams was different from both of those categories, and 60 milligrams is the indicated dose. That's your green bar here. A couple of take-homes about the use of fluoxetine in bulimia nervosa, and this is actually a slide created from data that was part of all of the studies that were presented to the FDA in the Eli Lilly effort to bring this medication to market. And here are your take-homes. You start to see the active medication effect on binge frequency separate from placebo, like on the first day. And you see the maximal effect very quickly. This is not depression rate. This is in two weeks, two to three weeks, you're seeing the improvement that you're going to see. That helps you prepare for what you're looking at as a patient has or doesn't have changes in response to the medication. Helps the patient also hang on and prepare for what this medication may do. Of note, when Lilly did these studies, they started everyone on 60 milligrams on the first day. I'm not saying that everybody ought to do that, but it's notable that it was well-tolerated used that way. In all of the studies we've done using 60 milligrams of fluoxetine, both in bulimia nervosa and in anorexia nervosa, we did the increase over five days. Two days at 20, two days at 40, and then we were already up to 60. Don't wait months to get someone to 60 milligrams. Get them quickly, and you'll see the results, hopefully, quickly. All right, what about binge eating disorder? All right, so this is just letting you know of a very nice review in the annals that describes all sorts of effective treatments for binge eating disorder, but here are a few take homes. We're aiming for abstinence, not just a little bit of improvement in these illnesses that include binge eating, whether it be bulimia nervosa or binge eating disorder. We can get there. We can get there much more of the time. CBT versus wait lists across many studies has got close to a 60% abstinence rate. So we're expecting good things. Listexamphetamine, which is the one medication where we do have FDA approval for the treatment of binge eating disorder, we've got something like a 40% abstinence rate. Serotonergic meds, and there are really a whole host of antidepressant medications that are in this category versus placebo, again, close to 40% abstinence rate. There are significant decreases in binge frequency associated with therapist-led CBT. Using that Fairburn book, if you like. It's easy to get. Guided self-help CBT, where the therapist is really just helping the patient, maybe not meeting all that frequently with the patient, but helping the patient make their way through that self-help book and kind of helping to explain, psychoeducate, and work on the exercises that are included. Interpersonal therapy, especially for individuals who really will describe that the binge episodes occur in the context of interpersonal difficulty. Antidepressant medications are helpful. List examphetamine, as I mentioned. Other medications that impact appetite and weight management, like topiramate, are also helpful at decreasing binge frequency. More points about medication treatment. Antidepressants are superior to placebo. Weight doesn't change significantly in response to antidepressant medication. So this is really an expectation that needs to be managed with the patient. Patients are sometimes thinking that they're coming to you and that weight is going to melt away. Your focus with someone who really has significant binge eating disorder is to help them normalize their eating, help diminish or help with the remission of binge eating entirely. But actually, the antidepressants don't change weight. They change behavior, which is its own curious scientific fact. Binge eating and weight do respond to medications that are known to impact appetite and weight. Topiramate is one of them. List examphetamine and stimulants are another. List examphetamine got itself FDA approved for bed. That doesn't mean it is the best or the only one. It just means that that's what Shire was able to assemble after its patent was going to run out for Vyvanse for ADD. They moved on to trying to see if there was benefit here. And there was. And it's real. And patients do find it useful when the medication is managed appropriately. These are just the data from that List examphetamine trial. And in this trial, placebo was compared with 30, 50, and 70 milligrams of List examphetamine. The 50 and 70 milligram doses were found to be superior to placebo. Generally, in the use, we march along different options. Not commonly 30, but maybe 40 will be just fine for an individual. Maybe there are some individuals for whom 30 works. It makes sense to try the lowest dose that works. These are medications where we know much less about safety in adults than we know in teens. And these are not uncommonly middle-aged individuals who are presenting for assistance with their binge eating disorder. OK, new treatments coming down the pike. I certainly would imagine you're curious about whether GLP-1 agonists are helpful for this population. So we don't know enough yet. Studies that specifically examine binge eating disorder are few. Binge eating behavior appears to decrease in studies where this is measured. And this is really just a glimpse from a study where obesity, the bar on the right, obesity-focused meds are compared with semaglutide. And no question that semaglutide has a more impactful effect on binge eating behavior than other obesity management agents. When they've combined the two, again, you see the option that includes semaglutide being more helpful at reducing binge eating behavior. If we can, the comparison wasn't made between the semaglutide only and