 First of all, thank you for having me. It's good to be here. My name is Saib Khalsa. I'm the director of clinical operations at LIBER and an associate professor at the University of Tulsa as well as most recently the director of the LIBER Float Clinic and Research Center. So today I'm pleased to talk with you about some new findings. The title of my talk is impact of flotation therapy on body image and anxiety in anorexia nervosa, a randomized clinical efficacy trial. And although it's remote in pandemic times, I'm very pleased to be with you virtually. I hope you're all doing well and are safe. So first I'd like to start out with I have no financial disclosures related to this talk. And also I'd like to emphasize at the outset of this talk that clinical trials are a huge team effort. And I want to start by really acknowledging the team that helped make this possible and especially the people who made strong contributions to this work whose names are listed in bold. In particular my longtime colleague and friend Justin Feinstein as well as Flux who's done a lot of wonderful analysis related to this project. Scott Mosman the director of the Eating Disorders Program at the Laureate Eating Disorders Center as well as Martin Paulus the scientific director of LIBER. So I've organized the talk into three parts kind of the past present in the future. I'm going to start by giving you a reminder of the previous study of floating anorexia nervosa that we undertook since it's been a few years since I've been back here and there's likely many people in the audience who are new. And then I'll talk about the present study which is a current randomized clinical efficacy trial and then end on the future implications of this work. So to begin with I'd like to introduce you to anorexia nervosa which is a rare affecting about one percent of the population but extremely deadly disorder. It has mortality rates on par with schizophrenia and bipolar disorder and it affects women 10 times to every one male although it does affect males. In current treatments for this condition the psychiatric disorder really only have moderate efficacy. We're really struggling right now to find ways to help people recover from eating disorders. Medications play a pretty limited role in treatment as you'll see and in terms of the definition of this disorder we don't have any real sort of objective biomarkers for indicating anorexia nervosa other than severe food restriction that causes a distinct weight loss. That combined with an intense fear of weight gain and a disturbance in body image perception as exemplified in this cartoon over here where we have a slender female looking in the mirror and seeing somebody that is obese and certainly not in proportion to the image of her body that everyone else sees. So those are the three classic diagnostic criteria for giving somebody an anorexia nervosa diagnosis. However we know that there are a lot of other associated features that are relevant to the disorder so heightened anxiety, high levels of obsession and perfectionism at the level of personality traits go hand in hand with this disorder. Oftentimes these are present even before somebody has a diagnosis and then difficulty recognizing internal body sensations and emotional states. So these are individuals who often feel full after eating a very tiny amount of food. They often experience abdominal bloating or cramping and as a disorder these latter aspects have not really been integrated into diagnosis but it's well known at a clinical level that they're very difficult to treat and they're poorly understood. So in that regard what I'm really going to focus on in this talk is body image and anxiety and in a separate line of research I focus a lot on understanding the manner in which these individuals have difficulty recognizing their internal body sensations. So I'd like to start with a sort of announcement I'm proud to indicate that the initial trial that we presented a few years ago at the Float Conference has been officially published. This is the manuscript you can find out, find it in this journal Frontiers in Psychology. It was published in October of last year and this was a safety study so it was what we call an early phase clinical trial. We were just getting a look at how patients with this disorder would respond when being exposed to the float environment. We really had no clear sense of how that would go. We had some hypotheses so we thought that it could be safe and we made several predictions. The first being that there would not be any adverse physical effects such as dizziness, falls, or blood pressure drops. These are things that are commonly encountered in the disorder, particularly when people are acutely underweight and about to be hospitalized or are hospitalized. We also thought that there would not be major adverse subjective effects. In other words, they would not worsen people's anxiety levels, their stress, or mood, or other aspects of their body sensation or body image. Again, these are things that are often heightened in these individuals and so we were aware of the possibility that they could be increased but we actually thought that floating would not adversely impact these effects. Finally, of course, something that's probably more familiar to most of you and the reason why many of you find floating so useful in your lives is that there's the possibility of positive effects on these variables. One thing when I talk about clinical trials with my students and one of the first things that I emphasize is that clinical trial is essentially a hypothesis test in which you develop a hypothesis and you use statistical methodology to ask a question in the universe