Assistant Professor Kelsey Hagan Hi, everyone.  Thank you for having me. My name is Kelsey Hagan.   I am an Assistant Professor of Psychiatry at  Virginia Commonwealth University in Richmond,   Virginia, in the United States, and I’m  also an Affiliate Editor for the Journal   of Child Psychology and Psychiatry. Today,  I’m going to be talking about other specified   feeding and eating disorders, or OSFEDs, and  providing an in-depth look of those disorders. So, to give you an overview of  what we’ll talk about today,   we’ll provide a history of other specified  feeding or eating disorders, or OSFEDs,   and discuss what they are. I’ll review each  example OSFED that’s provided in the DSM-5.   We’ll do a specific focus on a typical anorexia  nervosa, as this disorder has received a lot of   attention in the past decade, and rightfully so,  and then, we’ll discuss the future of OSFEDs. So, what is an OSFED? Beginning with the  DSM-III, eating disorders have included   a residual diagnostic category for “Clinically  significant presentations of eating disorders   that did not meet the full threshold criteria  for a specified feeding or eating disorder.” So,   at that time, anorexia nervosa and bulimia  nervosa were the only two specified feeding   or eating disorders. In DSM-III these were  termed ‘atypical’ eating disorders. You might   have heard the EDNOS, or eating disorder not  otherwise specified, and that was in DSM-IV. However, there were some big challenges  with this residual EDNOS category. One   was that it was pretty heterogeneous,  meaning that the symptoms that people   experienced were across the board and very  different. And so, in the treatment setting,   this hampered clinical communication,  because you might get someone come in   with an eating disorder or an EDNOS,  and you wouldn’t know what that means. You wouldn’t know if it meant  they had binge-eating and purging,   if they had lost a significant amount of weight,  if they had binge-eating only, for instance. And   what’s more is that most people with an eating  disorder fell into this diagnostic category,   and it was over 50% of people that fell  into this category. And the other thing   that was pretty tricky about this category is  some people felt, and I think rightfully so,   that EDNOS meant that the eating disorder was not  as medically severe or psychologically severe as   the threshold eating disorders, when in fact,  research suggested that people with an EDNOS,   or an OSFED, were just as severe as people  with the threshold eating disorders. And so, I’ll give an overview here of what this,  kind of, looks like, in terms of proportion of   cases. So, in the time of DSM-IV, we had about 15%  of people meeting criteria for anorexia nervosa,   about 30% of people meeting criteria for  bulimia nervosa, but then we had this whole   host of people out – that were meeting OSFED or  EDNOS at that time. So, to address the issue of   so many people falling into this residual  eating disorders category, the DSM-5 Eating   Disorders Workgroup was tasked with decreasing  the number of people who did not meet – or who   experienced clinically significant eating disorder  symptoms, but did not receive specific diagnoses. And so, to do this, the workgroup added  binge-eating disorder, which is when someone   experiences objectively large binge-eating  episodes, at least once per month, or once   per week, for three months. They also added an  unspecified feeding or eating disorder category,   which really encompasses when there’s not  enough information provided to make a diagnosis,   but we have some sense that something is  maladaptive about the eating behaviours   or eating cognitions. And then we –  they also developed the OSFED category,   with specific labels for distinct and  understudied conditions. And the purpose   of these specific labels was to generate some  research about each of the specific labels,   with the hope that in future revisions of the DSM,  we might be able to include that diagnosis or,   potentially, say that it might fit better within  another diagnosis that’s already specified. And so, now, I’ll go into briefly each of the  specified labels under the OSFED diagnosis and,   again, we’ll focus more on atypical anorexia  nervosa towards the end. So, one of the longest   standing other specified feeding or eating sin –  eating disorder presentations is something called   ‘night eating syndrome’. This was first described  by Alfred Stunkard and colleagues in 1955, and so,   it’s been around quite a while. The diagnostic  criteria, or proposed diagnostic criteria,   have changed quite a bit, but back in 2008, some  experts on night eating syndrome all convened,   and they came up with some expert  consensus criteria. They had six   specific criteria they listed, but the two that  I’ll highlight here are that there needed to be   significantly increased food intake during  the evening, or at night, and then, also,   that the person should be aware of their  nocturnal ingestions. So, this makes it   distinct from sleep-related eating disorder, where  people are not aware that they’re eating at night. The estimated prevalence of night eating syndrome  is about 1.1 to 4.6%. Now, night eating syndrome   has been around for quite a while, or at least,  we’ve recognised it for quite a while, and yet,   it’s still not incorporated into the DSM. There  was a lot of initial interest in this diagnosis   and a lot of research being conducted, up until,  I would say the early 2000s, to – up to 2010.   