the combination, but it looks as though semaglutide carries that whole comparison. It's not as though you're getting additional benefit for being on the combination. So what we don't know is whether there are risks for individuals with eating disorders taking these medications. We're certainly hearing anecdotally about people who had a history of a restrictive eating disorder, had a history of lots of distress around body shape and weight, beginning one of these medications either for their medical indications or because they want to give it a try and find that they are in a lot of old thoughts and behaviors that they thought they had left behind or that their providers thought that these patients may have left far behind. And there's worry about whether there's a vulnerable group who can get into trouble with these agents. With that said, they're everywhere, and certainly our patients with binge eating disorder are among those who are trying them for their weight, for their diabetes risk, and maybe for their binge episodes. All right, what about treatment for the newer eating disorders? So we know less about ARFID, but there's a lot of interest in taking a look at some of the strategies, the behavioral strategies that work for other conditions. And the group at Mass General, Cameron Eddy and Jenny Thomas, have written this, again, very user-friendly, informative book that is enjoyed by providers and families alike. A lot of psychoeducation, a lot of exercises for things that can be tried with information about how to do this in an FBT-like model for our youngest patients and in an individual model for our adult patients. A lot of the strategies are really exposure therapy-based. They've got some studies underway, NIH-supported studies underway, and I'm eager to see those results. But the resource is a good one. OK, I said we would talk a little bit about the neurobiology of eating disorders. Really, I'm going to talk about the neurobiology of anorexia nervosa. And for the last 10 or more years, we have been focused on a hypothesis that suggests that these behaviors are hard to change because individuals who have anorexia nervosa have actually adopted automated habit-like mechanisms, including brain mechanisms, that keep these behaviors active and make them hard to interrupt. That was a piece that Tim Walsh had published in the Green Journal. Just quickly on what a habit is, social science says that it's a frequently repeated behavior or set of behaviors that are constantly, consistently triggered by the same cues. I can say that when I get up, I read the newspaper. I do this every morning. I don't really think about it. It would take an effort to skip the step and leave the house without reading the paper. I'd feel weird if I didn't take at least a quick look at the paper. Cognitive neuroscience adds it's a learned behavior, something that we had to think about when we first developed the behavior, that persists relatively independent of the outcome. And this appears to be mediated by the dorsal striatum. I can think in my newspaper example that sometimes I don't even know why I'm looking at that newspaper. The last thing I want to do is read more about what's happening on the Columbia campus. But there I am. I pick it up. I take a quick look. I put it down. My friends, my fellow mammals, the rat, get placed in a cage and learn to press a lever to get an M&M. Very rewarding. They want to press that lever. They want to get an M&M. They get all the M&Ms that they can eat. They're too full to push the lever for another M&M. But you take that rat back into the cage, and they'll still push the lever because they've become reinforced about the experience of pushing the lever just as much as any impact that that may have. That behavior is housed in the dorsal striatum. We move from frontal circuits when we're thinking about and evaluating the options for and learning a new behavior. And we move that into posterior circuits when it becomes automated to make room for what's next. So how do you study decision making about food and anorexia nervosa? When we first started to want to do this, we thought, what are we going to do? We're not eating M&Ms. In fact, these patients aren't eating anything. How are we going to study the absence of a behavior? And then realized that actually patients are engaging in a behavior. They're engaging in a behavior multiple times a day. And it's a selection of a lower fat option rather than a higher fat option every chance they have to make a choice. We do this all the time. Patients certainly do this all the time. And they've got that inside their head or in front of them. What should I do now? The salad or the burger? So we developed a food choice task. And this was adapted from Hare and colleagues who had a food choice task that was for another purpose. We identified 76 different food items that range in calorie content and foods that were higher than 30% of nutritional fat were considered high fat. And the others were low fat. Here are some of the pictures that are in our task. So how does the task work? Patients are sitting in front of a computer screen. And they see this prompt. You're going to see a series of pictures of food for each picture. Please rate how healthy you think it is. And they'll see a food. They'll see a scale. And they push a button. Then they get the next screen. You'll see a series of pictures of food for each picture. Rate how tasty you think it is. And they'll see a picture, see a scale, and push a button. Then they're asked to choose. On each trial, you're going to see a reference food on the left. It's always going to be the same. On the right, the option will change. And