Paper Session I

Paper Session I
Friday 2:00 PM - 3:30 PM

Topic: Treatment of Eating Disorders (Adult) I
Club A, first floor
Kate Tchanturia, PhD, FAED & Tracey Wade, PhD, FAED

1.1: Trajectory of Weight Change Over the First 12 Sessions of Outpatient Therapy for Anorexia Nervosa and Relationship to Outcome

Tracey Wade, PhD, Flinders University, Adelaide, South Australia, Karina Allen, PhD, University of Western Australia, Perth, Ross Crosby, PhD, University of North Dakota School of Medicine and Health Sciences, Fargo, North Dakota, Christopher Fairburn, MRCPsych, Oxford University, Oxford , United Kingdom, Anthea Fursland, PhD, Centre for Clinical Interventions, Perth, Western Australia, Phillipa Hay, DPhil; FAED, University of Western Sydney, Sydney, New South Wales, Gini McIntosh PhD, University of Otago, Christchurch, New Zealand, Stephen Touyz, PhD, University of Sydney, Sydney, New South Wales, Ulrike Schmidt, PhD, King's College London, UK, Janet Treasure, PhD; FAED, King's College London, London, London, Susan Byrne, PhD, DPhil, University of Western Australia, Perth, Western Australia

The purpose of the study was to identify latent classes of trajectory of change in body mass index (BMI) between assessment and the twelfth session of outpatient treatment for adult anorexia nervosa. The relationship between class and outcome was then examined, and predictors of class membership were also examined. Participants were 120 people with anorexia nervosa who were randomised to one of three outpatient therapies. Four latent classes were identified; two classes had a baseline BMI around 17.5, but one group (N=19, 16%) showed a significantly greater increase in BMI over the first 12 sessions of therapy than the other (N=39, 32%). The third and fourth classes had a baseline BMI of 16.34 and 15.26 respectively, and neither gained weight over the first 12 sessions. With respect to outcomes, there was no differences between the classes in terms of therapy completion, but the class who had the greatest initial increase in BMI had significantly higher levels of remission (i.e., attaining a global Eating Disorder Examination Questionnaire score within one standard deviation of the community mean, a BMI > 18.5, and absence of binge eating/purging) at end of treatment and at 12-month follow-up than the other three classes. Treatment condition did not predict class membership. The only variables to predict class membership were approach coping style and lowest weight, where the class with the best outcome has higher levels of approach coping and a higher lowest weight. The extremely low levels of remission in the third and fourth class poses the question of whether outpatient therapies for anorexia nervosa should only include people with a BMI > 16.5.

1.2: Relapse Rates and Predictors of Relapse at One Year Follow-up After Outpatient Psychological Treatment for Anorexia Nervosa: Results from a Randomized Controlled Trial.

Anthea Fursland, PhD, Centre for Clinical Interventions, Perth, Western Australia, Tracey Wade, PhD, School of Psychology, Flinders University, Adelaide, New South Wales, Phillipa Hay, DPhil; FAED, School of Medicine & Centre for Health Research, Western Sydney University, Sydney, New South Wales, Stephen Touyz, PhD, School of Psychology, Sydney University, Sydney , New South Wales, Christopher Fairburn, MD; PhD, Department of Psychiatry, Oxford University, Oxford, Oxfordshire, Janet Treasure, PhD; FAED, Department of Psychological Medicine, Kings College London, London, London, Ulrike Schmidt PhD, Department of Psychological Medicine, Kings College London, London, London, Virginia McIntosh, PhD, Department of Psychological Medicine, University of Otago, Christchurch, New Zealand, Karina Allen, MPsych; PhD, Maudsley Hospital, South London and Maudsley NHS Foundation Trust, London, London, Ross Crosby, PhD, Psychiatry and Behavioral Sciences, University of North Dakota, Fargo, North Dakota, Susan Byrne, MPsych; DPhil; PhD, School of Psychology, University of Western Australia, Perth, Western Australia

Relapse rates following treatment for anorexia nervosa (AN) are reported as being notoriously high. This study examined relapse rates, and predicators of relapse, at 1 year post-treatment for adults (N=120) who participated in an Australian multicentre randomised controlled trial (RCT) comparing three outpatient treatments for AN (Specialist Supportive Clinical Management [SSCM], Maudsley Anorexia Nervosa Treatment for Adults [MANTRA] and Enhanced Cognitive Behaviour Therapy [CBT-E]). Treatment involved 25-40 individual sessions over a 10 month period. At the end of treatment full remission was defined as (1) having a BMI > 18.5 kg/m2; and (2) having a Global Eating Disorder Examination (EDE) score within one standard deviation of community norms. Participants were assessed, using a broad range of measures, at pre, mid and post treatment and at 6 and 12 month follow-up. At the end of treatment, 22% of participants (26/120) were in full remission (SSCM 9/40 [22.5%]; MANTRA 9/41 [22%]; CBT-E 8/38 [21%]). Of these participants, 20 (77%) were still in full remission at 1 year follow-up (SSCM 9/9 [100%], MANTRA 5/9 [55.6%], CBT-E 6/8 [75%]) and the remaining six were in partial remission (MANTRA 4/9; CBT-E 2/8). Another six participants (SSCM=1, MANTRA =1, CBT-E=4) who had not achieved full remission by end of treatment, continued to improve over time and did so by 12 month follow-up. Binary logistic regression analyses showed that higher BMI and lower EDE scores (albeit still not reaching normal levels) at end of treatment significantly predicted this positive trajectory. No significant predictors of relapse were identified. These results suggest that, with the treatments provided in this RCT, relapse rates for AN were remarkably low compared to previous studies. A good outcome (with regard to weight regain and normalization of core eating disorder psychopathology) by end of treatment indicates that these improvements are highly likely to be maintained.

1.3: An Examination of the Interpersonal Model of Binge Eating over the Course of Treatment

Anna Karam, MA, Washington University in St. Louis, St. Louis, Missouri, Dawn M. Eichen, PhD, University of California San Diego, San Diego, California, Ellen Fitzsimmons-Craft, PhD, Washington University in St. Louis, St. Louis, Missouri, Rick Stein, PhD, Washington University in St. Louis, St. Louis , Missouri, Robinson Welch, PhD, Washington University in St. Louis, St. Louis, Missouri, Denise E. Wilfley, PhD, Washington University in St. Louis, St. Louis, Missouri

This study evaluated the interpersonal model of binge eating and related eating disorder (ED) psychopathology over the course of two treatments (Interpersonal Psychotherapy and Cognitive Behavioral Therapy) in participants with binge eating disorder (N=159). Investigating treatment mediators is needed to better understand why and how evidence-based treatments are efficacious. The interpersonal model of binge eating posits that interpersonal problems increase negative affect, which triggers binge eating and ED psychopathology. Preliminary support for the model exists; however, the model has yet to be tested over the course of treatment. Thus, this study aims to fill an important gap and details how treatment achieves its effects. Change scores from pre-to post-treatment were calculated for the following variables: interpersonal problems (independent variable), negative affect (mediator), and ED psychopathology (dependent variables measured by the Eating Disorder Examination; objective bulimic episode (OBE), subjective bulimic episodes (SBE), eating restraint, eating concern, shape concern, weight concern, global). Bootstrapped mediation analyses, controlling for treatment group, were used and biased-corrected 95% confidence intervals (CIs) were computed. All variables decreased from pre- to post-treatment (ps<.05). In line with the model, over the course of treatment reductions in negative affect mediated the relation between decreases in interpersonal problems and reductions in OBEs (95% CI [0.41, 5.80]), shape concern (95% CI [0.11, 0.67]), weight concern (95% CI [0.03, 0.66]), and global (95% CI [0.07, 0.41]), regardless of treatment. Findings support that psychotherapy produces changes in the mechanisms of the interpersonal model of binge eating as theoretically proposed. Results preliminarily demonstrate how individuals improve through psychotherapy, which could lead to enhanced ED treatment by strengthening the identified active therapeutic components.

1.4: Guided Self-Help For Binge Eating: A Randomized Controlled Trial Comparing Two Forms of Support

Paul Jenkins, BSc; CPsychol; DClinPsy, Oxford Health NHS Foundation Trust, Oxford, Oxfordshire, Amy Luck, DClinPsy, Oxford Health NHS Foundation Trust, Oxford, Oxfordshire, Christopher Fairburn, DM, FMedSci, FRCPsych, University of Oxford, Oxford, Oxfordshire

Guided self-help (GSH) is often recommended as a first step for individuals presenting for treatment of bulimia nervosa (BN) or binge-eating disorder (BED). The optimal means of delivering guidance is, however, uncertain. Conventionally guidance is provided face-to-face in a limited number of brief sessions given by a non-specialist practitioner. A more scalable alternative is to provide guidance via asynchronous emails. The purpose of this study was to compare GSH delivered in person or via email with a waiting list control condition. The study recruited patients from routine referrals to two UK National Health Service eating disorder clinics. 180 adults (92.8% female, 81.7% White – British) were randomly allocated to a waiting list condition, face-to-face GSH, or email GSH (60 per group). The diagnostic distribution was as follows: BN 106 (58.9%); BED 39 (21.7%); and OSFED 35 (19.3%). Eating disorder psychopathology, psychosocial impairment, general psychological distress, and self-esteem were assessed at baseline and post-intervention. Both forms of GSH were associated with clinically significant change whereas there was little change in the waiting list condition. The attrition rate was higher in the email condition than in the face-to-face condition. These finding provide further support for GSH as a treatment for BN and BED. The study is one of the first to directly compare these two methods of delivering support in GSH although further work is needed to explore the optimal method for delivering guidance. This needs to take account of scalability and cost-effectiveness.

1.5: Can Weight Regain Account for Improvements in Core Eating Disorder Psychopathology and Comorbid Psychopathology in Anorexia Nervosa? Association Between BMI Change and Change in Eating Disorder Specific and General Psychopathology in the SWAN Study.

1.5: Can Weight Regain Account for Improvements in Core Eating Disorder Psychopathology and Comorbid Psychopathology in Anorexia Nervosa? Association Between BMI Change and Change in Eating Disorder Specific and General Psychopathology in the SWAN Study.

The aim of this study was to examine the relationship between weight regain and change in eating disorder psychopathology and broader psychopathology in patients receiving treatment for anorexia nervosa (AN). We used data from an Australian multicentre randomised controlled trial (the SWAN Study) in which 120 adults with AN were randomised to one of three outpatient psychological treatments (Specialist Supportive Clinical Management, Maudsley Anorexia Nervosa Treatment for Adults and Enhanced Cognitive Behaviour Therapy). Body Mass Index (BMI), core eating disorder psychopathology (measured using the Global Eating Disorder Examination [EDE] score) and levels of depression, anxiety, stress, perfectionism, obsessionality, interpersonal difficulties, self-esteem and clinical impairment were measured at pre-treatment, mid-treatment, post-treatment and at six and twelve month follow-up. Linear mixed models were used to examine whether improvement in BMI was associated with corresponding improvements in eating disorder psychopathology and the other more general psychopathology. The results showed, firstly, that weight regain was not significantly associated with an improvement in EDE scores. Secondly, while weight gain was accompanied by improvements on some of the secondary measures (clinical impairment, depression, self-esteem, obsessionality and interpersonal difficulties) it was not associated with improvements in anxiety, stress or perfectionism. Improvement in Global EDE score over time, however, was associated with improvement on these variables. Treatment condition did not affect any of these relationships. These results underscore the need for both weight regain and improvement in eating disorder psychopathology in recovery from AN.

1.6: A Comprehensive Nutrition Education Guide for Eating Disorder Clinicians with their Patients

Caitlin McMaster, BSc, Royal Prince Alfred Hospital, Sydney, New South Wales, Susan Hart, BSc; MSc; PhD, Royal Prince Alfred Hospital, Sydney, New South Wales

Nutrition intervention is a core element eating disorder (ED) treatment. However, there is a lack of evidence based, ED specific material to support clinicians in the nutrition education process. Dietitians at the Peter Beumont Day Program developed the Eating Disorder Healthy Eating (EDHE) Guide for adult ED patients consisting of EDHE Pyramid, Dietary Guidelines for ED Patients and patient meal plans To evaluate patients’ impressions of the EDHE Guide, 20 adult female ED patients (mean age=26.6 years) attending Royal Prince Alfred Hospital for treatment as a day program patient (n=8) or an outpatient (n=12) for a range of ED diagnoses (mean BMI = 23.7) were given 10 minute oral presentations of the EDHE Pyramid and a non-ED specific nutrition tool, the Australian Guide to Healthy Eating. Guided discussion sessions were used to examine which of the nutrition tools best addressed common ED nutrition beliefs. Qualitative data was assessed using thematic analysis. To assess nutritional adequacy of constructed meal plans, seven days of weight gain and weight maintenance meal plans were modelled using nutritional analysis programs. 19 out of 20 patients reported the EDHE Pyramid to be more helpful for working through an ED than the AGHE. Key reasons included relevance to treatment; reassurance; and clearer explanation about food groups. Modelled meal plans met Australian macronutrient distribution ranges for carbohydrate, protein and fat to reduce chronic disease risk and Australian Nutrient Reference Values for fibre, calcium, iron, zinc, vitamin D, sodium, potassium and fluid intake with the exception of vegan meal plans which were not adequate in vitamin D. This study suggests that a specialist nutrition education tool is more effective than a generalist nutrition education tool at addressing common ED nutrition beliefs and attitudes. Additionally, the EDHE Guide includes nutritionally complete meal plans suitable for patients with a range of nutritional requirements.