and get an answer. You really can only get effectively one answer. That's baked into something called a primary outcome. You can have a bunch of other questions in mind and you can look at other ideas, but from the standpoint of powering your study for a particular question, you really can only have one outcome. In our study, we had the primary outcome that there would be no adverse physical effects because we felt like if there were any adverse physical effects that we observed then that would even prevent the study of potential subjective effects for clinical benefit. We did look at these other effects as secondary outcomes in the paper that I just mentioned, but again we only had one primary outcome that we could really focus our inferences on. This was the structure of the study. It was a four-float study and again because it was the first time anyone had taken patients with a diagnosis of anorexia nervosa into a float environment, we used a graduated approach where they started by floating in a chair, which we considered to have some of the elements of flotation but not obviously the ones that involve being in a pool filled with epsom salt. We transitioned them to the open pool at Liber and then finally gave them exposure to the domed pool. This open pool was a feature that Justin sort of originated in collaboration with the folks at FloatAway and it's been very effective at helping individuals with heightened levels of anxiety tolerate exposure to an environment where they normally often report claustrophobia and whatnot. We had a bunch of wireless and waterproof sensors for measuring blood pressure and heart rate and we also gave them an experience of floating without sensors as well just so they could have that. We could look at subjective effects. So in a clinical trial you have to track things very carefully. So this is our flow diagram that describes how we brought people through the study. We considered 51 out patients with the diagnosis of AN for the study and we ultimately ended up only consenting 23 for the study. We were targeting to get a total of about 22 and we ended up with a little more than that. We had 21 who completed all four float sessions and we included all of them in the analysis. If you're interested in some of the details you can look at this clinicaltrials.gov identifier for how we designed the study at the outset before we started collecting any data. In terms of our primary outcome we were looking to see whether these patients would show orthostatic blood pressure changes. In particular would they have blood pressure drops when they stood up out of the float pool that could cause them to feel dizzy that could potentially lead them to fall and bonk their head when getting out of the pool and in this case we did not find any evidence of orthostasis either in systolic blood pressure which is basically these are all the dots reflecting each individual in the study. We would have had to have seen one of these dots or more below this red line. We did not see that. We didn't see that either for the systolic blood pressure or the diastolic blood pressure measure right there. So that finding gave us a lot of confidence in the ability to continue looking forward at doing other studies in this population. Although I'm not showing it here we looked at other subjective effects and like physical discomfort, other kinds of medical conditions. We did not see an increase in physical adverse effects in that study. In terms of subjective effects kind of the positive effects that we were considering might happen. We saw evidence for that. So it may not surprise you to see that we found stress and anxiety reduction. So this is a self-report scale basically an anxiety questionnaire that we gave patients before and then after each of their floats and what you can see is that these are the raw scores on the scale which has a range the way it scored from 20 all the way up to 80. So somebody with a state score of 20, a state anxiety score of 20 would be very low whereas somebody who is in the 40 to 50 range as you can see here is actually fairly elevated. Healthy individuals usually tend to be much lower. And so what you can see is that across each of the float sessions we saw statistically significant reductions in the amount of anxiety that they experienced after being in each of the float sessions. When we looked at measures of what's called effect size which is really kind of the distance between these two averages we found that the effect on anxiety was actually very large. And again this may not surprise all the float aficionados. I think this is kind of consistent with a lot of the anxiety and stress reducing experiences that people have in the absence of even using one of these anxiety scales. Now we also looked at body image and one thing that I didn't emphasize in the beginning of the talk is that body image is something that really is one of the last clinical characteristics of anorexia nervosa to change. So bringing people from an underweight state back into a normal weight state is the primary focus of treatment. It's well recognized that body image is fairly inflexible even by the time people are well enough to leave the hospital. So we were very pleased in that regard when we found some evidence of a change in body image disturbance and this is looking at something called the pump or the percentage of maximum possible change that looks at the amount of change that you can see across the range of an entire scale. In this case we saw some differences, statistically significant differences in a measure called body image dissatisfaction on something called the photographic figure rating scale. And those effect sizes