the computer has actually calculated an individual reference food, something that is neutral in health and taste according to someone's responses. That'll appear as the neutral food for them. And they're asked to make comparisons compared to that neutral. So for this patient, the neutral food is cherries. And they're asked whether they prefer cherries or M&Ms, cherries or oranges, et cetera. They're told you better be careful when you're making your choice because you're going to be asked to eat one of the things that you say you prefer after this study. So they shouldn't just try to impress me and say they feel like a steak. They should really pay attention. We had hypotheses that anorexia nervosa patients will choose high-fat food less often than will healthy controls, and that we'll be able to see evidence of a difference between patients and controls in the brain. So here's what happened. So first, we took a look at health. And we've done these studies now multiple times. We've got colleagues in other parts of the world using this task. And they've been doing these studies. And we've seen these findings over and over again. Everybody rates low-fat foods healthier than high-fat foods, healthy controls and patients. Difference in what you pick, but not by group. But when we get to taste, we see a difference. So those donuts, the normal response, the healthy control response would be to say, they're pretty tasty. Patients take that out of their taste list and say, no, I don't like donuts at all. And they really do have a difference by group on how they rate taste. And they certainly have a difference by group on how they choose with the patients choosing high-fat foods much less of the time than do controls. You'll notice that everybody chooses high-fat food a little bit less often than some others. And there's some calculation going in there of what do I really want right this minute to eat. But the patient and control group are different. The day after this study, we do our trusty multi-item meal. We have a long history of using a buffet meal to help measure what patients eat and what healthy controls eat in different patient groups at different points in treatment and what they eat. And we have some version of this kind of buffet that has been a longstanding part of our lab. Patients are given a simple instruction. This is your lunch for today. Eat as much or as little as you'd like. And we see what happens. And we've got scales underneath that red tablecloth. And we've got all sorts of ways that we measure every last morsel. What was notable is that that food choice task and what patients did in terms of their choices on the task very much correlated with what patients actually did in the lab when they were given a full meal to select. And at this point, that food choice task has become a very good proxy for us in knowing what patients actually do. Out in the lab and out in the world, this computer task seems to give us that information. What about the brain? So this was the first of what now is a few studies. It was very exciting. It was published in Nature Neuroscience. Our leads on this were Joanna Steinglass, the psychiatrist at Columbia, Karin Ford, who's a cognitive neuroscientist who used to be at Columbia and now is in Amsterdam. And look what happened here. Ventromedial region didn't look so different between patients and controls. But that dorsal striatum was engaged in patients and not in controls. And these are patients who are in a scanner and completing the food choice task at the same time. So we get to see what in the brain is lighting up as they're doing this work. And we've got the beginnings of information supporting that habit hypothesis. Very exciting for us. So just quickly, I thought we were going to have much more time. But what's interesting about this, yes, we're interested in mechanism. But we're really interested in mechanism if it can help inform how we think about treatments and treatments that might be more successful than some of what I presented to you earlier. And if we've got a part of the brain that's impacted, if we've got a set of behaviors that we really think explain a lot of what happens that make it so hard for patients to change, where does that let us go in terms of developing new treatments? It lets us go two ways. One is, can we do something that actually interrupts some of the neural circuits that we think are contributing? And is there something we can do behaviorally to interrupt some of what it is that patients are doing? And we're doing work in both of these areas. In terms of neural circuits, there's neuromodulation now. There's RTMS. And we've got our wonderful psychologist, Alexandra Muratore, pretending to offer RTMS to another wonderful psychologist in our group, Caitlin Lloyd. But she began really a pilot study taking a look at whether even a single session of high-frequency RTMS, whether active, is different than sham in terms of what people do on this food choice. Is there anything that moves in the needle? And in a small group of individuals who are actually among our inpatients, but they have normalized weight enough to participate in this study, we were very surprised. But we found a signal. We found a significantly different likelihood to choose high-fat items for these individuals when they received the real RTMS. The food choice was offered right after the treatment. It was almost simultaneous compared with the sham. And this has got her now working on an NIH-supported randomized control trial to really look at a course of RTMS compared with sham and see whether we can extend these findings. But interesting. In terms of behavioral approaches, where do you