Topic: Child and Adolescence I
Panorama, second floor
Kamryn Eddy, PhD, FAED & Nadia Micali, MD, PhD, FAED

2.1: Dynamic Interplay Among Eating Disorder Symptoms in a Transdiagnostic Sample of Treatment-Seeking Youth: Further Evidence for the Importance of Shape- and Weight-Related Concerns

Andrea B. Goldschmidt, PhD, Brown Medical School/The Miriam Hospital, Providence, Rhode Island, Ross D. Crosby, PhD, Neuropsychiatric Research Institute, Fargo, North Dakota, Li Cao, MS, Neuropsychiatric Research Institute, Fargo, North Dakota, Markus Moessner, DiplPsych, University Hospital Heidelberg, Heidelberg , Heidelberg, Kelsie T. Forbush, PhD, University of Kansas, Lawrence, Kansas, Erin C. Accurso, PhD, University of California, San Francisco, San Francisco, California, Daniel Le Grange PhD, University of California, San Francisco, San Francisco, California

Classifying eating disorders (EDs) in children and adolescents is challenging. In light of developmental considerations and high rates of diagnostic crossover, it is possible that the diagnostic boundaries between specific types of EDs are arbitrary. Understanding the dynamic interplay among ED symptoms within and between diagnostic categories may be useful for clarifying which core ED symptoms contribute to, and maintain, ED psychopathology in youth. Network analysis is an innovative statistical approach that has recently been used to understand which symptoms contribute most strongly to psychopathological disorders by identifying interrelationships among symptoms. We utilized network analysis to investigate associations among ED symptoms in a transdiagnostic sample of 629 treatment-seeking children and adolescents (90.3% female), aged 6-18 years (M age=15.4±2.2 years). An undirected, weighted network of ED symptoms was created using behavioral and attitudinal items from the Eating Disorder Examination (EDE) and the R package qgraph. Symptoms that reflected dissatisfaction with shape and weight were most strongly associated with other symptoms in the network. Empty stomach emerged as an important “bridge” between symptoms, and had close connections to all other ED symptoms in the network. Importantly, binge eating and compensatory behaviors were strongly connected to one anbut not with other symptoms in the network, which is consistent with previous research in adults. Taken together, results suggested that among children and adolescents presenting for ED treatment, shape- and weight-related concerns play a key role in the psychopathology of EDs, which supports cognitive-behavioral theories of ED onset and maintenance. Clinical interventions should target these symptoms early in treatment so as to achieve the greatest impact on other ED features.

2.2: “I put my trust in them.” How Adolescents and their Parents Respond and Change During Family Based Treatment - A Grounded Theory

Andrew Wallis, MFT, Eating Disorder Service, The Children's Hospital, Westmead, Westmead, New South Wales, Paul Rhodes, MSW, School of Psychology, University of Sydney, Sydney, New South Wales, Sloane Madden, PhD; FAED, Eating Disorder Service, The Children's Hospital, Westmead, Westmead, New South Wales, Jane Miskovic-Wheatley, DPsych, Eating Disorder Service, The Children's Hospital, Westmead, Westmead , New South Wales, Stephen Touyz, PhD; FAED, School of Psychology, University of Sydney, Westmead, New South Wales

The aim of this study was to understand the impact of Family Based Treatment (FBT) on interpersonal and intrapersonal relationships, and investigate the process of relational change during FBT. Participants were 16 adolescents and their parents who had participated in a larger randomised control trial where manualised FBT was provided after medical stabilisation in a specialist inpatient eating disorder program. All participants had DSMIV diagnosed Anorexia Nervosa. This qualitative study used constructivist grounded theory to analyse data. Adolescents and their parents were interviewed separately at least 6 months after their 20th session of FBT. Results indicated that interpersonal and intrapersonal change occurred for both adolescents and their parents across the course of treatment. The core theme that emerged from the data was relational containment indicating that a combination of treatment system therapy process and FBT tenets contained the parents, allowing them to interact with their adolescent in a more attuned and confident way, leading the adolescents to ultimately feel more secure due to increased trust and communication. Both adolescents and parents described improved interpersonal relationships after treatment for the family as a whole with adolescents also describing an improved sense of self. These results provide an important framework for understanding some of the relational processes that may underpin FBT working effectively. The findings of this study are synonymous with how parents provide a secure base in any time of crisis for their child, and indicate that FBT's initial focus and treatment sequence can not only restore physical health but reestablish and improve normal developmental processes after the crisis of being unwell with Anorexia Nervosa. Future directions for research are discussed and implications for treatment development suggested.

2.3: The Impact of an Intensive FBT Intervention on Caregiver Variables in Medically Hospitalized Youth with Restrictive Eating Disorders

Abigail Matthews, PhD, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, Claire Peterson, PhD, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, Laurie Mitan, MD, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio

The efficacy of family-based treatment (FBT) interventions for medically hospitalized youth with restrictive eating disorders (ED) has not been evaluated. This pilot study examined the impact of a brief FBT protocol on variables associated with a caregiver’s ability to implement FBT. Furthermore, it presented a unique opportunity to assess the characteristics of newly diagnosed youth requiring medical stabilization secondary to their ED, as well as 3- and 6-month follow-up weight data. The goals of this study were to: 1) assess parental self-efficacy and understanding about ED, pre- and post- an intensive FBT intervention; 2) examine the characteristics of medically hospitalized youth with newly diagnosed restrictive ED; and 3) assess 3- and 6-month weight gain following discharge. Forty-three adolescents with newly diagnosed AN (65.1%), atypical AN (30.2%), and ARFID (4.7%), and their primary caregivers, completed a 4-session FBT intervention. Mean admitting BMI was 16.94 (SD=1.7) and low heart rate was 38.6 (SD=6.8). Average length of stay was 13.22 days (SD = 7.4), with 17.1% requiring an admission greater than 20 days. Following our intervention, parental understanding about ED and caregiver self-efficacy related to caring for their child improved significantly (t=-7.031, p<.001; t=-5.95, p<.001). When examining pre-intervention parental self-efficacy, admitting BMI, age, EDE-Q Global scores, and parental understanding about the illness, parental understanding of the illness was the only predictor to contribute significant, unique variance to post-intervention parental self-efficacy (R2 = .31, p < .05; β = .39, p < .05). Mean BMI significantly increased from baseline to 3-month follow up (from m = 16.88 to m = 19.03; t = 8.82, p < .001) and from 3- to 6-month follow up (from m = 19.12 to m = 20.06; t = 4.36, p < .001). Moreover, at 6-month follow-up, 36.7% of our sample had been readmitted for a medical admission or had received inpatient or partial psychiatric hospitalization for ED symptoms. Preliminary findings from this ongoing study have implications for the use of an intensive FBT intervention during medical hospitalizations for restrictive ED and provide valuable information about presenting characteristics and trajectories for newly diagnosed youth with medical complications.

2.4: The Relative Importance of Peer Network and School Environments for Weight Perception and Dieting Behaviors in Youth

Tracy Richmond, MD; MPH, Boston Children's Hospital, Boston, Massachusetts, Carly Milliren, MPH, Boston Children's Hospital, Boston, Massachusetts, Kendrin Sonneville, PsyD; RD, University of Michigan School of Public Health, Ann Arbor, Michigan, Clare Rosenfeld Evans, ScD, University of Oregon, Eugene , Oregon

We aimed to determine the relative importance longitudinally of peer networks and the school environment on weight perception and dieting behaviors in young adulthood. Using data from Waves I and III of the National Longitudinal Study of Adolescent to Adult Health (Add Health), our outcome was reporting dieting to lose weight or keep from gaining weight in the past 7 days and accuracy of weight perception (self-reported weight perception compared to objective weight categories) at Wave III. We applied a multilevel modularity maximization algorithm for social network community detectionin sociocentric networks constructed from self-reported friendship nominations resulting in 16,735 participants in 1,503 distinct network communities nested within 125 schools. We used multilevel logistic regression to examine clustering of Wave III weight perception and dieting behavior by school and network communities in null models (no predictors) as well as models adjusting for individual demographics. The average age in the sample at Wave I was 16.2 (SD 1.7) years; 51% of participants were female, 52% non-Hispanic White, and 25% had overweight/obesity. Thirteen percent of participants reported dieting behaviors in the past week; 48% were inaccurate in their weight perception with 42% underestimating their weight. In null models, the peer network contributed more than four times as much to the variance in Wave III dieting as the school environment (σ2sch=0.10; σ2network=0.45); similar patterns were seen in weight perception (σ2sch=0.01; σ2network=0.05). This pattern held in models adjusting for age, sex, and race/ethnicity(dieting: σ2sch=0.07; σ2network=0.35; weight perception: σ2sch=0.01; σ2network=0.05 ). Peer networks had large and sustained contributions to variability in weight perception and dieting behaviors; peer networks contributed substantially more than school environments. Interventions targeted at preventing weight misperception and adoption of dieting behaviors should include peer-focused components.

2.5: Do Disordered Eating Behaviours in Girls Vary with Different School Characteristics? A UK Cohort Study

Helen Bould, BA; MA; MRCPsych; BM BCh, University of Oxford, Oxford, Oxfordshire, Bianca De Stavola, PhD, London School of Hygiene and Tropical Medicine, London, UK, Glyn Lewis, MRCPsych; PhD, UCL, London, UK, Nadia Micali, MRCPsych; PhD, UCL, London, UK

We hypothesise that rates of disordered eating behaviours will vary between schools in a large UK birth cohort. Previous work on eating disorders, disordered eating behaviours and schools has produced inconsistent results. A population study from the US found that differences in disordered eating behaviours between schools do not withstand adjustment for individual characteristics. In contrast, our previous work using Swedish record-linkage data found that rates of diagnosed eating disorders vary between schools, with higher proportions of girls and higher proportions of highly educated parents within a school being associated with greater numbers of diagnosed eating disorders. We used data from the Avon Longitudinal Study of Parents and Children (ALSPAC) to test the hypothesis that rates of self- and parent-reported disordered eating behaviours, and body dissatisfaction would cluster by school. We had complete data on disordered eating behaviours, school attended, and explanatory variables on 2146 girls at 263 schools at 14 years old; and 1769 girls in 273 schools at 16 years old. We used multilevel modelling to assess whether rates varied between schools and logistic regression to investigate the association between school characteristics and rates of disordered eating behaviours. At age 14, there was no evidence for clustering of body dissatisfaction by school, or for school characteristics being associated with body dissatisfaction. At age 16, again there was no evidence for clustering of disordered eating behaviours, but all girls schools had higher rates of disordered eating behaviours, whilst schools with higher academic results had lower rates of self-reported fasting and disordered eating behaviours. We found no evidence to support our hypothesis that levels of eating disorder symptoms vary between individual schools in England. However, there was evidence that rates of disordered eating behaviours may be higher in all girls schools, and in schools with lower academic performance.

2.6: Leave Me Alone, Help Me Recover: Adolescent Impressions of Family Therapy

Erin Parks, PhD, UC San Diego, San Diego, California, Anne Cusack, PsyD, UC San Diego, San Diego, California

While Family Based Therapy (FBT) remains the gold-standard treatment for adolescent eating disorders, one barrier to parents initiating FBT is their child’s objection to parental involvement and the parents’ subsequent fear of harming parent-child relationships. This mixed-method study examined adolescents’ impressions of parental involvement and parent-child relationships throughout FBT. Participants were 29 adolescents who had participated in family-based partial hospitalization (PHP) and intensive outpatient (IOP) eating disorder treatment. All participants completed a Likert scale survey and answered open-ended writing prompts on their impressions of family involvement in their eating disorder treatment. Analyses demonstrated that adolescents have significantly (p < .001) more positive impressions of family involvement post-treatment when compared to pre-treatment. This small, preliminary study showed that although adolescents are typically resistant to family involvement at the beginning of treatment, they generally view family involvement more favorably at treatment completion.

Topic: Neuroscience I
Club E, first floor
Laura Berner, PhD & Stefan Ehrlich, MD, PhD

3.1: Novel and Ultra-Rare Damaging Variants in Neuropeptide Signalling are Associated with Disordered Eating Behaviors

Michael Lutter, MD; PhD, Eating Recovery Center, Plano, Texas

Eating disorders develop through a combination of genetic vulnerability and environmental stress, however the genetic basis of this risk is unknown. To understand the genetic basis of this risk, we performed whole exome sequencing on 95 unrelated individuals with eating disorders (39 restricted-eating and 56 binge-eating) to identify novel damaging variants. Candidate genes with an excessive burden of predicted damaging variants were then prioritized based upon an unbiased, data-driven bioinformatic analysis. One top candidate pathway was empirically tested for therapeutic potential in a mouse model of binge-like eating. An excessive burden of novel damaging variants was identified in 157 genes in the restricted-eating group and 243 genes in the binge-eating group. This list is significantly enriched (OR=4.6, p<0.0001) for genes involved in neuropeptide/neurotrophic pathways implicated in appetite regulation, including neurotensin-, glucagon-like peptide 1- and BDNF-signaling. Administration of the glucagon-like peptide 1 receptor agonist exendin-4 significantly reduced food intake in a mouse model of ‘binge-like’ eating. These findings implicate ultra-rare and novel damaging variants in neuropeptide/ neurotropic factor signaling pathways in the development of eating disorder behaviors and identify glucagon-like peptide 1-receptor agonists as a potential treatment for binge eating.