had a range but on average they were in the medium effect range as you can see. So what was this scale exactly, this photographic figure rating scale. As you can imagine there's photographs with figures and there are ratings that are being made. So here's an example of an image. So basically individuals are looking at images of female bodies with the faces obscured. There's a range of different bodies and they're indicating the one that they think best reflects the body that they currently see as well as the body that they would like to have. Now in anorexia nervosa the typical pattern, let's say this was somebody's actual BMI, they might look at themselves and say oh well I actually, this is how I look. And then when you ask them what's the kind of body that you want to have, this is not an uncommon pattern for you to see. So this was certainly what we saw in our participants pre-float but the general pattern of the finding post-float was a bit of a normalization. So kind of something like this where you can see that there was a reduction in the difference between the current body that they saw and the ideal body such that it was more normalized to the actual body that they inhabit. And in this regard, you know this is a very visually based scale. We sort of interpreted this as if it's almost as if they saw themselves differently. So in terms of the conclusions from this study, we were sort of optimistic. We found that partially weight-restored outpatients with anorexia nervosa had tolerance to the physical and mental effects of floating. But more than that, we saw some acute anxiety reductions that were fairly large. This ended up being about 15% of the percent of maximum possible change. So you could say a 15% total reduction on the state-trading anxiety inventory. And then we also saw improvements in body image dissatisfaction. And those were smaller on the order of about 4.7% on this photographic figure rating scale. So based on these data, we decided to move ahead on what's called a randomized clinical trial to investigate the efficacy of floating in reducing these symptoms. So in the previous study that I just presented, there was really only one group that went through and experienced floating. We didn't have what's called a control or a comparator group. And so one criticism of that study could be that well, you know, everybody knew that they were floating and you would really have difficulty disentangling the expectations of floating from the study itself because you didn't have a control group. So the next step in that process of doing clinical trials would be to have a control group. So that's what we did. And so I'm going to present the results, initial results from that study here. What I'd like to emphasize another point is that the first study was a safety study. And this study is what we call a clinical efficacy study. So here our target really is improvement in symptoms related to the disorder. In other words, is there a therapeutic effect? Our hypothesis was that floating would reduce body image disturbance and anxiety symptoms. This time, we decided to focus on inpatients, so a more acutely ill group who were being treated in the Laureate Eating Disorders Program, which is housed in the same building as Liber. And we thought that floating would show evidence of reductions in anxiety and improvements in a body image relative to a usual care comparator group. And I'll explain what the usual care is in a moment. This was our goal. If you want to read more about the details of the study, you can look at clinicaltrials.gov, this identifier over here. We took a lot of what we learned from the previous study. In fact, we used a lot of the data that we gathered from the previous study. You conduct what's called a power analysis, which allowed us to say, based on the effects that we observed in the previous study, how many participants do we think that we would need to recruit in order to see an effect of floating versus usual care? And when we did that, what we found is that for our primary outcome, we would need about 44 patients in the flotation therapy arm and then 22 in the usual arm. We decided to randomize them on a 2 to 1 ratio such that more people were randomized to flotation therapy. And this was a true randomization. We had a randomization list developed by a statistician. So it wasn't that myself or the research coordinator involved could have got to pick who got to float based on their preference. This was really randomized. And we had several people, in fact, who were very unhappy that they were randomized to one arm versus another, but that was part of the agreement of them doing this study. So what we decided to do was to take these individuals and administer a larger number of float sessions. So instead of four float sessions, we doubled the number to eight. We also shortened the duration of the float to one hour. Part of this was pragmatic because these are inpatients on the previous unit. It's very difficult for us to get some of their time. But we also felt that a one hour float might be adequate enough to see certain effect. So what we did is people in the float group, we brought them from the unit down to the float clinic and then they floated. And the people who were in the usual care arm were receiving their regular clinical programming, their clinical treatment on an inpatient basis. And what we did is we found them on the unit and we just gave them the same measures at a certain time point and then about an hour later. And that would allow us to get a really good baseline of, well, if their symptoms were to improve simply because they knew they were in a clinical trial, then we would be able to detect