intervene? How do you develop a treatment that potentially stops habit? One is to decrease the intensity of a habit to figure out how we can decrease its pull. And the other is to think about how we can identify behavioral cues so that people can interrupt what it is they're doing. We've done one study, and now we're in the middle of a second. This was a real small pilot study. We did it with colleagues from the Midwest, Steve Wunderlich from Fargo and Carol Peterson from the University of Minnesota. And again, you see Joanna Steinglass and another colleague, Debbie Glassifer, who's at Columbia. And we developed a treatment that we called regulating emotion and changing habit. It was really trying to help educate patients about habit, see if that resonated with them, help them start to look at very, very small elements of their behavior that could be pulled apart and created in a bit of a chain. And certain patterns were very clear to the patients. There were behaviors they did to delay the start of their meal. There were behaviors they did to make a meal feel just right. They had to put their water here. They had to move their food there. And we wound up with all sorts of strategies that we worked out together with the patient to mix it up and try to interrupt what was going on. We had supportive psychotherapy as the comparison. And post-intervention measure that's called the habit rating index, the habit strength index, was different post-intervention in the REACH group and not in the supportive psychotherapy group. And more excitingly, when we took these patients to the lab to do that same multi-item meal the day after they completed their four sessions of this treatment, we actually saw a change in the REACH-receiving individuals compared with the supportive psychotherapy-receiving individuals. More was consumed in the lab for lunch after four weeks of this initiative, new initiative. This was enough for us to say, we've got to build on this. And we decided to build on this at a point where we really thought that risk of habit holding you back was the most severe. Patients relapse at high rates. I showed this to you before. And actually, as we analyzed, where do they have the highest risk of that relapse? It's in the first four to six weeks of their relapse prevention period. The likelihood of relapsing within 60 days is highest in that first month. So there is something that maybe we can intervene and do at the very start of their time without hospital that really has the potential for longer-term effect. So we've got a new REACH study. The word is the same, but the initials are different. We now call it Relapse Prevention and Changing Habits in Anorexia Nervosa, REACH+. And this is a six-month treatment. It's telehealth visits. There's extra support from a mobile platform. We have a creative way that we're designing this trial so that we actually have many different components of care that we're evaluating, components that are in a lot of other treatments, components that are taken from our habit-interrupting pilot study as well. And just giving you a sense of the way we structure this treatment, we've got behavioral aspects, cognitive aspects, motivation-enhancing aspects, questions about food monitoring, ways that we build and consolidate skills, and lots of different ways. Each one of these variables, we have a couple of ways that we imagine potentially being helpful. And we've got little randomizations. This unusual treatment trial design allows us to compare, for example, if we've got sessions that are dedicated to behavioral change, does it matter if you're eating in the session with the patient? Does it matter if you've asked them to eat between sessions and they come in and they talk about the experience? Does it matter which way you do this? And we have ways that we've designed this trial that half the patients do it one way and half the patients do it another way. And it's going to give us the power to determine whether it matters how we do behavior, whether it matters how we do the focus on thought that has to change. Our outcome measure, what we're targeting here, of course, is ability to maintain weight, ability to reduce relapse. Another picture that shows you the treatment. We have a website that patients are given access to. They log in. They see a whole bunch of tiles. And behind each of these tiles, there are lots of resources. So between sessions, they can get additional education. They can listen to podcasts. They can see videos. They can read recovery narratives. They can get worksheets and exercises that they can use toward moving this forward. And we'll see. We'll see how it goes on to work. So what did we go over today with next to no time left for questions. But we described eating disorders as disturbances in eating behavior that contribute to medical and or psychosocial impairment. There are six of them. We know the most about three of them. We know a little bit about ARFID. So that landed on our list as well. The presentations of eating problems seem to be changing somewhat. There are higher numbers of case presentations post-pandemic. There are higher weight presentations of conditions that resemble anorexia nervosa. We know that eating disorders come in all shapes and sizes, affect people from all sorts of diverse backgrounds. So ask everybody if they've got a problem or a history with an eating disorder. Treatments include behavioral therapies that focus on reversing eating behavioral disturbances and normalizing weight in low weight disorders. We know that medications are helpful for some of the disorders, not for anorexia nervosa. And a little bit we know that advances in neuroscience that