3.2: White Matter Alterations in Eating Disorders: Evidence from Voxel- Based Meta-Analysis and Systematic Review

Manuela Barona, BSc; MSc, University College London, UK

Identification of white matter differences in anorexia nervosa (AN) is a necessary step in order to better understand the underlying brain abnormalities in AN and target more effective treatments. However, little research has been done in this area and it has not yet yielded robust conclusions. The aim of this meta-analysis is to report white matter differences in women with AN compared to healthy controls. A comprehensive literature search was conducted in 2016 in order to identify studies comparing fractional anisotropy (FA) (the most widely used measure of white matter integrity) between patients with AN and controls. Results from the identified studies were combined to identify white matter differences using effect size-signed differential mapping (AES_SDM) meta-analysis in patients with AN (n = 138) compared to healthy controls (N = 141). A significant decrease in FA was identified in two posterior areas of the corpus callosum (CC) and one area in the pons. One area of increase FA was also found in the right cortico-spinal projections. Subgroup analyses revealed two specific areas in which we found decreased FA in adults with AN (right caudate nucleus and CC) as well as three specific areas in those with an active disorder (CC, pons, cortico-spinal projections and right anterior-thalamic projections). The areas identified have been found to be relevant for image perception and emotion and reward processing, thus might be playing an important role in the development and/or maintenance of AN.

3.3: Positron Emission Tomography Correlates of Outcome at 1 Year Follow up in Deep Brain Stimulation for Anorexia Nervosa

D Blake Woodside, MD, Toronto General Hospital, Toronto, Ontario, Nir Lipman, MD, Toronto Western Hospital, Toronto, Ontario, Eileen Lam, BSc, Toronto General Hospital, Toronto, Ontario, Andres Lozano, MD, Toronto Western Hospital, Toronto, Ontario

This presentation presents 1 year follow up data data on Positron Emission Tomography(PET) in a sample of 16 subjects who underwent Deep Brain Stimulation(DBS) for severe Anorexia Nervosa(AN). Clinical and psychometric data on this sample have been presented previously. Subjects enrolled in our DBS trail underwent metabolic brain imaging using FDG-PET at baseline, 6 and 12 months. Changes in cerebral glucose metabolism was compared from baseline to 6 to 12 months. Findings reported are significant at a t-threshold greater than 3.51(z>2.98, p<0.0029). Comparisons of 12 months to baseline showed decreases in metabolism in the superior frontal gyrus, left middle frontal gyrus, right anterior cingulate gyrus, subcallosal gyrus left caudate, bilateral putamen, bilateral thalamus (pulvinar), right medial globus pallidus and cerebellum (bilateral culmen, tuber and inferior semi-lunar lobule and left tonsil). Increases in metabolism were observed in posterior cortical regions including the right parahippocampal gyrus and middle temporal gyrus, and left inferior parietal lobule. The comparison of twelve months to six months showed greater decreases in metabolism over the course of DBS in left medial frontal gyrus, left anterior cingulate gyrus, and left globus pallidus. Greater increases in metabolism over time were observed in left superior temporal gyrus, and left inferior parietal lobule. Our current results extend our original findings. Activity within and immediately adjacent to the DBS target, the subcallosal and anterior cingulate, was significantly reduced with chronic stimulation, and parietal structures including the supramarginal gyrus and cuneus showed significant hyperactivity with stimulation. Both the cingulate and the parietal lobe have been repeatedly implicated in AN circuit models.. The cingulate is known to play a role in affective processing and the assignment of reward value to external stimuli, both processes known to be affected in AN patients. There are further direct anatomic projections from the SCC, along the anterior and dorsal cingulate, to the parietal lobe, suggesting in our study that a highly focal intervention can have a broad influence on neural circuits downstream from the DBS target in key AN-relevant structures.

3.4: White Matter Connectivity Strength in Adolescents with Anorexia Nervosa Across Taste Reward Pathways and During Weight Recovery

Brogan Rossi, BS, University of Colorado Denver, Aurora, Colorado, Megan Shott, BS, University of Colorado Denver, Aurora, Colorado, Marisa DeGuzman, BA; BS; Student, University of Colorado Denver, Aurora, Colorado, Tamara Pryor, PhD; FAED, Eating Disorder Center of Denver, Denver, Colorado, Guido Frank, MD, University of Colorado Denver, Aurora, Colorado

Anorexia nervosa (AN) is characterized by food restriction and severe underweight, and altered brain reward pathways could contribute to those behaviors. Here we tested whether white matter connectivity in adolescent AN is altered and whether weight recovery is associated with normalization of those fibers. We recruited 34 adolescents with AN and 33 healthy control adolescents (HC) who completed one diffusion weighed brain scan at beginning of treatment. Sixteen adolescents in each group completed two imaging sessions. Probabilistic fiber tractography was computed for fiber paths between brain reward circuit regions. A repeated measures ANOVA (multivariate Wilks lambda<0.001, p<0.012) indicated a significant connection strength interaction for brain scan by group for the right sided pathways basolateral amygdala to anterior cingulate (p<0.038), posterior insula to prefrontal cortex (p<0.027) and on the left from posterior insula to prefrontal cortex (p<0.003), medial orbitofrontal cortex (p<0.042) and middle orbitofrontal cortex (p<0.008), from anteroventral insula to caudate body (p<0.038) and orbitofrontal cortex gyrus rectus to caudate head (p<0.05). Post hoc analyses indicated that at scan 2 connection strength was smaller in the AN group compared to controls for the pathway left anteroventral insula to caudate body, but greater in AN compared to controls from left gyrus rectus to the caudate head. These results indicate that white matter connectivity changes during weight restoration in adolescent AN differently compared to healthy controls in a comparable time frame, with both connection strength increases and decreases in the AN group. These are preliminary results and we are currently collecting a larger sample. Nevertheless, the results indicate dynamic changes during recovery from AN in WM connectivity. Larger samples will test relationships with behavioral variables and whether connectivity strength is related to functional or effective connectivity.

3.5: Subliminal and Supraliminal Processing of Reward-Related Stimuli in Anorexia Nervosa

Ilka Boehm, DiplPsych, Technische Universität Dresden, Dresden, Saxony, Joseph A. King, PhD, Technische Universität Dresden, Dresden, Saxony, Fabio Bernardoni, PhD, Technische Universität Dresden, Dresden, Saxony, Daniel Geisler, Dipl.-Inf, Technische Universität Dresden, Dresden, Saxony, Maria Seidel, MSc, Technische Universität Dresden, Dresden, Saxony, Franziska Ritschel, DiplPsych, Technische Universität Dresden, Dresden, Saxony, Thomas Goschke, PhD, Technische Universität Dresden, Dresden, Saxony, John-Dylan Haynes, PhD, Charité Universitäts-Medizin, Berlin, Saxony, Veit Roessner, MD, Technische Universität Dresden, Dresden, Saxony, Stefan Ehrlich, MD, Technische Universität Dresden, Dresden, Saxony

Previous studies have highlighted the role of the brain reward and cognitive control systems in the etiology of anorexia nervosa (AN). In attempt to disentangle the relative contribution of these systems to the disorder, we used functional magnetic resonance imaging (fMRI) to investigate hemodynamic responses to reward-related stimuli presented both subliminally and supraliminally in acutely underweight AN patients and age-matched healthy controls (HC). fMRI data were collected from a total of 35 AN patients and 35 HC while they passively viewed subliminally and supraliminally presented streams of food, positive social and neutral stimuli. Activation patterns of the group * stimulation condition * stimulus type interaction effect were interrogated to invest potential group differences in processing different stimulus types under the two stimulation conditions. Moreover, changes in functional connectivity were investigated using the generalized psychophysiological interaction approach. AN patients showed a generally increased response to supraliminally presented stimuli in a region of lateral prefrontal cortex, the inferior frontal junction (IFJ), but no alterations within the reward system. Increased activation during supraliminal stimulation for food stimuli was observed in the AN group in the cuneus and the fusiform gyrus/parahippocampal gyrus. No group difference was found in respect of the subliminal stimulation condition and functional network connectivity. Increased IFJ activation in AN during supraliminal stimulation may indicate hyperactive cognitive control, which resonates with clinical presentation of excessive self-control in AN patients. Increased activation to food stimuli in visual regions may be interpreted in light of an attentional food bias in AN.

3.6: Macronutrient Intake Associated with Weight Gain in Adolescent Girls with Anorexia Nervosa

Traci Carson, BA; MPH, MPH expected May 2017, University of Michigan, Ann Arbor, Michigan, Charumathi Baskaran, MD, Neuroendocrine Unit, Massachusetts General Hospital, Boston, Massachusetts

Adolescents and women with anorexia nervosa (AN) are known to severely restrict fat intake and consume high amounts of fiber. However, data are limited regarding nutrition parameters that are associated with weight gain in AN. The objective was to prospectively investigate the macronutrient composition of diet associated with weight gain in adolescent girls with anorexia nervosa. This was a prospective study of 90 girls 12-18 years old; 45 with AN and 45 healthy normal-weight controls over a 6-12-month period. Subjects completed four-day food diaries and underwent body composition assessment using dual energy x-ray absorptiometry. Weight gain was defined as a ≥10% increase in BMI from baseline. Baseline clinical characteristics did not differ between girls with AN who did not gain weight (AN-0) versus those who did (AN-1) over the following 6-12 month period with the exception of percentage of calories from proteins (p=0.02). Total caloric intake did not differ between AN-0 and AN-1 at baseline or follow up. At follow up, compared to AN-0, AN-1 consumed a lower percentage of total calories from protein (p=0.001), and a higher percentage of total calories from fat (p=0.02). Compared to AN-0, between baseline and follow up, AN-1 had a significant increase in the percentage of total calories obtained from PUFA (p=0.007). Within the AN group, BMI at follow-up was associated positively with the percentage of total calories obtained from total fat (r=0.41, p=0.005), MUFA (r=0.44, p=0.002), and PUFA (r=0.33, p=0.03) at 6/12 months, and inversely with the percentage of total calories obtained from carbohydrates (r=-0.32, p=0.03) and proteins (r=-0.33, p=0.03). In conclusion, consuming a greater proportion of total calories from fat may assist in weight gain in adolescent girls with AN.

Topic: Body Image I
Club H, first floor
Scott Griffiths, PhD & Rachel Rodgers, PhD

4.1: The Psychosocial and Health Correlates of Drive for Muscularity in Young Adult Males

Trine Tetlie Eik-Nes, MSc, Norwegian University of Science and Technology/ Levanger Hospital, Norway, Trondheim, Sør-Trøndelag, Jerel P Calzo, ScD, San Diego State University Graduate School of Public Health, San Diego, California, S. Bryn Austin, ScD, Professor, Division of Adolescent and Young Adult Medicine, Boston Children's Hospital, Boston, Massachusetts, Aaron Blashill, PhD, Department of Psychology, San Diego State University, San Diego , California, Stuart B. Murray, PhD, Department of Psychiatry, University of California, San Fransisco, California

Drive for muscularity (DM) is the desire individuals have to develop a muscular physique. Cross-sectional studies indicate that DM may be associated with depression, muscle dysmorphia, disordered eating and muscle-building supplement consumption, but limited research to date has examined the psychosocial and health correlates of DM in the context of large scale epidemiologic cohort data. Using data from 3,255 males (ages 18-32 years) from the 2013 and 2014 waves of a nationally representative longitudinal Growing Up Today Study, generalized estimating equations estimated correlates (age, sexual orientation, weight status, education) of DM measured in 2013, and health outcomes (depressive symptoms, overeating, binge eating, purging and dieting, binge drinking, and muscle building product use (e.g., creatine, steroids) measured in 2014. DM was measured using the Drive for Muscularity Scale (DMS), which measures the degree of the respondents’ preoccupation with increasing their muscularity with higher scores indicating higher DM. DMS scores were entered continuously and we estimated how the odds of each outcome increased for each one unit increase of DM on the DMS. Models examining health outcomes adjusted for age, sexual orientation, weight status, education, and baseline (2013) health outcomes. DM decreased with age (β= -.0.03, 95% CI= -0.03, -.0.02, p<0.0001). Gay and bisexual males presented with higher DM compared to heterosexual males (β=0.18, 95% CI=0.17,0,43, p<0.0001). Higher DM was associated with increased odds of exhibiting elevated depressive symptoms (OR= 1.23; 95% CI= 1.05, 1.45; p<0.001), binge drinking (OR= 1.21; 95% CI= 1.01, 1.44; p<0.05), dieting (OR= 1.24; 95% CI= 1.07, 1.44; p<0.001) and use of muscle building products (OR= 4.51; 95% CI= 3.76, 5.43; p<0.0001). Health care providers should be made aware that DM may carry adverse health risks among young adult males.