that. We also planned to do longitudinal follow-ups six weeks, six months, and one year. And the goal here is to look at whether or not the effects that we would see, or at least that we were hypothesizing we would see would carry on over a longer term period, which if that was the case, it would tell us that there's additional measures of the validity and the utility of this in treatment settings. So what is usual care, this sort of UC acronym that I have here? So at the Laureate Eating Disorders Program, it's a highly individualized program. This is our campus. The eating disorder unit is over here. And then the kitchen and the dining hall is over here. And we have two tracks. We have an adolescent and an adult track. So we have adolescents between the ages of 13 and 17 receiving treatment and adults 18 and over. And it's a multidisciplinary treatment. So it involves physicians, whether they're psychiatrists, internal medicine, or other specialists as needed, nurses, licensed behavioral therapists, registered dieticians, a chef, a yoga teacher, and a school coordinator for the adolescents. So it's really a comprehensive treatment program. It also, in addition to medication, involves an intensive psychotherapy and relationship-based program with up to 40 hours of clinician-led individual group and family therapy each week. So as I mentioned, medications have a role, but it's rather limited in treating anorexia nervosa. And so therapy is quite a heavy component. There are five levels of care, starting with the most acute hospitalization when somebody is severely underweight. And then once their weight has been restored to a reasonable degree and they're no longer at certain medical risks like orthostatic blood pressure drops, dizziness, having a slowed heart rate, they can transition to something called residential care. And then as they move through these different levels, they get more and more independent. And so this is going to be important when we talk about what the inclusion and exclusion criteria are for the study. And I won't go into all these in detail. I'm just showing this to you. So you have a sense of the thoroughness and sort of all the criteria that we required in order for somebody to move ahead. So the primary diagnosis had to be of anorexia nervosa. They had to have a weight, a body mass index above sort of a very low level, no orthostatic hypertension. But importantly, they had to have gone from that acute stage of clinical treatment to more of a residential stage. And that allowed us to be reasonably confident that they could leave the unit, walk outside and walk down to our float clinic and float. We had a number of exclusion criteria, not no acute suicidal ideation or cutting. And as you can imagine, screening individuals with this level of detail who are this ill requires a lot of coordination. And so that really happened with close partnership with the Laureate Getting Disorders Program. So how did we define our outcome? So as I mentioned before, you really have one hypothesis test, which has to be crystallized into one primary outcome or endpoint. And this is how we did that. So we decided in looking at the effect sizes and seeing that we saw effects on anxiety as well as body image that we would decide to focus primarily on body image and particularly body dissatisfaction scores on the photographic figure rating scale because that aspect of body image is because body image is so difficult to change in clinical practice. We're reasonably sure that we would see something with anxiety as well. But we thought that if we could see a difference in body image, that would be particularly noteworthy. So we defined that outcome as body dissatisfaction, which is the current body image rating subtracted from the ideal body image rating assessed before and after each of the eight floats. So that's our sort of immediate primary outcome measure. And then again, assessed at six weeks, six months and one year follow up to see if there's a longitudinal sort of a long lasting impact. So that's the primary endpoint. We had a number of secondary endpoints. I won't go into all of them. Today I'll really be presenting results from this secondary endpoint, which was the change in anxiety level on the state trade anxiety inventory, the same scale that I showed you before, again, before and after each intervention. And then at six weeks, six months and one year follow ups. Okay. So in this regard, this is what we these are the participants that we brought through. So here's our participant flow diagram. We walk you through it. We considered 133 people as being eligible. And this included people that we knew of who are on the unit, not necessarily people that we consented or screened. We ended up consenting 86 people. And of course, in the COVID era, we ended up having to exclude some people who received COVID-19 diagnoses during their stay. And so what that we ended up with was we allocated or randomized 45 individuals to the flotation therapy arm and 23 to the usual care arm. We had a number of people who didn't complete all of their eight floats. Usually this was because they were discharged from the hospital before we could complete their floats. But you can see we ended up completing eight floats and eight usual care measurements from nearly 60 people. And in our analyses, we're able to integrate these people such that in total we actually had 67 people included in the analysis, which was one more than we anticipated needing based on our power analysis. So who were these people? I'm going to show you a very detailed chart and I'll walk you through it very briefly. But these are the numbers. And I think it's important that you be able