help elucidate illness mechanisms can allow us to really identify some mechanism-based treatment targets and develop some novel treatments that we hope will work. So I am happy to take questions. I know many of you are running to what's next. So please do what you need to do. But really, it's been a pleasure to be here. And I will stand up here if there are folks who've got some questions for me at the end of the session. Oh, it looks like we're going to ask real questions. That's just fine. Go ahead. Is this on? Okay. Yeah. Hi, my name is Rachel. I'm finishing my first year of medical school. So in your answer, feel free to assume I know next to nothing. My question is, what advice, if any, would you have for someone who or someone who is seeing someone who needs to lose weight for legitimate medical reasons, but has a history of disordered eating or an eating disorder? So your question is so timely and so important, and I think only a medical student could bring that degree of timeliness to what the field is struggling with right now. Many of us, right, if we're psychiatrists, we're physicians. There are many mental health providers who may not be physicians, but certainly for those of us taking care of the whole person and thinking biologically about the whole person, we're aware that there are medical risks associated with larger weight status. There's some sense in the lay public that maybe we should be worrying less about those folks, that there are some individuals who appear healthy in larger bodies, and that any kind of dieting or any kind of weight reduction is potentially problematic or could be contributing to the development of an eating disorder. I've been around long enough to know that a problem may not be present today and may go on to develop, and I do believe that people need very detailed discussions about the medical risks associated with a whole range of features, a whole range of vital signs, weight included. I think if someone has a history of an eating disorder or not, we ought to be following and monitoring carefully for the medical, hopefully benefits, as well as the mental health, maybe benefits, maybe risks, that develop with changes to the way someone eats. I think that fad diets, drastic diets, diet, diet, diets, the word is now becoming a bit of poison, probably aren't very helpful. On the other hand, optimizing nutritional intake, figuring out how to measure medical benefits. Certainly now with GLP-1 agonists and some individuals who may be starting at a place where that's something that's recommended, it shouldn't be a one and done. It shouldn't stop there. OK, we've got you on something that we should really be following folks and be able to monitor if any other problems develop inadvertently. Sure. Eating disorders are sometimes lumped in with addictive disorders, especially in the lay press. Can you comment on food as an addiction, please? So we actually had a grant a number of years ago where we were trying to work together with the substance use colleagues in our department and see if we could look at biological features that really seem to be shared. If we take a step back and think about the networks in our brain, the networks that do contribute to substance addiction probably were developed in those brains to keep us eating adequate amounts of food, to keep us reinforced by food. Nobody could have predicted that we'd be in a land of this kind of excess of food. But here we are, and some of these foods have been very specifically designed to make it very hard to say no and make us go back for more. So it's not surprising at all that we see certain similarities in those behaviors. The big difference when we try to treat folks with eating problems is that abstinence is not a possibility in terms of eating. So we somehow have to come up with strategies and approaches that help somebody eat, but eat more successfully. It's interesting, as GLPL1 meds are becoming certainly used with more prevalence, that some people will identify that there's a change to another addictive behavior. So yes, they're eating changes, but so does their smoking, or so does their appetite for alcohol. So I think it's just supporting that we do have some interesting overlaps. It's not exactly telling us what we need to do to help these folks who still have to eat their nutritious meals and snacks a day. Thanks for the question. So I have a multi-part question. I was wondering if you could advise on how, at what point do you decide that the patient needs to be hospitalized? What criteria do you, like how do you go about that as far as saying, OK, this person is a danger to themself or others? And in your experience, the best way to maintain that therapeutic alliance without completely destroying it? OK, so hard question, but there are some answers that I think are important. In the APA practice guidelines that was published earlier this year, there's a beautiful table. And now I'm going to just focus on the medical features that may suggest that hospitalization is the right level of care. There is a table that specifically for adults and for teens identifies which medical features, which tests, which vital signs should contribute to your considering hospitalization. So on the medical front, where to draw the line, there is some information. There's other things that have been published out there, but that's an easy one. No question when a patient meets for any other reason that you would think about hospitalization, like active suicidality. And I didn't say in the anorexia nervosa discussion that that is a significant risk. Mortality rates are high in the eating disorders, and most of all in anorexia nervosa, but a