4.2: Evaluating Brief Mindfulness and Cognitive Dissonance Intervention Strategies for Increasing Resilience to the Adverse Effects of Thin-Ideal Media Exposure on Body Image and Eating Disorder Risk Factors

Melissa Atkinson, PhD, University of the West of England, Bristol, Avon, Phillippa Diedrichs, PhD, University of the West of England, Bristol

Media images that promote an unrealistic thin appearance ideal have a well-established negative effect on body satisfaction and other risk factors for eating disorders. Traditionally, interventions have focused on evaluating negative risk factors and have neglected a parallel goal of assessing positive factors which may promote a resilient body image. This study assessed whether brief training and practice of a cognitive technique could result in improved body appreciation and body-related psychological flexibility (Aim 1), and buffer against the effects of later media exposure (Aim 2). Undergraduate women (N= 202, Mage = 19.90, SD = 2.75) completed baseline trait and state measures of eating disorder risk and protective factors and then watched a randomly allocated 15-min training video (mindfulness, cognitive dissonance, documentary control). One week after training, participants completed a second set of trait measures, then underwent a media exposure exercise before completing a final set of state measures. After 1 week, participants in mindfulness and dissonance conditions reported higher body appreciation (d = .45 and d = .51, respectively), and lower internalisation (d = .43 and d = .44, respectively) compared to control (Aim 1). Following media exposure, mindfulness and dissonance participants reported higher state weight satisfaction (d = .43 and d = .60, respectively), and lower appearance internalisation (d = .48 and d = .62, respectively) and perceived pressures (d = .54 and d = .67, respectively) compared to control (Aim 2). Brief portable body image and eating disorder prevention interventions using both mindfulness and cognitive dissonance have the potential to produce resilience to the negative effects of idealised media images and aid in reducing serious body image concerns and eating disorders. Future research is necessary to test mediation of proposed protective factors, and to assess the potential for longer-term resilience.

4.3: Acceptability and Feasibility of a Dissonance-Based body Image Intervention for Girl Guides and Girl Scouts: Qualitative Results from a Dissemination and Implementation Study Across Nineteen Countries.

Nadia Craddock, BSc; EdM, Centre for Appearance Research, University of the West of England, Bristol, Bristol, Phillippa Diedrichs, PhD, Centre for Appearance Research, University of the West of England, Bristol, Bailey Powe, BSc, Oregon Research Institute, Eugene, Oregon, Eric Stice, BS; MA; PhD, Oregon Research Institute, Eugene , Oregon

The World Association of Girl Guides and Girl Scouts (WAGGGS), the largest youth organization globally for girls, partnered with the Dove Self-Esteem Project to implement a positive body image intervention, Free Being Me (FBM), globally. FBM was adapted from the dissonance-based eating disorder (ED) prevention intervention, The Body Project. Since 2013, FBM has been disseminated in over 120 countries to 3 million young people. This study explored key stakeholders’ views on the adoption and implementation of Free Being Me, to inform future efforts to broadly implement body image and ED prevention interventions. National team members and local group leaders (N= 40) from nineteen countries took part in semi-structured interviews. Respondents shared their views on the acceptability, appropriateness and feasibility of implementing the intervention on a local and national scale. Interviews were audio-recorded, transcribed verbatim, and analyzed thematically. FBM was perceived as a valuable, timely, and effective intervention to improve girls’ body image. The intervention reportedly had a positive impact on girls, staff and leaders, and the wider community. Leaders mostly found it easy and enjoyable to deliver. The length and mode of delivery however, were challenging for some member organizations. Consequently, leaders often adapted the programme to suit the needs of their girl guides, culture, and organizational infrastructure. WAGGGS’s global co-ordination of financial, training, and organizational support was beneficial to implementation. However, the scale of delivery required from WAGGGS was a challenge for some member organizations, and a lack of infrastructure, future funding, and workforce capacity were seen as obstacles for sustainable long-term delivery. To ensure the effectiveness and sustainability of broadly implementing evidence-based body image and ED prevention interventions, solutions for organizational and funding challenges will be essential.

4.4: Profiles of Muscularity Concerns and Muscularity-Oriented Behaviors Among French Young Women

Marilou Girard, MA, University of Toulouse 2 Jean Jaurès, Toulouse, Occitanie, Henri Chabrol, MD; PhD, University of Toulouse 2 Jean Jaurès, Toulouse, Rachel Rodgers, PhD, Northeastern University, Boston, Massachusetts

In recent years, the female appearance ideal has become increasingly toned, firm and muscular, and there is a growing evidence of the importance of muscularity in women’s body image. In addition, it has been suggested that the internalization of this female athletic ideal might be associated with a range of deleterious eating and exercising behaviors. The aim of the present study was therefore to contribute to the emerging literature on muscularity concerns by exploring the different patterns of muscularity concerns and muscularity-oriented behaviors among French young women. A sample of 492 French female college student, of mean age 20.88 (SD = 2.72), completed a questionnaire assessing sociocultural influences, internalization of appearance ideals, appearance comparison, body dissatisfaction, disordered eating symptoms, drive for muscularity and psychological functioning. A cluster analysis revealed three groups: A group with “no interest in muscularity” (59%; n = 285), a group characterized by “muscularity concerns and behaviors” (10%; n = 49) and finally, a group characterized by “muscularity concerns only” (31%; n = 152). The three cluster groups differed from each other in terms of partner pressure, interest in the media, internalization of the thin and muscular ideal, and drive for thinness, with the “muscularity concerns and behaviors” group displaying the highest means. Furthermore, perfectionism scores were higher in this cluster group compared to the two other clusters. These findings confirm that muscularity is an increasingly important dimension of women’s body image in Western contexts, and suggest that partner and media influences might contribute to the development of muscularity-oriented behaviors among young women that may be detrimental to their health. Increasing our understanding of muscularity concerns is critical with a view to informing prevention programs for body image disorders and eating disorders among young women.

4.5: Mediators of a Change in Bulimic & Muscle Dysmorphia Symptoms in The Body Project: More than Muscles

Tiffany Brown, PhD, University of California, San Diego, San Diego, California, K. Jean Forney, MS, Florida State University, Tallahassee, Florida, Dennis Pinner, BS, Florida State University, Tallahassee, Florida, Pamela Keel, PhD, Florida State University, Tallahassee , Florida

The Body Project: More than Muscles (MTM), a dissonance-based (DB) intervention, has demonstrated efficacy in reducing eating disorder risk factors (e.g., bulimic symptoms) and more male-specific risk factors for muscle dysmorphia. The present study tested whether reductions in body-ideal internalization and drive for muscularity mediated intervention effects for these outcomes. Data were drawn from a randomized controlled trial in which 99 males were randomized to either a 2-session DB intervention (n=52) or a waitlist control condition (n=47). Mediation models were conducted using bias-corrected bootstrapped confidence intervals (CIs) to test the indirect effects of condition via the posited mediating variables (change in mediator from baseline to post-intervention) on the dependent variables of interest (change in bulimic/muscle dysmorphia symptoms from baseline to 4-week follow-up). Replicating results from previous trials, body-ideal internalization partially mediated intervention effects on bulimic symptoms (CI: -1.92 to -0.27) among males. Extending these results, body-ideal internalization also partially mediated the effect of condition on muscle dysmorphia symptoms (CI: -4.68 to -0.37). Examining the male-specific target of drive for muscularity, results support that drive for muscularity fully mediated condition effects on both bulimic symptoms (CI: -2.49 to -0.19) and muscle dysmorphia symptoms (CI: -12.19 to -0.46). Results are the first to demonstrate that targeting internalization of the male-specific muscular ideal causes reductions in both bulimic symptoms and muscle dysmorphia symptoms. Results provide preliminary support for body-ideal internalization and drive for muscularity as causal risk factors for bulimic and muscle dysmorphia symptoms among males and support male-specific theoretical models of risk for both eating and appearance-related disorders.

4.6: Challenging Fat Talk: An Experimental Investigation of Reactions to Body-Disparaging Conversations.

Suman Ambwani, PhD, Dickinson College, Carlisle, Pennsylvania, Megan Baumgardner, BA, Vanderbilt University, Nashville, Tennessee, Cai Guo, BA, Stanford University, Stanford, California, Lea Simms, BA, Dickinson College, Carlisle , Pennsylvania, Emily Abromowitz, Student, Dickinson College, Carlisle, Pennsylvania

The “Fat Talk Free Week” eating disorder prevention campaign posits that “fat talk” (i.e., body-disparaging/objectifying conversations) contributes to myriad negative health outcomes, a premise that has been corroborated by several experimental and correlational studies. However, two important questions remain: 1) why do people engage in these harmful conversations, and 2) could we use feminist theory to develop a feasible conversation alternative to break the cycle of fat talk? The current experimental study examined women’s responses to fat talk and feminist-based challenging fat talk scenarios through a vignette paradigm. The experimental vignettes were developed and pilot-tested with a sample of undergraduate women (N=32). Undergraduate women (N=266) at a small liberal arts college in the Northeastern United States then participated in a two-part study: in Part I, they completed online questionnaires assessing demographics, body dissatisfaction, baseline/typical fat talk engagement, feminist identification, and socially desirable responding, and in Part II (individual experimental sessions completed one week later), they viewed either the “fat talk” or “challenging fat talk” vignettes and then completed self-report measures assessing fat talk engagement, perceived acceptability of the vignette, social likeability of the target character, and mood. Results indicated that participants rated the “challenging fat talk” scenario as more socially attractive, the target character as more likeable, and this condition yielded less negative affect and fat talk engagement. Moreover, baseline body dissatisfaction, baseline fat talk tendencies, negative affect, and low levels of feminist identification all predicted post-exposure fat talk engagement. Present findings raise important questions about social conformity and women’s fat talk engagement, and offer possibilities for attending to feminist-based conversation as complements for extant fat talk prevention efforts.

Topic: Personality and Cognition
Meeting Hall 1A, first floor
Melanie Brown, PhD & Joanna Steinglass, MD

5.1: Stability and Change in Personality Following Treatment

Johanna Levallius, BSc; MSc, Karolinska Institute, Stockholm, Claes Norring, PhD, Karolinska Institute, Stockholm, David Clinton, PhD, Karolinska Institute, Stockholm, Stockholm, Brent Roberts, PhD, University of Illinois, Urbana-Champaign , Illinois

There’s a growing body of evidence of the close relationship between personality and mental illness, and personality can be said to represent a transdiagnostic perspective. Little is known of the long-term relationship between eating disorder and personality as measured by the five-factor model. The aim of this study was to measure stability and change in personality following two different psychotherapies. 209 adults with eating disorder (ED) enrolled either in a four-month intensive, multimodal psychodynamic group-therapy (DAY) or four-six month internet-based supported cognitive behavioral therapy (iCBT). ED diagnosis, symptoms and personality (NEO PI-R) were assessed at baseline, termination and at 6-month follow up. Structural equation modeling was used to analyze overall change. Recovery rate was 71% in DAY and 55% in iCBT. Neuroticism decreased significantly while Extraversion, Openness and Conscientiousness increased. The two predictors, treatment and outcome, had little influence on pattern of change. At the facet-level, there was a high degree of rank-order stability (rmean = .69). Still, on average, 28% reliably changed in any given facet; and there were differences in change based on treatment and outcome. This study lends support for the possibility of personality change, emphasizes the relevance of facet-level study and that personality change play a role in recovery.

5.2: The Uncertainty Principle: A Review of Intolerance of Uncertainty in Eating Disorders

Melanie Brown, MA; PhD, Icahn School of Medicine at Mount Sinai, New York, New York, Lauren Robinson, BSc; MPhil, Institute of Child Health, University College, London, Milano, Giovanna Cristina Campione, PhD, Child Psychopathology Unit, Scientific Institute, IRCCS Eugenio Medea, Bosisio Parini, Milano, Kelsey Wuensch, BA, Icahn School of Medicine at Mount Sinai, New York , New York, Nadia Micali, MD; PhD, Icahn School of Medicine at Mount Sinai, New York, New York, Tom Hildebrandt, PsyD, Icahn School of Medicine at Mount Sinai, New York, New York

Intolerance of uncertainty is a dimensional construct studied extensively in anxiety disorders that may have relevance in eating disorders as a vulnerability and underlying neurobiological mechanism. This systematic review and meta-analysis aims to synthesize the evidence of intolerance of uncertainty in eating disorders. A search of electronic databases was conducted, and manuscripts were selected. Meta-analysis utilizing the Intolerance of Uncertainty Scale (IUS) revealed that intolerance of uncertainty was significantly elevated in women with eating disorders when compared to healthy controls (SMD = 2.07, 95% C.I. 1.24 to 2.79; p<0.0001). Significant differences were also found when comparing women with anorexia nervosa to controls (SMD = 2.16; 95% C.I. 1.14 to 3.18; p < 0.0001) and women with bulimia nervosa to controls (SMD = 2.03; 95% C.I. 1.31 to 2.80; p < 0.0001). Findings suggest that intolerance of uncertainty might be a transdiagnostic feature of eating disorders, with particular importance in anorexia nervosa. A maladaptive cognitive style characterized by intolerance of uncertainty may represent a risk and maintenance factor for eating disorders. Evidence of a neurobiological basis for intolerance of uncertainty, assessment of the construct, and the clinical impact of intolerance of uncertainty in developing novel, targeted interventions for eating disorders will be discussed.

5.3: Exploring the Clinical and Neuropsychological Profile of a Non-Clinical Sample of ADHD Comorbid with Eating Disorders.

