[MUSIC PLAYING] We are the Association for Child and Adolescent Mental Health or ACAMH for short. And this is ACAMH Learn. My name is Dr. Virginia Carter Leno. I'm a Wellcome career development fellow at the Centre for Brain and Cognitive Development. And I think Professor Mandy's talk has really set the scene really nicely for some of the themes I'm going to build on today, primarily talking about when we're thinking about autism and eating disorders, developmental mechanisms, and community priorities. So I think part one of the talk really nicely established that there is an increased prevalence of eating disorders or a higher risk for eating disorders in autistic people and some potential reasons behind this phenomenon, largely using qualitative work, hearing from autistic people themselves with lived experience of eating disorders about some of the reasons why they think they develop those eating disorders in the first place. But there isn't actually that much empirical work testing these putative hypotheses. So actually using data to work out the associations between these constructs and find evidence for these mechanisms that we think are driving vulnerability to develop or experience eating disorders in autistic people. So I'm a developmental psychologist. So I think about things from early development into the lifespan. And I think this here is a really helpful approach to take as well. So I'm going to give you two examples of some work we've done trying to use this longitudinal or developmental approach to test some of the hypotheses that well so clearly laid out in his presentation. And then the second part of this talk will actually be outlining the idea of people with lived experience, really giving them a voice and using their experiences and their opinions to guide where we focus our research. And so I'll give an example of this approach because I really think this should be applied more broadly in the field of autism research but is really helpful in focusing where we might want to put our attention and our future efforts in this field. And then I just wanted to acknowledge here-- so generally, research is a team sport. And most of this work has been conducted with this group of people here. So myself and Professor Will Mandy, who spoke before, also Dr. Moritz Herle at King's College London, and Johanna Keller, who was a research assistant with us during some of these projects. So when seeking to understand the link between autism and eating disorder or disordered eating, we can start to think about some of the factors or mechanisms which might be kind of in the middle between autism and eating disorders, which could explain this association. So one kind of group of things to think about is autistic characteristics. And I think Professor Mandy laid out very clearly some of the things that autistic people and community stakeholders think could be important characteristics to think about, such as sensory processing and interoception, preference for predictability, tendency to develop fascinations, and autistic cognitive or social styles. But I think what Professor Mandy's work also really clearly highlights is this idea of experiences of distress. We could also think about this as mental health problems, as a way of how we measure experiences of distress in the population. So these are the two groups of potential risk factors that I'm going to focus on with the examples I'm going to talk through. So the first kind of piece of empirical evidence, trying to test some of these hypotheses that have come from people with lived experience and qualitative work, was zooming in on fussy eating. And this is something which emerges quite early in childhood and is very typical developmental phenomena that children can be quite fussy about the things they want to eat. When I say fussy, I mean-- I'll give you an example of the questions we ask-- but only wanting to eat specific things or just finding kind of eating a range of food more difficult. Now, in the general population, this has been proposed potentially as a kind of temperamental difference in children which could make them more likely to experience disordered eating later in life. People haven't really thought about this through the lens of autism, though, even though there's quite a lot of evidence to suggest that autistic people, when they are children, do show more restrictive patterns of eating. So this piece of evidence work was trying to bring together what we know from autism research and what we know from eating disorder research to test a putative explanation for why autistic people might be more vulnerable to developing eating disorders. So what we did was we used data from an existing cohort. I would also say as part of this work, we actually did a review of all of the longitudinal cohorts, which are kind of easily accessible, that we could find to see who had measured relevant constructs to thinking about autism risk factors and eating disorders. And it's actually surprisingly few and far between. So I think that is another piece of future work, is building longitudinal data sets which measure things, which are really important to understanding autistic mental health. These were collected quite a while ago. And so maybe they don't measure everything we want. But they had enough that we could test some of these hypotheses. So this data set is Children of the 90s or ALSPAC, which is a population representative data set of all children born in a defined geographical region who were then followed up from early in life until adulthood. And so from this, what we used is they had measurement of autistic traits at multiple times-- so when children were 7, 11, and 14. This data set also had measurement of fussy eating. And let me just give you an example of when I say fussy eating, this is what the researchers were asking parents. So how worried are you because your child is choosy? How worried are you because your child has feeding difficulties? And how worried are you because your child is refusing food? And so a higher score would reflect parental concern about their child's fussy or