significant percentage of that high rate is due to suicide risk. So if we've got someone who we believe to be not just chronically, look, you're doing something that's dangerous to yourself, but acutely suicidal, you would make the same decision around hospitalization that you would for anybody, even someone without an eating disorder. They may land on a general unit. They may not land on a specialty unit. It's nice if a specialty unit is available. The alliance is always difficult when we're pushing someone to get a treatment that they are reluctant to get or they think is overkill, so to speak. It's always difficult. We're always doing that dance, and we stick to the message of we're trying to help you help yourself. We're trying to do what we believe to be safe. We really have no choice but to put your safety above all else, and we try. And we try to repair whatever ding to that occurs if somebody winds up saying, but I was traumatic to be there. It is notable, and actually in this talk yesterday by the Hopkins group, they had data around patients who come in extremely angry, feeling coerced about treatment at the start, and then when they're asked at different intervals, they're less and less coerced. They're more convinced that the hospitalization may have been necessary. So we can ride that, because there are going to be a number of patients who can look back on it not as angry as you fear. Thank you. Sure. Tony. Hi. Thank you for the very nice talk. So over the years, my specialty has morphed into trauma. And I never list myself as an eating disorders expert, but I've treated a lot of eating disorders because of the huge connection between trauma and eating. I don't know in terms of literature how many studies have focused on the trauma aspect. I know in my clinical work, the more I keep the focus on healing the underlying trauma, not all the time, but oftentimes the eating disorder symptoms actually start to dramatically improve. You had mentioned about habits. Some of the more integrative treatments for trauma, things like EMDR, hypnosis, acupressure. My original specialty was addictions. I've noticed that all of those treatments actually make the work a lot easier. And I think it's because the cognitive piece oftentimes gets pushed to the back, as you guys are actually finding neurobiologically. I'm just wondering in terms of, and I think the stats for the connection between people with eating disorder histories with the trauma history is anywhere in the literature from 50% to 80%. Just wondering how much research is actually focusing on adding a trauma-focused treatment of whatever and seeing, are you getting better outcomes? So I don't know that there are formal studies that I can cite about whether trauma-focused treatment specifically is being measured against not offering trauma-focused treatment. The word about trauma appears to be history of trauma is higher across all sorts of psychiatric presentations. And that certainly includes the eating disorders. The most recent, actually I was writing a review and the editors had all sorts of questions about what exactly is meant by the trauma that's associated. And is it this, and is it this, and is it this? And the most recent publications put together childhood maltreatment, which is really a whole range of traumatic experiences. So we no longer have to distinguish how much of it is sexual abuse or other kinds of trauma, how much is bullying, or other kinds of childhood experiences. But childhood adverse experiences, including these experiences where there really is a mistrust and mis-action on the part of an adult who should be responsible that impacts the life of a young person, happens at some significant frequency in individuals who endorse eating disorder symptoms, eating disorder diagnoses. Whether that's the same, different, or more or less compared to other psychiatric populations, I don't know. What we haven't done, I mean, we'll often have patients who are requesting trauma-informed care. We try our best to cobble something together. But I don't know of studies that have actually looked at whether approaching it one way versus another has a different outcome. Thank you. Sure. Hello. I work mainly with people with eating disorders in outpatient setting. And sometimes, very often, even olanzapine seems to be perfect for patients with anorexia. But at least in my patients, it's very hard to persuade them to take it. And I was wondering if you have any tips on how to do it, how to talk to them about it. Because, of course, they read about this medication and know that it's bad. Two things that I find helpful. Now that we know that there was a different version of this when I was trying to recruit people into a trial and I didn't know whether the medicine would or wouldn't be associated with weight change, I had some pilot work that I could reference. But here, I can really let them know it is a modest change. It is not a dramatic. There's no question that the accounts that they'll read on the internet that somebody gains 60 pounds on olanzapine does not happen to these patients. And I'll show them the article if they would like to take a look. The other thing that I found very helpful and I did use in recruiting patients is there's a small amount of literature that suggests this isn't a metabolic change. It's not as though they take the pill, their metabolism changes, and the same way they used to eat, all of a sudden now causes weight gain. We do have evidence that that's not the case. Patients find that greatly relieving. They have a tremendous amount of confidence that they can control what they choose to take in or not. And the idea that the hope is that the