Bruno Nazar, MSc; MD; PhD, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, Sergeant, MSc; CPsychol; PhD, Professor, Vrjie University, Amsterdan, London, Janet Treasure, MD; PhD; FAED, Professor, King's College London, London, UK, Paulo Mattos, MSc; MD; PhD, Professor, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil

The comorbidity of Attention Deficit / Hyperactivity Disorder (ADHD) comorbid with Eating Disorders (ED) is of interest to define a subset of patients that might have different prognosis and treatment response. Previous studies revealed that these patients present higher frequencies of other comorbid disorders and higher levels of disordered eating behaviours when compared to patiens with either ED or ADHD alone. The association of ADHD with obesity has also been largely demonstrated but reasons for this association need further investigation with the presence of an ED being a possible candidate. We have investigated if subjects with ADHD+ED (n=16) differed from ADHD only (n=35) and from Controls (n= 39) in demographic measures (BMI, socioeconomical level), self report questionnaires for depression, anxiety, impulsivity and in neuropsychological tasks exploring vigilance (continuos performance test (CPT)), executive functioning (Visual and Digit Span) and decision making (Iowa Gambling Task). The ADHD+ED group had significantly (p<.001) higher weight (+13kgs; +4 BMI points) than ADHD only and Controls. Also, ADHD+ED presented a significantly (p<.001) greater number of current Hyperactive/Impulsive symptoms than ADHD only group. Finally, ADHD+ED presented significantly (p<.001) greater omission errors on the CPT and a trend (p=.053) for more disadvantegeous choices on the IGT when compared to the other groups. We concluded that patients with the comorbidity ADHD+ED have higher distractibility and a poorer decision making which might impair their ED treatment. When approaching patients with ADHD+ED clinicians should consider these characteristics when using psychotherapy tools that rely on self monitoring and vigilance. ADHD patients that start to gain weight should be screened for the presence of an ED.

5.4: Attentional Bias Modification Reduces Chocolate Consumption: The Role of Habitual Craving

Eva Kemps, MPsych; PhD; BPsych&Ed, Flinders University, Adelaide, South Australia, Marika Tiggemann, BA; PhD, Flinders University, Adelaide, South Australia, Ebony Stewart-Davis, BPsych(Hons), MPsych

High levels of food craving are characteristic of individuals who binge eat, such as those with binge eating disorder, bulimia nervosa and some obese individuals. One neglected contributing factor to this unwanted (over)consumption is attentional bias for craving-related food cues. This study examined whether attentional bias modification can reduce craving-driven consumption. Using a modified dot probe task, 176 women (Mage = 20.19 years; MBMI = 23.17 kg/m2) were trained to direct attention away from, and thus avoid, craving-related food cues (pictures of chocolate). Chocolate was chosen because it is the most commonly craved food in Western cultures. Chocolate consumption was measured by an ostensible taste test. Habitual chocolate craving was measured by the Craving sub-scale of the Attitudes to Chocolate Questionnaire. Results showed that habitual chocolate craving moderated the relationship between attentional re-training and chocolate consumption. Specifically, while individuals with low levels of chocolate craving ate less chocolate following attentional re-training, individuals with high levels of craving actually ate more. It is possible that exposure to chocolate in the taste test may have triggered a craving in individuals with high levels of chocolate craving. This suggests that more extensive re-training may be required to combat craving-driven consumption in these individuals than the single training session used here. Theoretically, the results are consistent with cognitive-motivational models of craving which hold that craving-related cognitive biases drive consumption. At a more practical level, they highlight the need to tailor cognitive bias modification protocols to accommodate diagnostic characteristics of individuals with disordered eating, such as high levels of food craving.

5.5: Enduring Changes in Decision Making in Patients with Full Remission from Anorexia Nervosa

Trevor Steward, MSc, Bellvitge University Hospital-Idibell, Ciberobn, Barcelona, Barcelona, Gemma Mestre-Bach, MSc, Bellvitge University Hospital - Idibell, Ciberobn, Barcelona, Barcelona, Zaida Agüera, PhD, Bellvitge University Hospital— Idibell, Ciberobn, Barcelona, Barcelona, Roser Granero, PhD, Idibell, Ciberobn Fisiopatologia Obesidad y Nutrición Idibell, Ciberobn Instituto Salud Carlos III, Barcelona , Spain, Isabel Sánchez, PhD, Bellvitge University Hospital— Idibell, Ciberobn, Barcelona, Spain, Nadine Riesco, PhD, Bellvitge University Hospital— Idibell, Ciberobn, Barcelona, Spain, Iris Tolosa-Sola MSc, Bellvitge University Hospital— Idibell, Ciberobn , Barcelona, Spain, Francisco Tinahones, PhD, Idibell, Ciberobn, Barcelona, Spain, Felipe F Casanueva, PhD, Idibell, Ciberobn, Barcelona, Spain, Cristina Botella, PhD, Idibell, Ciberobn, Barcelona, Spain, Rafael de laTorre, PhD, Idibell, Ciberobn, Barcelona, Spain, Jose M Fernández-Real, PhD, Idibell, Ciberobn, Barcelona, Spain, Gema Frühbeck, PhD, Idibell, Ciberobn, Barcelona, Spain, Susana Jiménez-Murcia, PhD, Bellvitge University Hospital— Idibell, Ciberobn,Barcelona, Sapin, Fernando Fernández-Aranda, PhD, Bellvitge University Hospital— Idibell, Ciberobn, Barcelona, Spain

The aim of this study was to evaluate decision making performance in anorexia nervosa patients before beginning treatment and a one-year follow up. Anorexia nervosa patients (n = 42) completed the Iowa Gambling Task upon admission to a 3-month day-hospital treatment program and at a one-year follow-up. Patient Iowa Gambling Task performance was compared to age-matched controls (n = 46). Anorexia nervosa patients displayed poorer performance on the Iowa Gambling Task at admission compared to controls (p<.001). Patients with full remission (n = 31; 73.9%) at the 1-year follow-up improved Iowa Gambling Task performance (p = 0.007), and scores were similar compared to those of controls (p = 0.557). Anorexia nervosa patients with partial/no remission at follow-up (n = 11; 26.1%) did not improve Iowa Gambling Task scores (p = 0.867). These findings uphold that enduring remission from anorexia nervosa can reverse decision-making impairments, and that such impairments might be most likely explained by clinical state rather than a trait vulnerability.

5.6: Differences in the Ability to Delay Monetary Gratification Across the Eating Disorder Spectrum

Savani Bartholdy, BSc; MSc, Institute of Psychiatry, Psychology & Neuroscience, King's College London, London, London, Samantha Rennalls, BSc; MSc, Institute Of Psychiatry, Psychology & Neuroscience, Kings College London, London, UK, Iain Campbell, DSc, Institute Of Psychiatry, Psychology and Neuroscience, Kings College London, London, UK, Ulrike Schmidt, MD; PhD, FRCPsych, Institute of Psychiatry, Psychology & Neuroscience, King's College London, London , UK, Owen O'Daly, PhD, Institute of Psychiatry, Psychology & Neuroscience, King's College London.

Bulimia nervosa (BN) and binge eating disorder (BED) have been associated with greater reward-related impulsivity, reflected by a reduced capacity to delay gratification (i.e., greater temporal discounting [TD]) whereas the opposite has been found for anorexia nervosa (AN). However, differences in TD have not yet been directly compared between eating disorders. This study is the first to investigate the capacity to delay gratification across the eating disorder spectrum and compared to healthy individuals. A total of 94 women participated in the study: 66 women meeting the DSM-V criteria for an eating disorder (28 AN, 27 BN, 11 BED) and 28 healthy women (HC). Reward-related impulsivity was assessed via self-report (Delaying Gratification Inventory (DGI) questionnaire) and a hypothetical monetary TD task. Clinical variables were assessed using the Eating Disorder Examination Questionnaire (EDE-Q), and the Depression, Anxiety and Stress Scale (DASS-21). A main effect of group was observed for both TD performance andself-reported reward-related impulsivity on the DGI. Self-reported impulsivity was greater in both women with BED and BN compared to women with AN and healthy women, higher rates of TD were only observed in women with BN compared to AN, implicating the importance of perception of impulsivity or loss of control in BED. Moreover, a poorer self-reported capacity to delay gratification was associated with greater illness severity and frequency of binge eating. BMI and mood (DASS-21 total scores) significantly correlated with TD and DGI measures, however including these variables as nuisance covariates did not influence the result. These findings hold implications for the treatment of altered behavioural control in the context of different eating disorder diagnoses.

Topic: BED and Obesity
Meeting Hall 1B, first floor
Alexandra Dingemans, PhD & Christine Peat, PhD

6.1: Sweet Taste Preference in Binge-Eating Disorder: A Preliminary Investigation

Erica Goodman, BA, University of North Dakota, Grand Forks, North Dakota, Lauren Breithaupt, MA, George Mason University, Fairfax, Virginia, Hunna Watson, PhD, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Christine Peat, PhD, University of North Carolina at Chapel Hill, Chapel Hill , North Carolina, Jessica Baker, PhD, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Cynthia Bulik, PhD, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Kim Brownley PhD, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.

Research suggests that individuals with sweet taste preference are at greater risk for binge eating; however, it is unknown whether sweet liking affects health outcomes and eating behaviors in those with binge-eating disorder (BED). Forty-one adults (85.3% female, 82.9% White) with BED completed a sweet taste test, including ratings of pleasantness of five solutions (0.05M to 0.83M), and self-report measures of eating disorder behaviors and food cravings. A subset of participants completed an oral glucose tolerance test (OGTT; N=21) and a 24-hr dietary recall (N=26). ANCOVA tests were used to compare those who were sweet likers (SL [N=18]; rated 0.83M most pleasant) vs. sweet dislikers (SDL[N=23]; all others) on outcome variables; a Fischer’s exact test was used to compare these groups on diabetes status. Small (.01), medium (.06), and large (.14) effect sizes (ηp2) were examined. SL and SDL did not differ on BMI or global eating psychopathology. Compared with SDL, SL reported numerically higher binge-eating frequency in the past 28 days (20.2 vs. 14.0 episodes) and higher 24-hr caloric intake (2972.2 vs. 2125.1 kcals) and protein intake (440.9 vs. 327.0 kcals; ηp2’s=0.06-0.15); SL exhibited significantly smaller postprandial insulin area under the curve (AUC; ηp2 =0.36) and numerically smaller postprandial glucose AUC (ηp2=0.17) and insulin sensitivity index (1.83 vs. 2.96, ηp2=0.14). All group differences in food cravings were associated with small effect sizes (ηp2’s<0.025). Based on the OGTT and World Health Organization criteria for diabetes, there was no difference between SL and SDL on diabetes status (no, pre-, diabetes). Though preliminary due to limited sample size, results indicate that individuals who prefer sweet taste and have BED may be at increased risk for binge-eating, higher intake of macronutrients, and diminished insulin sensitivity. Understanding the role of sweet liking may contribute to a reduction in medical consequences in this population.

6.2: Bariatric Patients, Weight Regain and Psychiatric comorbities: Systematic Review

Maria Francisca Firmino Prado Mauro, MD; Student, Instituto de Psiquiatria - Universidade Federal do Rio de Janeiro, Brazil, APO AE, Rio de Janeiro, Marcelo Papelbaum, MD; PhD, Instituto de Psiquiatria - Universidade Federal do Rio de Janeiro, Brazil, APO AE, Rio de Janeiro, José Carlos Appolinario, MD; PhD, Instituto de Psiquiatria - Universidade Federal do Rio de Janeiro, Brazil, APO AE, Rio de Janeiro, Marco Antônio Alves Brasil, MD; PhD, Universidade Federal do Rio de Janeiro UFRJ, Rio de Janeiro , APO AE, Brazil, João Regis Ivar Carneiro, MD; PhD, Universidade Federal do Rio de Janeiro UFRJ, Rio de Janeiro, APO AE, Brazil

Bariatric surgery (BS) has been recognized as a gold standard treatment for obesity. Nevertheless, among those who underwent surgery, there is a significant group of individuals who exhibits weight regain (WR). Several factors are proposed to contribute to a poorer prognosis, including the presence of general and eating psychopathology. The objective of this study was to developed a systematic review of the literature of the relationship between psychiatric comorbidity and WR in obese individuals submitted to BS. This review was made according to PRISMA guidelines. A structured search, using a previous discussed set of key-words, was done in several databases (PubMED, Web of Science, Cochrane, Scopus and Psycoinfo), looking for studies that investigated the association between eating disorders and general mental illnesses and WR in patients submitted to BS with at least 18 months of follow-up. A total of 2311 articles was first screened. After the stepwise selection procedure, using specific inclusion criteria, by two independent reviewers, 15 articles were included in the review. Although several authors examined specific causes for insufficient weight loss, only few studies focused on factors that lead to WR in the post-bariatric period. Some limitations of the studies included the absence of a common definition of clinically significant WR, the lack of structured diagnosis of mental disorders and the small sample sizes. Nevertheless, the majority of studies found an association between the presence of psychopathology and WR. Specifically, post-bariatric patients who presented with WR exhibits higher rates of binge eating disorder and eating behaviors (grazing, loss of control and pickling and nibbling). This review outlined the potential re-emergence of maladaptive eating behavior for those submitted to BS and its impact on the long-term weight maintenance. However, a consensus should be discussed in order to define how to address WR in clinical studies.