However, it has remained an OSFED, and it seems  as though it might stay in that category. There’s   some initial evidence suggesting that it might  better fit criteria for binge-eating disorder,   or we might need to consider whether  it’s more of a sleep-related disorder. The second diagnosis that’s pretty  well-known is purging disorder. So,   this was first formally recognised  by Pam Keel and colleagues in 2005,   although research has – Researchers have  talked about this in some way, in some fashion,   and under different labels, since the 1980s. So,  just to re – to orient you to what purging it,   it’s “The forceful expulsion of stomach  or intestinal contents, to influence body   weight or shape or to compensate for eating.” And  then DSM-5 described it as “Recurrent purging to   influence body weight and/or shape, in the  absence of objective binge-eating episodes.” And two really important things to highlight  here about this are that a minimally healthy   body weight is required, so a person cannot  be underweight to have purging disorder,   and loss of control eating is not required,  as well. And there’s been some debate about   whether or not someone should ex – have  experiences of loss of control eating or not,   and that’s part of the research  question for purging disorder. The prevalence of purging disorder is about 1.3 to  6.2%, and in terms of the future of this diagnosis   in the DSM, there’s limited data on the long-term  course, although Keel and colleagues have been   working on this research. And what’s interesting  is people with purging disorder only represent   about five to 10% of people seeking  treatment for an eating disorder. Now, there’s been some talk in the field  and some discussion of whether or not   purging disorder should be subsumed  under the bulimia nervosa category,   or should remain its own distinct entity,  with some research suggesting that purging   disorder has a higher accrued mortality rate  than bulimia nervosa. So, the jury, I think,   is still somewhat out on whether purging disorder  will be its own category, will remain an OSFED,   or get subsumed under bulimia nervosa, but  nonetheless, this is a pretty severe presentation. And then one other OSFED label is subthreshold  bulimia nervosa, or binge-eating disorder,   which I’ve lumped together here. These are  essentially just less – presentations wherein   behaviour – behavioural frequently of binge-eating  for both bulimia nervosa and binge-eating   disorder, and frequency of inappropriate  compensatory behaviours for bulimia nervosa,   fall below the threshold of what the specified  diagnoses require. And so, as you’ll note from   DSM-IV to DSM-5, the behavioural frequency  criteria for bulimia nervosa changed from the   behaviours needing to occur at least twice a week,  on average, over three months, to once per week,   on average, over three months, and then there’s  also the addition of binge-eating disorder. The DSM workgroup recognise that some  people with these disorders might   experience behaviours less than once per week,  on average, and so, it might be the case that   they would go on to develop a full threshold  presentation, or maybe they had only had the   symptoms for a short period of time and  were then likely to go on to develop it. So, the prevalence of these  syndromes is somewhat high,   and in terms of the future of sub-threshold  BN and binge-eating disorder, research has   found little meaningful difference between full  and threshold – sub-threshold bulimia nervosa,   in terms of comorbidities and impairments. And  for binge-eating disorder, or sub-threshold   binge-eating disorder, research has  also suggested that this might apply to   children and youth who don’t experience those  objectively large binge-eating episodes yet. So, now, turning our attention to atypical  anorexia nervosa, which is going to be the   focus of most of the rest of the presentation,  I’ll give a brief overview of the history of   this presentation and then talk about some  research that I was involved in that was a   critical synthesis of the literature so far.  