medicine will help make it a little easier for them to make healthier choices and take in different foods and have a different sense of appetite, that has felt greatly relieving to patients because they're like, OK, I get it now. Either that'll happen, we'll see what that feels like, but probably not me. There isn't a sense that just being in the room with the pills and boom, they're going to change in their weight. So I do share that information to whatever extent. I don't want a patient to say, I want to take it and I want it not to work. On the other hand, I want them to give something a try. Thank you very much. Thank you for this presentation. I had a question about the statistic between the nine year to 22 year drop in patients with anorexia nervosa who, or I guess, increase in recovery rates. Exactly. So I was curious, in these patients that do recover over that 13 year period, is there something they have in common with regards to more outpatient follow up, more inpatient hospitalization? I don't know the answer. I don't know the answer to that. We did a focus group, before we did this REACH study, we did a focus group with some recovered patients to try to ask them. And they had all been on our inpatient unit, but some of whom many years before. But we wanted to know their perspective if we were to set up a relapse prevention treatment, what components they thought helped them. And again, we published the findings there. But I'm not sure what to make. It was a small group and all the rest. So a number of them talked about needing to get beyond all of the rules and particulars of their treatment in order to actually do best. Several of them said they think that the treatment pieces laid a groundwork that was very important for them. But for example, they all said we had to stop weighing ourselves so much. We understand why, when we were really underweighed and we were first getting started, that was a very important thing to monitor. But ultimately, I had to say no more of that in order to, or whatever. So actually, no, it wasn't like some treatment strategy, some staying in treatment for forever. That wasn't what they described. They described having to get to some point where they went from the supports being needed to taking those training wheels off. That gave them some sense of agency. We built a lot of choice. We weigh our patients. It's a trial. It's relapse prevention right after they've been in the hospital. That's an outcome measure. We do leave that in. But we did want there to be a lot of choice of what patients have access to, what they may engage with. We're tracking what they participate in. It's a much shorter trial. It's not 22 years. But I don't know really what. But there's some information about individual reports where there are folks who kind of settled down. It's interesting that it's recovery. It's not just improvement that they're describing. They're really describing getting into that recovery category. And remember, there are some individuals who aren't participating in the study and other kinds of things. But I don't know more than that. Thank you. Thank you very much for the talk. Two questions. First, any attempts to evaluate the microbiome of these patients? So that's number one. Number two, this is somewhat antiquated. But I remember when I was at White years ago, they had a subsection. And I just looked it up on the internet. And apparently, they still do for eating disorders. And I found it to be very unsuccessful. And then a lot of times, they would take patients who failed and send them over to Renfrew, which I found to be a cult. So I'm just curious of you. And then those patients would disappear into the ether, the ones who went to Renfrew. All right, I'll try to answer all of your questions. So the first one's microbiome. There are groups that are very interested in microbiome differences across a range of conditions. I don't know what to do with the finding that individuals with eating disorders have different microbiome than individuals without eating disorders. Because individuals with eating disorders are eating different things. And there are lots of contributions to what make the microbiome different. No one's looking at these folks before they develop the eating disorder. And yet, they're trying to expand and say, that may be a piece of the vulnerability, I don't know, or the secondary effect. But there are, including the same group that does the genetics work at UNC, they're interested in microbiome across a range of eating problems. And I'm sure that they will have more to say about that in the years ahead. But there's an interest. The second thing is, when you talk about white, you're talking about the analytic institute. So amidst these studies that have modest impact on eating disorder patients doing a little bit better, psychodynamic psychotherapy has been looked at too. And it also is OK. None of these treatments help people go much from their starting BMI's of 15 to a finished BMI of 16 and a half. There's a modest improvement, but it's not a loss. They're staying in treatment. They're working on something. They don't seem to be sliding backward. But it's certainly not superior to some of these other effects. And I don't know how much behavioral change there really is. Renfrew. So Renfrew was the first of what now are many, many, many privately held residential treatment programs. Renfrew's not a cult. But these programs all are privately managed. They're not affiliated with academic medical centers. They are single focus. They're specialized care for one set of conditions, kind of like some of the substance use treatment programs that are out in the wilderness doing