6.3: Reliability and Validity of the Stanford Integrated Psychosocial Assessment for Bariatric Surgery

Lianne Salcido, BA, PGSP-Stanford, PsyD Consortium, Palo Alto, California, Alina Kurland, MS, PGSP-Stanford PsyD Consortium, Palo Allto, California, Debra Safer, MD, Stanford University Department of Psychiatry and Behavioral Sciences, Palo Alto, California, Sarah Adler, PsyD, Stanford University, Palo Alto, California, Lilya Osipov, PhD, Stanford University, Palo Alto, California, Lindsay Wakayama, BS, PGSP-Stanford PsyD Consortium, Palo Alto, Claifornia, Natasha Fowler, BS, Stanford University, Palo Alto, California

Emerging evidence has shown that disordered eating behaviors play a role in suboptimal weight loss outcomes among bariatric surgery patients; however, there have been few studies indicating reliable behavioral variables that predict which patients are most at risk. Existing measures are limited by insufficient characterization of disordered eating (e.g. binge eating) and other psychosocial behaviors. We developed the Stanford Integrated Psychosocial Assessment for Bariatric Surgery (SIPABS)—an 18-item screening tool spanning 4 psychosocial domains that demonstrate associations to post-surgical outcomes. The SIPABS was applied to pre-surgical evaluations within the clinical charts of 60 patients who had bariatric surgery in 2012. Review was conducted by 4 raters (2 psychologists and 2 doctoral students). Inter-rater reliability was calculated averaging Cohen’s Kappa (K) for each rater set in the 4 domains: 1) Readiness for Surgery (K=.6, p= .36); 2) Availability and Functionality of Social Support (K=.43, p = .26); 3) Psychopathology/Disordered Eating (K=.65, p<.001) 4) Coping (K=.61, p< .001). Moderate to good inter-rater reliability was found. Predictive validity using linear regression revealed that higher SIPABS scores, using all 18 items, significantly predicted suboptimal weight loss at 2 years post surgery (F=2.615, p=.01). Specific items of “Understanding of the Bariatric Surgery Process" and “Presence of Psychopathology” significantly predicted worse weight loss outcomes. Pearson’s correlations showed compliance (r=.413, p=.007), availability of social support (r=.413, p=.007), coping with stress (r=.352, p=.002), and eating disorder psychopathology (r=-.385, p=.012) significantly correlated to weight loss outcomes 2 years post surgery. Findings suggest associations between disordered eating, behavioral factors, and poor post-surgical outcomes. Early identification of risk due to disordered eating and other behaviors may allow for targeted interventions.

6.4: Comparative Effectiveness of Treatments for Binge-Eating Disorder: Systematic Review and Network Meta-Analysis

Christine Peat, PhD, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Nancy Berkman, PhD, RTI International, Durham, North Carolina, Kathleen Lohr, PhD, RTI International, Durham, North Carolina, Kimberly Brownley, PhD, University of North Carolina at Chapel Hill, Chapel Hill , North Carolina, Carla Bann, PhD, RTI International, Durham, North Carolina, Katherine Cullen, BA, RTI International, Durham, North Carolina, Cynthia Bulik PhD, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

The Agency for Healthcare Research and Quality 2016 systematic review on the management and outcomes for binge-eating disorder (BED) revealed several efficacious treatment options including both pharmacological and psychological interventions. In the current review, we provide information on the comparative effectiveness of these treatments. We searched for BED treatment articles published from inception through mid-November 2015 and updated the MEDLINE® search through mid-May 2016. We selected only studies meeting predetermined inclusion and exclusion criteria and used the Cochrane risk-of-bias tool to assess the risk of bias of the included studies. In all, 30 trials (12 comparing pharmacological interventions and 18 comparing psychological interventions) revealed 27 treatment comparisons. We were able to evaluate 1 pharmacological comparison (second-generation antidepressants [SGAs] vs lisdexamfetamine) with network meta-analysis; we evaluated the 26 comparisons of psychological interventions – primarily cognitive behavioral therapy (CBT) and behavioral weight loss (BWL) – qualitatively. Across these treatment comparisons, only 3 significant differences emerged: lisdexamfetamine was better for increasing binge abstinence than SGAs; therapist-led CBT was better for reducing binge-eating frequency than therapist-led BWL, but BWL was better for reducing weight. Thus, both pharmacological and psychological interventions are effective at reducing binge eating and improving associated outcomes; however, few differences between these treatments emerged. The current results suggest that patients and clinicians can make informed choices among several effective treatment options (e.g., CBT, lisdexamphetamine) depending on patient preference and treatment goals. Such information can empower individuals in making decisions about treatment; however, additional research is needed to determine which interventions might be more effective for individual patients with BED.

6.5: Binge Eating Disorder and Food Addiction: Effects on Weight Loss and Attrition during Behavioral Obesity Treatment

Ariana Chao, APRN-BC; PhD, University of Pennsylvania, Philadelphia, Pennsylvania, Jena Shaw, PhD, University of Pennsylvania, Philadelphia, Pennsylvania, Zayna Bakizada, BA, University of Pennsylvania, Philadelphia, Pennsylvania, Naji Alamuddin, MD, University of Pennsylvania, Philadelphia , Pennsylvania, Rebecca Pearl, PhD, University of Pennsylvania, Philadelphia, Pennsylvania, Christina Hopkins, BS, University of Pennsylvania, Philadelphia, Pennsylvania, Emilie Pinkasavage BS, University of Pennsylvania, Philadelphia, Pennsylvania, Nasreen Alfaris, MD; MPH, Massachusetts General Hospital, Boston, Massachusetts, Robert Berkowitz, MD, University of Pennsylvania, Philadelphia, Pennsylvania, Thomas Wadden, PhD, University of Pennsylvania, Philadelphia, Pennsylvania

There is a lack of consensus about whether individuals with binge eating disorder (BED) or food addiction (FA) benefit less from behavioral weight loss (BWL) relative to those without these conditions. This study examined differences in weight loss and attrition between participants with and without BED participating in a 14-week BWL intervention for obesity. We also explored the effects of meeting criteria for FA or not. Data were from 178 participants (mean age=44.2±11.2 yrs; baseline BMI= 40.9±5.9 kg/m2; 88.2% female; 70.8% Black) who followed a 1000-1200 kcal/d diet and attended weekly group lifestyle modification sessions. BED diagnosis was assessed using the Questionnaire on Eating and Weight Patterns-5 and confirmed by clinician interview. Participants completed the Yale Food Addiction Scale and weight was measured weekly. We used an intention-to-treat analysis to compare reductions in initial weight among participants with and without BED or FA. Analyses were conducted adjusting for demographic factors and baseline weight. Six (3.4%) participants met criteria for BED. Percent weight loss did not differ between individuals with (9.40±1.59%) and without BED (9.16±0.30%; p=0.88). Two of the 6 (33.3%) participants with BED dropped out of the program compared to 26 of the 172 (15.1%) participants without BED (p=0.24). Twelve (6.7%) participants met criteria for FA. Individuals with FA lost significantly less of their initial weight (6.85±1.17%) compared to those without FA (9.32±0.30%, p=0.04). Four of the twelve participants (33.3%) with FA did not complete the program, compared to 24 of the 162 (14.5%) participants without FA (p=0.10). Our results suggest that BWL produces similar weight loss effects among individuals with and without BED. Individuals who meet criteria for FA may need additional support during BWL treatment.

6.6: Validity of DSM-5 Indicators of Binge Eating Episodes in Obese Adults: An Ecological Momentary Assessment Study

Andrea B. Goldschmidt, PhD, Alpert Medical School of Brown University, Providence, Rhode Island, Ross D. Crosby, PhD, Neuropsychiatric Research Institute, Fargo, North Dakota, Li Cao, MS, Neuropsychiatric Research Institute, Fargo, North Dakota, Stephen A. Wonderlich, PhD, Neuropsychiatric Research Institute, Fargo , North Dakota, Carol B. Peterson, PhD, University of Minnesota, Minneapolis, Minnesota, James E. Mitchell, MD, Neuropsychiatric Research Institute, Fargo, North Dakota, Scott G. Engel PhD, Neuropsychiatric Research Institute, Fargo, North Dakota

DSM-5 criteria for binge eating disorder (BED) include 5 features of binge eating episodes which are intended to aid diagnosticians in determining the presence and severity of loss of control over eating. The validity of these features is currently unclear. We examined the extent to which DSM-5 indicators predicted self-reported binge eating (i.e., endorsing both overeating and loss of control) among 50 obese adults participating in an ecological momentary assessment (EMA) study. EMA timestamps were used to approximate “eating much more rapidly than usual” (i.e., eating episodes ≤-1 SD from each participant’s average eating episode duration). Self-reported pre-episode hunger levels, post-episode fullness, pre-episode feelings of shame + eating alone, and post-episode feelings of disgust, depression, and/or guilt were used to approximate the remaining 4 features: “eating until uncomfortably full,” “eating large amounts of food when not physically hungry,” “eating alone due to embarrassment over how much one is eating,” and “feeling disgusted with oneself, depressed, or very guilty after overeating,” respectively. Individual regression models, adjusted for body mass index and BED status (full- or subthreshold: n=8; no BED: n=42), revealed that binge eating was associated with lower pre-episode hung

Topic: Risk Factors
Forum Hall, second floor
Kyle De Young, PhD & Debra Franko, PhD, FAED

7.1: Factors Related to the Eating Disorders of Male Japanese Junior High School Students: A Longitudinal Population Study Comparing 2010 and 2015

Gen Komaki, MD; PhD, International University of Health and Welfare, Ohkawa, Fukuoka, Mitsuhiko Tojo, MEd, Okayama University Graduate School of Education, Okayama, Kanagawa, Motonari Maeda, MEd, Joshibi University of Art and Design, Sagamihara, Kanagawa

The prevalence of eating disorder symptoms among the early adolescent male population of Japan was studied. More than 10,000 Japanese junior high school students aged 12 to 15 years residing in urban (City A) and rural (City B) areas were enrolled. Data of 2,551 boys in 2010 and 2,930 in 2015 was obtained by self-administered questionnaires that included the Eating Disorder Examination Questionnaire (EDE-Q 6.0) and twenty-two items related to risk factors for eating disorders. Four subscales; “restriction” (R), “eating concern” (EC), “shape” and “weight” concerns (SC and WC); were assessed, as were “binge eating” behaviors (BE) and “inappropriate compensation behaviors: purging behaviors such as self-induced vomiting (V), laxative abuse (LA), and Fasting. The prevalence of eating disorder symptoms scoring higher than 4.0 increased from 0.5% to 0.9% for R (p=.07), 0.1% to 0.3% for EC (p=.13), 0.8% to 1.4% for SC (p=.03), and 0.6% to 1.3% for WC (p=.01). Clinically significant BE (more than four times a month) increased from 4.2% to 5.3% (p=.002), V from 1.6% to 2.2% (p=.03), LA (more than twice a month) from 1.0%, and 1.8% (p=.00), and Fasting from 2.8% to 3.3% (p=.001). However, in comparison of the two areas studied, City A showed significant increases for almost all of the subscales, BE, and compensation behaviors, along with an increase in the prevalence rate of the items related to risk factors for eating disorders. In contrast, the findings in City B did not show similar changes. Our longitudinal population study of male Japanese junior high school students indicates that the increase in prevalence over the past five years may be related to changes in the risk factors for the eating disorders of early adolescents.

7.2: Developmental Premorbid BMI Trajectories of Adolescents with Eating Disorders in a Longitudinal Population Cohort

Zeynep Yilmaz, PhD, University of North Carolina, Chapel Hill, North Carolina, Nisha Gottfredson, PhD, University of North Carolina, Chapel Hill, North Carolina, Stephanie Zerwas, PhD, University of North Carolina, Chapel Hill, North Carolina, Cynthia Bulik, PhD, University of North Carolina, Chapel Hill , North Carolina, Nadia Micali, MD; PhD, Icahn School of Medicine at Mount Sinai, New York, New York

Body mass index (BMI) plays a defining role in eating disorder (ED) diagnosis; however, whether child and adolescent BMI predicts later ED risk is not well understood. The present study examined whether developmental BMI trajectories represent a significant prospective risk factor for EDs. Using data from the Avon Longitudinal Study of Parents and Children (N=1839), we used subject-specific conditional growth models to describe premorbid BMI trajectories for individuals with anorexia nervosa (AN; N=261), bulimia nervosa (BN; N=333), binge eating disorder (BED; N=126), and purging disorder (PD; N=145), starting at 1.5 months and ranging to age 15.5 or 12 months prior to ED onset, whichever came first. Child and adolescent BMI was calculated for 22 time points via face-to-face assessments and maternal-report questionnaires. Self-report data on ED status were collected at ages 14, 16, and 18. Distinct developmental trajectories emerged for EDs at a young age compared with the no ED developmental trajectory. Most notably, the average growth trajectory for AN departed significantly lower than other trajectories by age 2 for females and age 7 for males (compared with no ED group: intercept=-1.72[SE=.19] BMI units below at 100 months; slope =-.12[SE=.02] BMI units slower growth per month). The mean BMI trajectories for BN, BED, and PD were not statistically distinguishable from one anbut they were consistently higher than the mean trajectories for AN and no ED group (compared with no ED group: intercepts range from 1.21[SE=.12] to 1.39[SE=.29] BMI units higher at 100 months; slope=.12[SEs .02 to .04] BMI units faster growth per month). While maternal ED status did not predict child ED, a 1 unit decrease in maternal BMI was associated with a 1.08 factor increase in the odds of developing AN, and a 1 unit increase in maternal BMI was associated with a 1.05, 1.08, and 1.13 factor increase in the odds of developing BN, BED, and PD, respectively Taken together, our results provide important clues about the role of premorbid metabolic factors and weight in the etiology of EDs. Especially for AN, premorbid low weight may be a key biological risk factor. Observing children whose BMI trajectories persistently deviate from age norms for signs of disordered eating could potentially assist with identifying individuals at high risk for EDs.