restrictive eating. So maybe I'll just go back here. And this was quite a large data set. So we have nearly 9,000 people who had measurement of the constructs we were interested in. And then finally, at age 16, we had self-reported disordered eating. We took kind of a broad approach to this. It wasn't a specific sort of diagnosis, like anorexia nervosa or binge eating disorder. It was more asking about symptoms of eating disorders and treating them as a more continuous construct, because, actually, the numbers of people in this sample who met criteria for specific types of eating disorder who kind of had that diagnosis was quite low. But if you treat these as more continuous traits, then there is actually quite a lot of variation in people reporting what you would consider symptoms of eating disorder. But they might not always have a diagnosis. So it's kind of a broader approach to measuring these constructs that we're interested in. And then because we have all this data measured at multiple time points, it gets a little confusing. And one way to treat repeated measures of data is to do structural equation modelling, where we basically are interested in the intercept, which is the starting point. So this would be age seven of someone's level of fussy eating or level of autistic traits. And then the slope, which is how they change from age seven until our final time point-- so through childhood into adolescence, are they increasing? Are they decreasing? Some people's fussy eating might decrease as they get older. That's generally the trend. But some people, it might persist. Some people, parents even might report they think it's getting worse. And so this type of statistical approach allows us to capture these kind of dynamic developmental changes, which we think are really relevant in understanding our outcome of disordered eating at age 16 in adolescence. So what did we find? Well, I should also add here that we did pre-register as much of the analysis as possible because we did have quite clear hypotheses that we thought-- autistic traits, your starting point or your intercept. At age seven, that would predict your change in fussy eating over time such that if you had higher autistic traits, you would either show a stable level or an increase in fussy eating over time. And then this, in turn, fussy eating would be predictive of disordered eating. So if you have higher levels of parent-rated fussy eating and it's a kind of steeper slope of change, so it's not decreasing, this, in turn, means you're more likely to report disordered eating in adolescence. And that's basically what we found. So we did find this hypothesised pathway going from your starting point or your intercept to autistic traits, leading up to fussy eating. So more likely parents to be reporting fussy eating, which, in turn, was associated with a higher likelihood of reporting disordered eating in adolescence. And just to give you a kind of sense of how much of the variation in eating disorders could be explained or how much of the link between autistic traits and eating disorders could be explained, although it's statistically significant, it's not a huge amount. So this kind of pathway, moving from autistic traits up to fussy eating and then back out to disordered eating, that could explain about 4% of the association between autistic traits and eating disorders. 4% is not a huge number. And so this is telling us, yes, fussy eating might be important. But actually, there are probably a range of other factors which can help us to understand this link between autistic traits and disordered eating. That's just to give you the statistical details of the test for this mediation pathway in blue here. So I guess the take home message from this first piece of work-- there are a couple of results that I think are really important. So overall, fussy eating does decline between 7 and 13 years in this general population sample. But this decline was shallower in children with autistic traits. So that's what that association is picking up. The children who had higher autistic traits, that decline in fussy eating was less rapid or shallower. And then this change in fussy eating predicted eating disorder behaviours in adolescence. This association of what we call the indirect pathway, that mediation pathway, is significant but was small. And there's some association. So when we unpicked exactly which aspects of disordered eating might be driving this effect, it might be binge eating specifically. And this is really interesting because there are interventions developed to target eating problems in autistic children-- so kind of autism specific interventions that already exist. And although it's kind of early days, one was deemed feasible with promising preliminary results. So this is highlighting putative targets for interventions that could be developed for autistic children or children with autistic traits specifically to target fussy eating early in development. And this, in turn, allows us to take a more preventative approach and maybe would mean that those children are less likely to go on to experience disordered eating in adolescence. So that was our first example. And now, we were really focusing on autism-specific characteristics. We could think about fussy eating as maybe being reflective of the kind of preference for sameness or restrictive patterns of behaviour associated with autism. And in the second piece of work, then we're thinking a bit more broadly in a way. So a lot of the work that's been discussed and actually this whole talk is about autism. But actually, autism and ADHD and other neurodevelopmental condition really often co-occur. So there are some estimates that around 40% of autistic people would also meet diagnostic criteria for ADHD, at least in adolescents in representative samples. So that's a huge number. And actually, a lot of the work has not thought about, yes, we're kind of beginning to delineate the associations between autism and disordered eating. But where does ADHD factor in this picture? Does it have the same sort of associations? Does it change the nature