whatever. Renfrew's in Philadelphia, not quite wilderness, but then again. And it really varies. So some patients go, and they go again, and they go again, and they make no progress. And other people say that was very helpful because it finally made me. Some people say, I never wanted to go back there, and that's why I was able to take next steps. I don't know what to say. But in our current health care landscape, hospitalizations have shortened tremendously. Health insurance are pushing patients out of that highest level of care and telling patients that their only option for 24-7 care that's available next is the less expensive non-hospital, but still 24-7 options. Renfrew is one, but there are many other now bigger chains of these residential programs. My concern about those residential programs are the industry-sponsored support. It's not industry like pharma industry. It's industry like venture capital industry. It's a for-profit model. Their incentives are a little funky. Their pressure to present outcomes is zero. So we have concerns about there being great variability in what gets delivered. But again, some people will say it helped them, and some people will say it didn't. Thank you. I just wanted to know, in your clinical practice, what is the prevalence or risk of seizures that you see with eating disorders? So the risk of seizures is primarily due to hyponatremia. So when patients have a very high water intake and aren't being carefully medically monitored and aren't properly educated about the risk of so much water, they may, over time, reduce their sodium to the points where the levels are very, very low and they can have a seizure. This is what sometimes brings someone in to treatment. They've had a seizure somewhere, and then the whole thing gets worked up. They may not have otherwise really known what. It's hard to break people of this behavior, again, because water is filling and non-caloric. Water is part of a set of habits, quite possibly. And to let someone know that that's really potentially going to kill them is tough. Again, some people will sort of wake up and say, I better do something about this. So the numbers, it varies. But it's usually from that behavior. We can see it in anorexia nervosa or bulimia nervosa, because our patients with bulimia nervosa have these purging behaviors. They become quite dehydrated from vomiting or laxative use. And then they will refill. They'll be thirsty. They don't want anything with a calorie. They'll drink a lot of water. Our society has people carrying around jugs of this. And so they feel quite normative as they're taking in very large amounts of water. But it can be dangerous. Thank you so much for your lecture. And it's a little related with that. I was wondering if you can use thoughts or data about naltrexone at 150 XL and, sorry, Welbutrin at 150 XL and naltrexone 25, like half of the tab versus Contrave. Welbutrin 150 or what? And naltrexone 25 versus Contrave, right? So like cheaper cost and whatnot. OK, so let's talk about the Welbutrin part first. So this is complicated because there's a big black box warning for all eating disorders with Welbutrin, right? So that has to do with the increased risk of seizures associated with this medication. And there were a couple of studies. Bulimia nervosa is really where this was focused on. There was a thought that the electrolyte imbalance that's sometimes seen in some of these patients was a piece of the reason why. But then this was seen also in an anorexia nervosa sample. And the thought was, oh, goodness, this may be across multiple eating disorders. And then that black box warning occurred. On that chart I had shown about bulimia nervosa versus placebo trials, there's one that had Welbutrin. And patients stopped binge eating pretty successfully on Welbutrin. And there are some patients who will come into the treatment settings, and they've been placed on Welbutrin. What do I think? I mean, it's not recommended that anyone be on Welbutrin, but the reality is some people are on it. I think that in the initial studies there may have been elements of the first preparations, not the Excel, but the first shorter acting Welbutrin preparations where the rate of getting to blood level and the speed at which that was metabolized and removed added to the seizure threshold risk. I think that there have been far fewer seizures in general seen in association with Welbutrin in these longer acting preparation, first the medium long acting and now the Excel. So we're probably dealing with a safer version of Welbutrin. I do let patients know that if there's any purging, and actually the very low weight patients too, they're not getting Welbutrin from me. But if someone has binge eating disorder and there is a reason that Welbutrin makes some sense or someone has really stopped in some of their eating behavior, there are some people who hold on to, stay on, start Welbutrin. Contrary of itself, there are some people who will say that that is additionally helpful at decreasing the urges. There are others that don't find it all that useful. It's not been something that I use in any regular way. And there were questions with naltrexone from the beginning. Maybe it helps, maybe it doesn't. So there have been some trials with naltrexone and bulimia nervosa from back when. I don't know more about it than that. OK, thanks. I think we do. Thank you.

eating disorders

anorexia nervosa

bulimia nervosa

binge eating disorder

ARFID

DSM-5 criteria

Dr. Evelyn Atiyah

family-based treatment

fluoxetine

neurobiological research

dopaminergic pathways

COVID-19 pandemic

social media influences

early intervention

individualized treatment