to see this information to evaluate it for yourself. So they were all females. And that's based on the requirements of the Laureate Aiding Disorder Program. They did not differ in age. They were all fairly young, around 20 years on average. No differences in the years of education. Importantly no differences in the amount of diagnoses of major depression, generalized anxiety disorder, or obsessive-compulsive disorder, which are commonly co-occurring in this population. No baseline differences in their level of body image dissatisfaction. This is important if you're going to be looking for differences between groups. You don't want the two groups to start out on an uneven level. Similarly, no differences in their levels of trait anxiety on this same state-trade anxiety inventory. And with anxiety levels of 60, you can bet that that's pretty high on that scale. They also had no difference in their body mass index at admission or the lowest BMI that they had ever experienced. We did see, interestingly, some evidence. So there was, for whatever reason, statistically, greater numbers of people taking psychotropic medications in the float group and maybe even somewhat in terms of having a longer illness duration. But again, that was, we assume that was occurred by chance because these people were randomized. And importantly, they did not differ in all of these other clinically relevant measures. Okay, so I've been talking a lot about the setup of the study. What did we find? The first thing is that when we looked at the average float duration, the average float duration was about 49 minutes. So most people lasted most of the time. And in terms of our primary outcome on body image, we actually found an effect. So when we looked at the acute changes in body image dissatisfaction on the photographic figure rating scale, these little dots reflect all of the data for people in each of the arms, for each of their floats. So we're looking at the average effect, pre-float versus post-float across all eight floats, which is how we designed it. And these horizontal or these solid lines reflect sort of the change, the average change in each group. And what you can see is that there's, you know, these, these group, these lines are not parallel, they're not overlapping. And that's associated with a statistically significant group by time interaction, whereby the people in the float group actually had a lower degree of body image dissatisfaction. When we looked at the, the sort of effect size, we found that the float group had a 4.9% reduction of that pump scale, whereas the usual care really showed a minimal, if any, reduction. When we looked at the effect size, this was actually a bit smaller than what we saw with the initial study. But again, this is a statistically significant finding and it, and it is in line with the outcome that we predicted. Just for reference, the, to remind you that the pump reduction that we saw in the outpatient study, the safety study was 4.7% versus 4.9%. So I'm pretty confident that, that this is a legitimate impact of floating on this aspect of body image. Now, this is taking the float effect across all eight floats. What I'll show you next is data looking at the body image dissatisfaction for each of the eight sessions that we did. And what you can see is that for the most part, the effect is consistently occurring across all the sessions. For whatever reason, in sessions five and six, we don't really see the groups showing a difference. And I'll come back to that point when I, at the end of the talk, when I talked about the implications. So that was our primary outcome. Again, one hypothesis test and we found evidence in support of the hypothesis. But we also have other hypotheses that we can test. And so in terms of our secondary outcome, in terms of anxiety, this, this was state anxiety, we actually see a much larger effect. You can see that these lines are much further apart. The statistical significance value or the p value is much, much, much lower. And when we look at the effect size, in this case, we saw a pump reduction of 20% versus a 0.2% reduction in the usual care group, basically no reduction. That's a very large effect size. And this, again, reflects the average change across all the sessions. For reminder, with the initial study, the pump change was actually 15.2%. So the anxiety reduction in the inpatients of 19.7% is actually even larger. And so I think this is quite meaningful. It's evident at each individual session, even more so than for the body image results, you can see that the usual care group really showing minimal reductions and the flow group showing large reductions. One thing that you can see from this pattern is that the usual care group kind of started out with higher anxiety levels. And then over time, their anxiety in relation to just having anxiety measured changed somewhat. The flow group didn't seem to have as much change. But again, what we're really looking at is the difference between the pre and the post flow measurement for the flow group and then the usual care measurement as well. So very statistically significant finding on our secondary outcome with a very large effect size. Okay. So that's most of the data that I planned to show you today. I want to sort of review where we're at and give you a sense of just what goes into these kinds of studies. So we first randomized, we randomized our first patient in April of 2018. We enrolled our last patient in March of 2021. So our enrollment is 100% complete. And if you recall, we were going to do follow up up to a one year time point. So our last follow up is actually not scheduled to March of 2022, which is why I'm not presenting any of