of associations between autism and disordered eating? Thinking clinically, would it mean that you would be tailoring your intervention slightly differently? I think that's the next frontier of work, which hasn't really been thought about that much. But it's definitely worth bearing in mind when we're thinking about autism and disordered eating, ADHD is something we do need to consider as well because of that really strong co-occurrence. So this work, we measured both autistic and ADHD traits. And by including them in the same statistical model, you basically are better able to understand what is specific to autism and what is specific to ADHD. Then here, we're focusing on anxiety as a mediator. And this was really inspired by the idea of this general distress experienced by and reported by autistic people with eating disorders. This is something that might be driving their patterns of disordered eating. And if this is the case, we would expect it to function as a mediator. So it would sit between autistic traits and disordered eating. Also, in this piece of work, we looked at two different types of eating behaviours. So we're moving again from a diagnosis of, for example, anorexia nervosa or binge eating disorders. And we're kind of focusing on how people report that they eat-- the patterns of behaviour involved in eating, how they eat food, the decisions they make. And so here, some factor analysis-- we did suggested two main groupings of eating behaviours. So one which is restrictive and emotional-- so this is restricting what you eat, choosing to eat less, or eating in response to emotions experience. And then we have externally driven eating. And this is, for example, maybe being more responsive to things in your environment as you eat. You see something really delicious. And you immediately want to go and eat that thing even if you're not hungry, for example. So we can think of these behaviours as maybe being earlier on the pathway or involved in the emergence of eating disorders but are not clinical diagnoses themselves but are very helpful in trying to understand and unpick these pathways to what could end up as an eating disorder being seen in the clinical setting. So we use the same data set. Again, we're measuring slightly different things here. So at age seven, we have autistic traits. But now, we're also measuring the two main characteristics of ADHD, inattention and hyperactivity and impulsivity. Then at age 10, we have measurement of anxiety from a kind of broader psychiatric assessment that was conducted called the Daubert. And then at age 13, we have self-reported eating behaviours, which are the two categories I mentioned of emotional and restrained eating or externally driven eating. And again, we have quite a large sample because this is a population representative cohort, where we have about 7,500 people who have complete measurements on either autistic traits or fussy-- sorry, not fussy eating. That should say ADHD. And what did we find? So again, we're setting up this model to reflect the longitudinal nature and nature of this data. And so we have over here our traits of neurodevelopmental conditions, key covariates. Then in the middle, we have anxiety measured in mid-childhood. And then we have our eating behaviours measured in adolescence on this hand side. And so the purple pathways are relating autistic traits to our outcomes. So we see an association between autistic traits and anxiety, which is really not surprising given that we know the risks for mental health problems is higher in people with an autism diagnosis or autistic traits. And we also see pathways from autistic traits to both types of eating behaviours. So higher autistic traits more likely to report emotional restraint eating but also externally driven patterns of eating. Then for hyperactivity and impulsivity, we actually only saw a pathway to anxiety here. So this kind of facet of ADHD wasn't in this analysis related to our eating behaviour outcomes. And then for inattention, we, again, see a pathway to anxiety and then just one pathway to externally driven eating. So higher levels of inattention and childhood was more likely to report externally driven patterns of eating in adolescents. So taking that first step, you can see there's some overlap. So both autistic traits and inattention are both linked to externally driven eating but also some specificity. So it's just autistic traits that's associated with emotional and restrained eating. And this is, obviously, really relevant for thinking about all the pieces of work that Professor Mandy outlined in the first part of the talk about the overlap between autism and anorexia nervosa or restrictive eating behaviours. Then the next stage was to look at anxiety as our mediator. So does this explain some of these patterns affecting the way we think it does? And so just looking at the path between autistic traits and emotional and restrained eating, we did find a significant indirect effect of anxiety. So that is telling us that some of this path from autistic traits to emotional and restrained eating can be explained by anxiety. Again, it's not a huge effect. I think it was around 15%, which is higher than our fussy eating measurement before. But obviously, it doesn't tell us everything. It cannot explain all of the association between autistic traits and emotional restrained eating. But it's telling us that anxiety is important in understanding this association here. So just to sum up that other piece of work-- so autistic and ADHD traits in childhood may share some eating behaviour phenotypes in adolescents, whereas others seem to be more specific to autism. And emotional difficulties, here as measured through anxiety, may underpin this association and could be a really important potential target for intervention. And I was just looking at recent papers out on this subject. And actually, I think this one chimes quite nicely, which came out this year, which is a qualitative piece of work. But one of the themes that come through is the idea of regulating with food. So maybe certain patterns of eating-- which