So, atypical anorexia nervosa was actually   first described in 1973, and the definition  resembled what we know of it today, mainly. And then, in the early 2000s, what started  happened was eating disorder treatment centres   began seeing a dramatic increase in the number  of patients who met all the DSM criteria for   anorexia nervosa, except low weight, and  the medical complications that they were   presenting with were pretty commensurate  to those with low weight anorexia nervosa,   or typical anorexia nervosa, and they also were  reporting pretty severe psychological symptoms,   as well. And so, to address this  issue and spark more research,   the DSM-5 workgroup added and described atypical  anorexia nervosa, and it was defined as “All of   the criteria for anorexia nervosa being met,  except that despite significant weight loss,   the individual’s weight is within  or above the normal range.” And so, my colleagues and I undertook  a review earlier this year – well,   that was published earlier this year, and  we started it last year, to basically,   synthesise the status of the research on atypical  anorexia nervosa and delineate future steps that   are needed for our understanding of this  disorder. And so, to do this, we conducted   a systemic review and meta-analysis in accordance  with PRISMA guidelines, using a standard set of   guidelines to really regulate and standardise how  we conduct meta-analyses in literature searches. So, our overall aims were essentially to compare  atypical anorexia nervosa to typical anorexia   nervosa and non-eating disorder controls on  psychological, medical and sociodemographic   variables, including gender and race. We also  sought to conduct a critical synthesis of   definitions of significant weight loss. As you saw  from the definition that’s provided in the DSM,   significant weight loss is pretty ambiguous,  and so, we don’t quite know what that means. And so, our methods were that we searched some  major databases and also, the grey literature,   which means literature that is not yet  published. We specifically targeted PsyArXiv,   which hosts pre-prints or – of work that  might have been submitted for peer review,   as well as ProQuest, which tends to host  master’s theses and dissertations. We   searched literature that was published from 2013,  or the publication of DSM-5, up to September 2022,   which was when our manuscript was submitted.  And our search term was quite simple,   it was just ‘atypical anorexia nervosa’. And  in terms of the studies that we included,   we included those that were published in  English, or an English translation was available. We used that DSM-5 definition of atypical  anorexia nervosa. We also included studies that   had novel data on at least ten individuals with  atypical anorexia nervosa, and the studies needed   to compare atypical anorexia nervosa to typical  anorexia nervosa or non-eating disorder controls. So, this is – over here, to the right, is a  flowchart of our systemic review process. In   total, we identified 293 unique manuscripts, and  then after the screening process, we ended up   with 24 manuscripts for inclusion in our review,  and now, I’ll dive into some of the results. So,   we conducted a meta-analysis, which,  essentially, compare averages across   the studies between groups. And so, here, we  conducted a meta-analysis on body mass index,   which is the standardised measure of  weight, of individuals with atypical   anorexia nervosa versus anorexia nervosa.  And this was more of a sanity check, and so,   of course, we found that people with atypical  anorexia nervosa had higher BMIs than those   with typical anorexia nervosa. We also found  that people with atypical anorexia nervosa   had lower BMIs than those without an eating  disorder, so those non-eating disorder controls. In terms of the results of our synthesis of  definitions of significant weight loss, we   were surprised to find that only three out of the  24 studies described how they defined significant   weight loss. Two of those studies required a  reduction of at least 10% in body weight and then,   another study was specifically examining  psychological symptoms that were associated   with different levels of weight loss from their  – from the participants highest lifetime weight,   and those weight loss values were defined as  five/ten/15% weight loss from the lifetime   highest weight. But what we did find  was that most studies simply required   persons or participants to be at or above a  minimally healthy body weight for their height,   which might have been something  like a BMI or at least 18.5. And so, in terms of eating pathology, what we  found here was quite interesting and has been   echoed in another meta-analysis by Sarah Johnson,  that focused specifically on eating pathology   and different types, and that was that on  almost all measures of eating pathology,   individuals with atypical anorexia nervosa  scored higher than individuals with anorexia   nervosa. We couldn’t conduct a meta-analysis  here because of the heterogeneity of measures. Now, turning toward other psychological  symptoms, I’ll highlight depression here,   as that’s one that’s commonly comorbid  