the follow up data here. So in terms of the status on this study, you know, we're kind of a glass nearly full. I'd prefer to look at it that way than nearly empty. And I am pleased with our follow up data. Our follow up rates at this point are quite high. Usually what you want to see in a clinical trial is an 80% or greater follow up rate. And usually over time, you see a drop off. So we're getting good follow up. And I'm confident that by the time we are able to get our last follow up, we'll be able to analyze the follow up data and have more information to share in that regard. Okay. So in terms of the implications of this work moving forward, let's talk about that. So first, this randomized control trial or RCT demonstrates floating's efficacy in reducing anxiety and body image symptoms in inpatients with eating with anorexia nervosa. There's evidence of clinical efficacy. And there's sort of two points that I want to emphasize in this regard. The first is that any change in body image is a big deal, given the rigidity of this construct in treatment. And we're definitely seeing evidence of that. We don't know if that if the evidence is sustained over the long term. But I've shown pretty clearly that that we do actually move one of these measures of body image as a result of float. Now, the second thing is that we found a pretty large effect size reduction in anxiety. So in this population, we know that medications for anxiety such as benzodiazepines are pretty ineffective in treating anxiety and eating disorders. They also have side effects such as tolerance and they have addictive potential. And so in this regard, floating could potentially be viewed as a behavioral benzodiazepine or a non-pharmacologic option, one that has few side effects. And as you can see, has a pretty strong acute impact on anxiety levels. So what would we use this for? What are potential implications in a clinical context? I'd like to highlight two that I think of, but when I move to the breakout room, I'd be curious to hear any other thoughts from you. The first is the potential for float assisted augmentation of psychotherapies that are focused on body image or anxiety. There are psychotherapies that are specifically focused on each of these. So the idea would be that maybe somebody floats and you get them into a state where they've experienced a change in their body image or their anxiety. You then work with them in a psychotherapeutic context using other tools and techniques, maybe cognitive reappraisal techniques or other behavioral techniques. Alternatively, maybe you could start with a psychotherapy session working on those techniques and then bring somebody into a float environment. So I think there's a lot of exploration and investigation that could be had in that front. The other thing that I think is very interesting and worth pursuing is float assisted augmentation of meal-related anxiety. So we know that the surest way to make somebody with an norexia nervosa anxious is to put food in front of them. That's a main hallmark of the disorder. Some people view them as anxiety disorders where the focus of fear is food. So if you can reduce anxiety to such a large extent with floating, what happens then when you put a meal in front of them? What happens then in treatment where that usual care of that sort of 40 hours plus of full-time work where they're trying to put on weight again in a very supervised setting, learn how to eat healthy? What happens if you use floating as a potential tool to help them overcome their fear of food? So I think those are two avenues that would be would sort of be probably immediate and natural next step. Of course, there could be others and I look forward to hearing more. Okay, so to conclude, I think we've shown so far that in these two studies that individuals with norexia nervosa across the illness spectrum can safely tolerate the flotation therapy experience. I think that's important that we didn't know that before. We now have evidence of short-term improvements in body image disturbance and acute anxiety levels across two clinical trials. The magnitude of the effect size is small for body image, but it's very large for anxiety and the overall change is consistent. So I think it's reliable and I think it's reasonable to expect that you would see that in other settings. We also saw it with the first float, so it may be that in order to have these impacts, you don't necessarily need to have multiple floats if all you're targeting is an acute anxiety reduction or trying to budge body image in a particular way for treatment. Maybe one or two floats is enough to begin working with somebody on these challenges. As I mentioned, the magnitude of the effect is consistent. We don't know whether these effects are sustained over the long term and one thing that I can tell you we know from the clinical literature is that these patients frequently relapse even within the first year of being discharged from the hospital. So it's not uncommon for people to re-encounter triggers that they're eating disorder and to have challenges maintaining their clinical improvement. So I'm not exactly sure what we'll see in our longer-term follow-up, but as soon as we have it, I'll report it. The final thing is that really the neurobiological mechanisms of these effects await discovery. So we don't know what's happening in the brain in relation to the anxiety reduction and the body image change, and this may be a fruitful avenue for additional investigative inquiry. So on that note, I would like to conclude and thank you for your attention and wish you a happy float conference, safe travels, and I look forward to speaking with you further about this.