I also think Professor Mandy's work spoke to kind of a coping strategy for very difficult emotional experiences. And so if we could address some of these emotional difficulties and think about how these might be driven by maybe a poor person environment fit and how we could improve that, then, in turn, that could be a really important preventative approach so that autistic people or neurodivergent people maybe don't go on to experience eating disorders later on in the lifespan. So I've kind of given you a flavour of two pieces of empirical research where what we were really trying to do is test hypotheses which came from speaking to people with lived experience. And I think there's much more work to do there, either using existing data sets or where we don't have measurement of the things we think are really key, going out and collecting some new data ourselves to test them. But if we just take a step back from all the statistics and models and things like that and think about who decides on research priorities-- so often, what is researched is decided by academics. And this is definitely true in the field of mental health research and autism research, although I think, thankfully, this is really shifting, as people from the autistic community have really pushed for their voices to be heard more and involve more in these key decisions about what we focus on, what we do research on. And this is definitely true for clinically relevant questions like this. So understanding community and stakeholder perspectives is really key to developing research that will improve the lives of autistic people who experience disordered eating. And at the end of the day, I think that's what we're all hoping for here. So this is just an example of a kind of priority setting approach and a project that we undertook last year to understand the priorities of neurodivergent people with lived experience of disordered eating. And I hope by talking through, it will give you a sense of how these projects work but also really guide future research to make sure that it is in line with the priorities of people with lived experience. So what we did is we broadly followed the principles laid out by the James Lind Alliance, which is a nonprofit association, which has loads of really helpful information about how to conduct priority setting partnerships such as this. So first, we did an online scoping survey with neurodivergent participants with lived experience of disordered eating to just broadly ask, what are the sorts of priorities that research in this field should be focusing on? And in the little pop-out box, I've just given some data about the type of people who took part. And this is mostly to highlight. We cannot consider this a representative sample. The sample was primarily female, primarily people who reported white ethnicity and primarily people who had an undergraduate degree or higher. So all of this should be taken with a pinch of salt. I'm not saying this is not helpful. And I really think these exercises are. But we cannot say this is representative of the general population view. All participants confirm they either identified as neurodivergent or had a specific diagnosis and had experienced disordered eating. And here, we wanted the survey to be quite broad and very inclusive so people didn't have to have a diagnosis to take part. But we just asked them which diagnosis they kind of felt both best fitted their experiences or if they'd actually received the diagnosis clinically. So 62% sample endorsed the diagnosis of autism. I think the next most frequent diagnosis was ADHD. And there was a lot of co-occurrence. So we asked them to give us some research priorities. And then we read through all of the responses and reached a consensus on 15 topics that kind of frequently mentioned or broadly captured what people talked about. Then we rewrote these topics as empirical research questions and took them through to an in-depth workshop. So we kind of started very broad. This was an online survey. And now, we're going through to an in-depth workshop, where we're going to really talk about things in greater detail. And so here, we had 14 people take part in a two-hour workshop, which was facilitated by a company that had lots of experience running these sort of public engagement and community consultation events. Again, we have a broader split on gender lines. But majority endorsed white ethnicity and had an undergraduate degree or higher. And 86% endorsed the diagnosis of autism. But everyone confirmed they identified as neurodivergent and had experienced disordered eating. And so we asked them to do some different exercises, different ways to think about the topics, and then narrow it down to a list of top 10 topics which were ordered by priority. And we summarised these in two themes. So just to give you a sense of how it worked, because I think sometimes it's quite helpful to see things, is we started off by getting people to think about, who would benefit from research, looking at the overlap of neurodivergence and eating disorders? Who really needs to know about the research? What are the different communities? Then we ask people things like, which topic is most urgent? Different topics, how would they impact different societal groups? just to really get people thinking about things in a kind of 360 degree way before we ask them to rank in different priorities. Really, really fascinating discussions. And so many things came up in these groups, which I think as a researcher, I really hadn't thought about or considered. So it's really kind of underscores the value of talking to people with lived experience to guide the research and make sure that you're measuring and thinking about things that's actually relevant for people who have this lived experience of things that you're researching. So this is just another visual example of how we ask people to rank things. So these are our 15 questions. The text is probably a bit small on the screen here. But it's all outlined in the manuscript. So we asked people which would most benefit. And we have our stakeholders along the