with eating disorders. What we found   was that individuals with atypical anorexia  nervosa reported more depression than those   with typical anorexia nervosa. Now, in terms of  controls, not surprisingly to us was that those   with atypical anorexia nervosa reported more  depression than non-eating disorder controls. Now, turning toward the physical symptoms, what  we found were that three studies of adolescents   reported similar cardiovascular symptoms, and  – but perhaps a somewhat lower frequency. So,   the cardiovascular symptoms, such as  bradycardia, were similar between atypical   anorexia nervosa and typical anorexia  nervosa adolescents. We also found that   bone mineral density was higher in atypical  anorexia nervosa than in anorexia nervosa,   and one study found no difference,  and I’ll go into that in a moment. And then there were two studies on  central nervous system morphology,   or grey and white matter density, and there was  no difference from controls here, but there was   not a comparison between anorexia nervosa and  atypical anorexia nervosa in this study. And so,   comparing bone mineral density in atypical  anorexia nervosa and anorexia nervosa,   we found here higher bone mineral  density in atypical anorexia nervosa,   but I do want to note that we – there is only  three studies here, so more research is needed. And in terms of menstrual disturbance, so  menstrual cycle is often affected by eating   disorders, and is particularly affected by eating  disorders characterised by extreme restriction,   such as anorexia nervosa. So, we wanted to  understand what this looked like in atypical   anorexia nervosa versus anorexia nervosa. What  we found was that menstrual cycle disturbance   was less frequent in atypical anorexia  nervosa than anorexia nervosa, but again,   I want to highlight that there’s only a few  studies here, and more research is needed. Now, turning towards the sociodemographic  characteristics, and how they compare across   the two groups, really intriguingly, what we  found was that a greater proportion of males   – or atypical anorexia nervosa had a greater  proportion of males than anorexia nervosa. And   then, with race, we also found that atypical  anorexia nervosa had a greater proportion of   non-white individuals than those with anorexia  nervosa, which was also quite intriguing to us. And so, to wrap it up and sum it altogether,  compared to typical anorexia nervosa,   atypical anorexia nervosa reported similar or  higher levels of eating disorder psychopathology,   similar levels of other psychological  symptoms. There are significant medical   consequences associated with atypical anorexia  nervosa, and a greater proportion of non-white   persons and men with anorex – or with  atypical anorexia nervosa. Taken together,   these findings were really intriguing to us,  because people with atypical anorexia nervosa   often report for – when they do report their  symptoms and report for care, potentially,   identifying as a non-white person, and also,  identifying as male, might not conform to   stereotypes that providers have about who gets  an eating disorder. And so, you know, this might   have implications for how someone with atypical  anorexia nervosa gets treatment and accesses care. And so, what remains unknown about atypical  anorexia nervosa? So, as you can see,   there is a lot of heterogeneity in how  folks were defining significant weight loss,   and I think as a field, we need to figure out  how are we going to define significant weight   loss? Is it even important to define significant  weight loss? What does that look like? And also,   the long-term course of atypical anorexia  nervosa is not well understood. And so,   thinking about the future of this disorder,  a lot of people have called for a unified   diagnosis of anorexia nervosa, with perhaps  a low white – a low weight specifier,   in order to recognise the severity  of atypical anorexia nervosa. And bringing it all home, and thinking about  the future of OSFED and what it’s going to look   like in the future revisions of the DSM, so, the  introduction of OSFED to the DSM did meet the goal   of increasing the prevalence of the specified  feeding – the specified eating disorders,   rather. So, the prevalence of anorexia nervosa  increased, as did the prevalence of bulimia   nervosa, but what’s really important to know is  that OSFED still remains the most common eating   disorder diagnosis and constitutes up to two  and a half times more cases than the specified   disorders. And, again, importantly, OSFED is  associated with substantial disease burden,   and even more so than the specified eating  disorder diagnoses, according to recent research. So, thinking about the future of these disorders,   there’s ongoing consideration that’s needed  for eating disorder diagnoses to potentially   decrease the prevalence of this disorder, and  then increase the specificity of other disorders.