7.3: Attachment and Hypothalamus-Pituitary-Adrenal axis functioning in patients with Eating Disorders

Alessio Maria Monteleone, MD, Department of Psychiatry, Second University of Naples, Naples, Campania, Umberto Volpe, MD; PhD, Department of Psychiatry, Second University of Naples, Naples, Campania, Francesca Pellegrino, MD, Department of Psychiatry, Second University of Naples, Naples, Campania, Giovanna Fico, MD, Department of Psychiatry, Second University of Naples, Naples , Campania, Francesco Monaco, MD, Department of Psychiatry, Second University of Naples, Naples, Campania, Palmiero Monteleone, MD; Professor, Department of Medicine, Surgery and Dentistry, University of Salerno, Salerno, Salerno

Attachment theory postulates that different attachment styles influence the development of individual’s self-esteem and modulate the individual’s ability to cope with stressful events by responding with adequate affective regulation. Life events and emotion regulation are involved in the onset and maintenance of eating disorders (ED) and insecure attachment style has a high prevalence in ED subjects. The variables mediating the relationship between attachment style and ED psychopathology have not been explored sufficiently. To assess the role of the endogenous stress response system in this relationship, we investigated the effect of attachment style on the hypothalamus-pituitary-adrenal (HPA) axis functioning in ED patients. Fifty-two women with ED and 25 healthy women filled in the Experience in Close Relationship questionnaire to evaluate their attachment style, and collected saliva samples in the morning to measure the Cortisol Awakening Response (CAR). Thirty ED patients and 15 healthy women took part also into the Trier-Social-Stress-Test (TSST). During TSST, subjective anxiety was measured by the State-Trait Anxiety Inventory and saliva samples were collected to measure cortisol levels. Avoidant attached patients showed an enhanced CAR compared to both anxious and secure attached patients. In the initial TSST phase (threat appraisal), cortisol levels decreased in both secure patients and healthy subjects but not in insecure attached patients. In the TSST stress response phase, the cortisol increase was higher in insecure patients and delayed in avoidant ones. Finally, in the TSST recovery phase, cortisol levels returned to pre-test values in all but not in the avoidant patient group. In insecure patients, the mean state anxiety score was higher than in secure ones. These results suggest, for the first time, that in adults with ED attachment styles influence HPA axis and psychological responses to stressful events, which could have a role in the pathophysiology of ED.

7.4: Eating Rate is Increased Among Disinhibited Eaters Whose Attention is Directed Elsewhere

Kyle De Young, PhD, University of Wyoming, Laramie, Wyoming, Taylor Perry, Student, University of North Dakota, Grand Forks, North Dakota, Terra Towne, MA, University of North Dakota, Grand Forks, North Dakota, Erin Murtha-Berg, MA, University of North Dakota, Grand Forks , North Dakota, Nicole Della Longa, MA, University of North Dakota, Grand Forks, North Dakota, Erica Goodman, BS, University of North Dakota, Grand Forks, North Dakota, Alexandra Thiel MA, University of Wyoming, Laramie, Wyoming, Alexis O'Halloran, Student, University of North Dakota, Grand Forks, North Dakota, Alex Karie, Student, University of North Dakota, Grand Forks, North Dakota

Eating more rapidly than normal is a common feature of binge-eating episodes. We sought to test two factors – one situational and one dispositional – that might impact eating rate to understand what puts individuals at risk of eating rapidly, and perhaps binge eating. We hypothesized that eating while allocating attentional resources elsewhere would lead to an increased consumption rate. We also hypothesized that this would especially be the case for individuals prone to losing restraint over their eating (i.e., those who are disinhibited). Thirty-eight individuals (67% women; 42% reporting past-month binge eating) participated and completed measures, including the Three-Factor Eating Questionnaire Disinhibition subscale, at baseline. They then were instructed to drink for five minutes at a consistent rate from a large cup containing 32 oz. (960 kcals) of Boost meal replacement shake while performing a computerized attention task (Multi-Attribute Task Battery-II), which varied in difficulty (low, medium, high; order counterbalanced across participants). Their consumption rate was measured twice every second with a hidden scale. A mixed-effects linear model indicated the presence of a three-way interaction (attention task difficulty X disinhibition X time; p<.001) while controlling for sex and BMI. Under more attention-demanding conditions, participants drank the shake faster; this difference was especially strong for those higher in disinhibition. Results support the hypothesis that individuals prone to losing restraint over their eating eat faster when their attention is allocated elsewhere. This study highlights the importance of attention allocation and behavioral tendencies in determining eating rate, which is critical to understanding the mechanics of binge-eating episodes. Future research should examine whether the perception of loss of control is in part influenced by consumption rate, connecting objective indices of eating with subjective experience.

7.5: Model Behavior: How the Interpersonal (IPT) Model Predicts Disordered Eating in a College Longitudinal Study

Kerstin Blomquist, PhD, Furman University, Greenville, South Carolina, Erin Jackson, Student, Furman University, Greenville, South Carolina

The Interpersonal (IPT) Model of binge eating proposes that interpersonal stressors lead to either low self-esteem or negative affect, which then triggers disordered eating. Clinically-significant eating disorders typically develop in late adolescence (college years), with 26% of college women and 10% of college men reporting disordered eating and 44% engaging in binge eating. In addition, new and unique interpersonal stressors arise in college. However, studies testing the IPT model have been limited to predominantly non-college-aged, female samples and cross-sectional designs. To address this gap, the current study tested the IPT model in a longitudinal sample of undergraduate men and women (N=245). At 3 time points during their first year of college, participants completed self-report measures of interpersonal stress, self-esteem, negative affect, eating loss of control, and disordered eating. Using Hayes’ PROCESS macro to test for mediation, results revealed that self-esteem significantly mediated the relationship between interpersonal stress and eating LOC severity (Z=3.09, p=.002, K2=.10) and disordered eating (Z=2.10, p=.036, K2=.053). Likewise, negative affect significantly mediated the relationship between interpersonal stress and eating LOC severity (Z=2.91, p=.0036, K2=0.010), eating LOC frequency (Z=2.81, p=.0049, K2=.091), and disordered eating behaviors (Z=3.43, p=.0006, K2=.13). Self-esteem did not significantly mediate the relationship between interpersonal stress and eating LOC frequency. Our findings provide longitudinal support for the IPT model in a mixed gender, undergraduate sample, suggesting that targeting interpersonal stressors in college may be a useful avenue for eating disorder prevention.

7.6: Insecure Attachment and Early Maladaptive Schema in Disordered eating: The Mediating Role of Rejection Sensitivity

Tara De Paoli, Student; BPsych(Hons), The University of Melbourne, Melbourne, Australia, Isabel Krug, PhD, The University of Melbourne, Melbourne, Australia, AsPr. Matthew Fuller-Tyszkiewicz, Deakin Univesity, Melbourne, Victoria, Australia

The current study aimed to assess insecure attachment and the early maladaptive schema domain of disconnection and rejection in the context of disordered eating, and offer rejection sensitivity as a mediator of this relationship. The sample consisted of 108 female participants with a current or lifetime eating disorder (ED) diagnosis (age M=25 years) and 508 female healthy control participants (age M=21 years). Participants were asked to complete a number of self-report measures related to insecure attachment (anxious and avoidant), maladaptive schema (emotional deprivation, abandonment, mistrust, social isolation, and defectiveness), rejection sensitivity (interpersonal and appearance-based), and disordered eating behaviours. Invariance testing conducted between the ED and the healthy control groups indicated that the model was structurally variant (i.e. different between groups). Path analysis indicated that the overall model demonstrated good fit (RMSEA=0.054, CFI=0.988, TLI=0.979). For both the ED and the healthy control groups, attachment anxiety, abandonment schema, interpersonal rejection sensitivity, and appearance-based rejection sensitivity were directly associated with disordered eating (p<.05). However, indirect effects indicated differences between groups. For the ED group, anxious attachment was associated with disordered eating through multiple pathways involving emotional deprivation schema, abandonment schema, interpersonal rejection sensitivity, and appearance-based rejection sensitivity. There was also an indirect effect for emotional deprivation schema on disordered eating through appearance-based rejection sensitivity (β=.068, p<.05). For the control group, indirect effects were found for emotional deprivation schema on disordered eating through interpersonal rejection sensitivity (β=.245, p<.05) and appearance-based rejection sensitivity (β=.038, p<.05). The results supported the hypotheses, indicating that both interpersonal and appearance-based rejection sensitivity are important mediators for the relationships between insecure attachment, maladaptive schema, and disordered eating. These findings may inform treatments targeting interpersonal functioning for those presenting with disordered eating.

Topic: Comorbidity and Risk Factors for ED
Meeting Hall IV, second floor
Anja Hilbert, PhD & Stephen Wonderlich, PhD, FAED

8.1: Examining the Relationships Between Compulsive Exercise, Quality of Life and Psychological Distress in Adults with Anorexia Nervosa

Sarah Young, DClinPsy/PhD candidate, University of Sydney, Sydney, New South Wales, Stephen Touyz, MPsych; PhD; FAED University of Sydney, Sydney, CV4 7AL,, Caroline Meyer, PhD, The University of Warwick, Coventry, CV4 7AL,, Jon Arcelus, MD; PhD, University of Nottingham, Nottingham , NG1 5BH, Paul Rhodes, PhD, University of Sydney, Sydney, New South Wales, Sloane Madden, MBBS; PhD; FAED, Sydney Children's Hospital Network, Sydney, New South Wales, Kathleen Pike MS; PhD; FAED, Columbia University, New York, New York, Evelyn Attia, MD; FAED, Columbia University, New York, New York, Ross Crosby, PhD; FAED, Neuropsychiatric Research Institute, Fargo, North Dakota, Phillipa Hay, DPhil; FAED, Western Sydney University, Sydney, New South Wales

Compulsive exercise in patients with anorexia nervosa (AN) can be performed to avoid or manage negative psychological symptoms such as anxiety, distress and low mood. Previous research in community samples has shown that exercise driven by shape or weight concerns, and/or to avoid guilt is associated with severity of eating disorder symptoms and poorer quality of life. The current study assessed the relationships between compulsive exercise, quality of life and psychological distress in a sample of outpatients with AN. Participants were 78 adults with AN, enrolled in the multi-site randomized controlled trial “Taking a LEAP forward in the treatment of anorexia nervosa”. At baseline and across treatment, participants completed the Eating Disorder Examination-Questionnaire (EDE-Q), Compulsive Exercise Test (CET), Short Form-12 Health Status Questionnaire (SF-12), Eating Disorder Quality of Life (EDQoL) scale, Kessler-10 item distress scale (K-10), Padua Inventory, and the Anorexia Nervosa Stages of Change Questionnaire (ANSOCQ). Baseline correlations demonstrated that compulsive exercise was associated with greater eating disorder psychopathology and higher psychological distress, but poorer quality of life. Regression analyses across treatment found that higher levels of compulsive exercise at baseline directly predicted lower motivation to change after 20 sessions of treatment. The change in levels of compulsive exercise (on the CET) from sessions 1-10 predicted changes in mental-health quality of life and eating disorder quality of life after 10 sessions. Further research is required to ascertain which aspects of treatment can be most effective in improving quality of life and motivation to change outcomes for patients who exercise compulsively.

8.2: Compensatory Behaviors among a Racially Diverse Sample of Undergraduate Women

Amanda Bruening, BA; MA, Arizona State University, Tempe, Arizona, Marisol Perez, PhD, Arizona State University, Tempe, Arizona  

Research investigating the use of compensatory behaviors is largely inconsistent among racial minorities. Early studies among Blacks and Hispanics demonstrated that individuals were less concerned about their weight, engaged in less exercise, and were typically heavier, thereby supporting the notion that these groups were less at risk for eating disorders. However, more recent research has found that rates of bulimia nervosa and binge eating disorder are comparable to Non-Hispanic Whites. Despite these contrary findings, only one study has examined the differences in eating pathology among racial minorities in a community sample. Though norms for undergraduate women have been published using the Eating Disorder Examination Questionnaire (EDE-Q), which is often regarded as the gold-standard self-report measure for assessing compensatory behaviors, the majority of the sample (88%) were White. No study to date has attempted to establish norms for minority college women. Given the increasing rates of eating pathology on college campuses in tandem with the rising number of minority women attending college, it is of critical importance to examine how rates of various compensatory behaviors differ across racial groups among undergraduate women. A benchmarking analysis will be performed on a sample of approximately 2,910 undergraduate women. Similar to other studies, the sample recruited was predominantly White (n = 1696) followed by Hispanic/Latina (n = 524), Asian (n = 480), African American (n = 181), and Native American (n = 48). As there is limited knowledge of compensatory behaviors among Native American women, this subgroup was included despite the low sample size. Results of the current study extend not only our knowledge of how compensatory behaviors differ across various racial groups, including those overlooked in the current literature, but also the clinical utility of the EDE-Q.