bottom. So for example, people with lived experience, the general public, service providers, people in power-- these were all chosen by people in the workshop as the groups they saw as really key to benefiting from this research. And then we have, which has the most potential to? And again, change in societal attitudes, improve health care practitioners' ability to act, improve people with disordered eating's ability to act, and generally improve the health of neurodivergent individuals. And then we just ask them to rank things as most urgent. And the darker colours here reflect things that were ranked more highly. So you're already kind of getting a sense of some of the things, topics which are being picked out as more important than others. But I think this is probably an easy way to look at things. So after we boiled all this information down, then we have our top 10 rankings. And I think a helpful way to think about things is maybe kind of clustering them into two themes. So the first theme is improving outcomes. And this came across so clearly in that people really felt-- and I think this mirrors some of the things discussed in part one of these talks-- that current approaches for disordered eating were just not serving neurodivergent people. So the number one question, which was ranked as the highest priority, is, how could treatment for disordered eating be improved for neurodivergent individuals? Then the next kind of one from that theme is, what is the impact of neurodiversity training or knowledge in eating disorder clinical services? So again, in part one of the talk, I think there was some discussion of how maybe a lack of knowledge or understanding of how autism may present or autistic characteristics can actually hinder the provision of eating disorder treatment. And so this question was more speaking to, if we did train people who are going to frequently come into contact with people with eating disorders and probably people who are neurodivergent, would that be helpful? Would that improve clinical services? Would that mean better outcomes for autistic and neurodivergent people? That's a question that, I think, is really important but has not yet been answered. The next one, which I thought was really interesting, is, which treatment interventions for disordered eating are actively unhelpful for neurodivergent individuals? So I think there's this kind of underlying assumption sometimes in mental health research that maybe psychological treatments don't come with unhelpful or unintended or negative side effects, that that's something solely relevant to pharmacological interventions. But actually, people were telling us in these groups that sometimes things they'd been offered had been very unhelpful-- had not worked but actually made people feel worse. So that's something that neurodivergent people with lived experience really wanted to know more about. What could be identified as unhelpful? Because then we need to not offer this to neurodivergent people. And then the last one in this theme is, would better information and education reduce risk for neurodivergent people? So people in the groups talked about, had they known that there was an increased risk for eating disorders in autistic people, this might have helped them to understand some of their behaviours or maybe put things in place so that it wouldn't have moved on to clinically relevant disordered eating, which was having a really negative impact on their day to day life. So people really wanted more information that they themselves could read and could take on to understand their own experiences. And then the second theme that came across was about identifying causal mechanisms. So in addition to proving outcomes, I think, there really is value. And people with lived experience think it's important to do this work, understanding some of the autism or neurodivergent-specific factors, which might increased risk for disordered eating. So this was the number two ranking. But in this theme, the first one, what are other factors that may increase risk of disordered eating in neurodivergent people? And the second question is kind of similar. Is there a difference in the underlying reasons for disordered eating? People ask specifically about understanding the link between certain ADHD characteristics, sensory sensitivities, and different thinking styles-- for example, black and white thinking. So I think this is a really helpful call for researchers to go out and begin to answer some of these questions that people with lived experience really see as important things for us to focus on. So just to summarise the things I've talked about today-- so at the moment, I think we really do lack understanding of some of the autism-specific causal and maintaining factors, which will impact upon how we offer different interventions, how we develop them, and how they're rolled out in the community. So understanding some of these autism-specific factors is really key to providing targeted and therefore probably more effective interventions. So some work has been done using longitudinal data, which really allows you to begin to unpick some of these pathways and highlight potential intervention targets. And the work I've done and I've talked about today has suggested both fussy eating early in childhood and maybe anxiety in mid-childhood could be really important targets for support. I will say this work-- what I would like to do and what I think we should move towards is modelling person environment fit better. This is mostly focused on individual characteristics. But what are the aspects of the environment that are causing this distress in the first place? And how can we improve things to lower risk for disordered eating in autistic people? And then just to end on, any future research should be really in line with what people with lived experience want. And so I think here, hopefully, we've given some ideas and some suggestion for what people could focus on in the field of autism and eating disorders to make sure that their work is very much in line with community priorities. Thanks. [MUSIC PLAYING]