8.3: The Role of the 5-HTTLPR VNTR in Moderating Psychosocial Risks for Disordered Eating Pathology in Adolescence: Findings from the Australian Temperament Project (Est. 1983).

Vanja Rozenblat, BA; MPsych, PhD Candidate, The University of Melbourne, Parkville, VIC, Eleanor Wertheim, PhD, La Trobe University, Bundoora, VIC, Ross King, PhD, Deakin University, Burwood, VIC, Isabel Krug, PhD, The University of Melbourne, Parkville , VIC, The ATP Consortium, PhD, The Australian Temperament Project, Royal Children's Hospital, Parkville, VIC

The purpose of this study was to investigate whether sexual abuse, physical abuse, depression and emotional control interacted with a functional variable number tandem repeat (VNTR) in the promoter region of the serotonin transporter gene (5-HTTLPR) to predict disordered eating symptoms. The sample included 672 participants (49.8% female) from the Australian Temperament Project (ATP), a 33-year longitudinal study of social-emotional development. At age 15-16 years, participants completed the Bulimia and Drive for Thinness scales of the EDI-2, the Short Mood and Feelings Questionnaire, and an ATP-devised emotional control scale. At age 23-24 years, 481 of the original 672 participants retrospectively responded to four questions regarding childhood sexual abuse and parental physical punishment. We observed no association between disordered eating and sexual abuse, or disordered eating and a mild-to-moderate form of parental punishment. We did, however, observe associations with depression, emotional control, and severe parental punishment (where effects lasted beyond one day). There was also some evidence of interaction between 5-HTTLPR and severe parental physical punishment in predicting bulimic symptoms (p = .017). This effect persisted after controlling for gender, BMI and ethnicity, but did not meet the Bonferroni adjusted threshold (p < .005). Findings suggest that the 5-HTTLPR is not involved in risk for disordered eating; however, type and severity of exposure are likely to be important prognostic indicators of adolescent disordered eating.

8.4: What Does gender have to do with it? Associations Between Bender-Linked Personality Traits and Eating Pathology

Vivienne Hazzard, MPH; RD; Student, University of Michigan School of Public Health, Ann Arbor, Michigan, Kendrin Sonneville, RD; ScD, University of Michigan School of Public Health, Ann Arbor, Michigan

The purpose of this study was to examine relationships between gender-linked personality traits (instrumentality and expressivity) and eating pathology, as well as to investigate depressive symptoms as a potential mechanism to explain such relationships. Data for this study were collected in Wave III of the National Longitudinal Study of Adolescent to Adult Health (Add Health) from a subsample of respondents (n=3,737) who completed the short form Bem Sex Role Inventory (BSRI). Using median split scoring for instrumental (“masculine”) and expressive (“feminine”) trait scales as measured by the BSRI, respondents were categorized as undifferentiated (low in both), masculine (high instrumentality), feminine (high expressivity), or androgynous (high in both). Self-reported outcomes assessed were disinhibited eating (overeating and/or loss of control eating), purging (vomiting and/or using laxatives) to control weight, and lifetime eating disorder (ED) diagnosis. Depressive symptoms were assessed by nine questions from the Center for Epidemiologic Studies Depression Scale (CES-D). Sex-stratified logistic regression analyses were run, accounting for survey design and adjusting for age, race/ethnicity, income, and parental education. In young adulthood (mean age=22.5 years, SD=0.12), prevalence of disinhibited eating was 8.7% in females and 5.9% in males, prevalence of purging was 0.9% in females and 0.0% in males, and prevalence of lifetime ED diagnosis was 4.5% in females and 0.2% in males. Compared to females categorized as undifferentiated, females categorized as masculine had lower odds of disinhibited eating (OR: 0.36; 95% CI: 0.15, 0.85), and females categorized as androgynous had lower odds of having ever been diagnosed with an ED (OR: 0.38; 95% CI: 0.15, 0.99). After adjusting for depressive symptoms, the association between masculinity and disinhibited eating remained significant, but the association between androgyny and ED diagnosis did not.

8.5: When Grit Goes Bad: The Interaction of Autism Symptoms and Grittiness in the Prediction of Eating Disorder Symptoms

Elizabeth Velkoff, BA, Miami University, Oxford, Ohio, Christopher Hagan, MA, Florida State University, Tallahassee, Florida, April Smith, PhD, Miami University, Oxford, Ohio

The purpose of the present study was to examine whether symptoms of autism spectrum disorder interact with grit (the ability to sustain interest and effort on long-term goals) to predict eating disorder (ED) symptoms. Previous research indicates that individuals with anorexia nervosa (AN) show a number of similarities to those with autism spectrum disorders, particularly in terms of cognitive functioning. One characteristic of autism spectrum disorder is a pattern of routines, rituals, and need for sameness. This is reflected in difficulties with set-shifting, or the ability to move from one task to ana deficit also common among individuals with AN. It may be the case that this tendency toward sameness may manifest as grit, allowing individuals with autistic traits to pursue goals with sustained effort. However, when one of those goals is severe weight loss, as in AN, or other disordered eating behaviors, grit may serve a less adaptive purpose. The present study tested the hypothesis that autism spectrum symptoms would interact with grit to predict ED symptoms. Specifically, we hypothesized that among individuals with high levels of grit, elevated autism spectrum symptoms would be associated with more severe ED symptoms. Participants (N = 140) were recruited online and through ED treatment centers across the United States and completed self-report questionnaires assessing for autism spectrum symptoms, grit, and ED symptoms. Results indicated that there was a significant interaction (p = 0.055) between grit and autism spectrum symptoms in predicting ED symptoms. Probing this interaction revealed that at high (p = 0.045) but not low (p = 0.386) levels of grit, greater autism spectrum symptoms are associated with more self-reported ED symptoms. Replication results from four independent samples will also be discussed. These findings point to a potential individual difference factor by which autism spectrum disorders are related to EDs, specifically through elevations in grit. Specifically, these findings may indicate that grit, a trait which serves a positive function in many contexts by facilitating the pursuit of long-term goals, may go awry in some cases, facilitating the dangerous weight loss characteristic of AN.

8.6:The risk of Eating Disorders comorbid with ADHD: A systematic review and meta analysis

Bruno Nazar, MSc; MD; PhD, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, Joseph Sergeant, PhD; Professor, Vrjie University, Amsterdam, Rio de Janeiro, Paulo Mattos, MD; Professor, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, Janet Treasure, PhD; FAED, Professor, King's College London, London, UK

There has been interest in whether people with Attention-Deficit/Hyperactivity Disorder (ADHD) are at higher risk of developing an Eating Disorder (ED). The aim of this study was estimate the size of this association with a meta-analysis of studies.We retrieved studies following PRISMA guidelines from a broad range of databases.Twelve studies fitted our primary aim in investigating ED in ADHD populations (ADHD = 4.013/Controls = 29.404), and 5 exploring ADHD in ED populations (ED = 1044/Controls = 11292). The pooled odds ratio of diagnosing any ED in ADHD was increased significantly, 3. 82 (95% CI:2.34-6.24). A similar level of risk was found across all ED syndromes [Anorexia Nervosa = 4.28 (95%CI:2.24-8.16); Bulimia Nervosa = 5.71 (95%CI: 3.56-9.16) and Binge Eating Disorder = 4.13 (95%CI: 3-5.67)].The risk was significantly higher if ADHD was diagnosed using a clinical interview [5.89 (95%CI:4.32-8.04)] rather than a self-report instrument [2.23(1.23-4.03)]. The pooled odds ratio of diagnosing ADHD in participants with ED was significantly increased, 2.57 (95%CI:1.30-5.11). Subgroup analysis of cohorts with binge eating only yielded a risk of 5.77 (95%CI:2.35-14.18). None of the variables examined in meta-regression procedures explained the variance in effect size between studies.People with ADHD have a higher risk of comorbidity with an ED and people with an ED also have higher levels of comorbidity with ADHD. Future studies should address if patients with this comorbidity have a different prognosis, course and treatment response when compared to patients with either disorder alone.

Topic: Epidemiology
Meeting Hall V, second floor
Bryn Austin, ScD, FAED & Anna Keski-Rahkonen, MD, PhD, MPH

9.1: A 30-Year Longitudinal Study of Body Weight, Dieting, and Disordered Eating Symptoms

Tiffany Brown, PhD, University of California, San Diego, San Diego, California, K. Jean Forney, MS, Florida State Unversity, Tallahassee, Florida, Kelly Klein, MS, Florida State University, Tallahassee, Florida, Charlotte Grillot, MA, Florida State University, Tallahassee , Florida, Todd Heatherton, PhD, Dartmouth College, Hanover, New Hampshire, Pamela Keel, PhD, Florida State University, Tallahassee, Florida

Etiological models for eating disorders are, in part, derived from sex differences found in epidemiological patterns. As such, evaluating sex differences in the natural course of eating pathology and posited risk factors across the lifespan is necessary to fully test these models. Nine-hundred men and women from a northeastern university completed surveys on eating attitudes and behaviors at 10-year intervals from late adolescence (Mean (SD) age = 20 [2] years) to later-adulthood (Mean (SD) age = 50 [2] years). Associations between posited risk factors and disordered eating over time were analyzed using multilevel modeling. DSM-5 eating disorder diagnoses decreased over the 30-year span for both sexes, with no significant sex difference in point prevalence at age 50 (men: 3.9%, women: 7.7% for any eating disorder, χ2(1) = 2.95, p = .09). Body mass index increased over time in both sexes (p < .001). For men, weight perception, dieting frequency, and drive for thinness increased over the 30-year span (all p-values <.001). In contrast, for women, weight perception and drive for thinness decreased over time, and dieting frequency reached a nadir in midlife, with an increase at age 50 (all p-values <.001). The relationship between dieting and disordered eating grew weaker over time for both sexes, suggesting that increased dieting around age 50 may reflect a healthy response to age-associated weight gain, as opposed to a pathological response to body image disturbance. Notably, eating disorder diagnoses remain prevalent in later adulthood for both sexes. Results imply that current risk models require refinement to account for developmental trajectories in which dramatic sex differences observed in late adolescence diminish over time.

9.2: Maternal Eating Disorders and Perinatal Outcomes: A Three-Generation Study in the Norwegian Mother and Child Cohort Study

Hunna Watson, PhD, MPsych; MSc; University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Stephanie Zerwas, PhD, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, -, Leila Torgersen, PhD, Norwegian Institute of Public Health, Oslo, -, Kristin Gustavson, PhD, Norwegian Institute of Public Health, Oslo , Elizabeth Diemer, PhD, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, Gun Peggy Knudsen, PhD, Norwegian Institute of Public Health, Oslo, -, Ted Reichborn-Kjennerud MD; PhD, Norwegian Institute of Public Health, Oslo

An existing body of research indicates that mothers with a history of eating disorders are at risk of adverse birth outcomes. In turn, adverse birth outcomes have been associated with a higher likelihood of eating disorder onset in adult offspring, possibly reflecting a transgenerational cycle of risk. Previous studies of the relationship between maternal eating disorders and adverse perinatal outcomes have failed to control for familial transmission of perinatal event phenotypes, which may confound the association. In a unique design afforded by the Norwegian Mother and Child Cohort Study (MoBa) and Medical Birth Registry of Norway, we linked three generations through birth register records and maternal-reported survey data. The aim was to determine if maternal eating disorders increase risk after parsing out the contribution of familial transmission of perinatal events. The samples were 70,881 grandmother-mother-child triads for analyses concerning eating disorder exposure during pregnancy and 52,348 for analyses concerning lifetime maternal eating disorder exposure. As hypothesized, eating disorders predicted a higher incidence of perinatal complications even after adjusting for grandmaternal perinatal events. For example, anorexia nervosa immediately prior to pregnancy was associated with smaller birth length (relative risk = 1.62, 95% confidence interval = 1.21, 2.15), bulimia nervosa with induced labor (1.21; 1.07, 1.37), and binge-eating disorder with several delivery complications, larger birth length (1.26; 1.17, 1.35), and large-for-gestational-age (1.04; 1.02, 1.06). Maternal pregravid body mass index and gestational weight mediated some associations. Exposure to eating disorders increases the risk of adverse birth outcomes, independent of familial transmission of perinatal events.

9.3: Other Specified and Unspecified Feeding or Eating Disorders Among Women in the Community

Anna Keski-Rahkonen, MD, MPH, PhD, University of Helsinki, Helsinki, Finland, Linda Mustein, MD, MPH, PhD, University of Helsinki, Helsinki, Finland

Our goal was to examine the occurrence, course, and clinical picture of the DSM-5 residual categories 'Other Specified Feeding or Eating Disorder' (OSFED) and 'Unspecified Feeding or Eating Disorder' (UFED), to describe potential subtypes, and to evaluate whether the subdivision of the residual category appears meaningful. We screened women from the 1975-79 birth cohorts of Finnish twins (N = 2825) for lifetime eating disorders using questionnaires and the SCID interview. This analysis characterizes women who reported clinically significant eating disorder symptoms but did not fulfill diagnostic criteria for DSM-5 anorexia nervosa (AN), bulimia nervosa (BN), or binge eating disorder (BED). Thirty-eight women (21% of those with an eating disorder) fell in the residual OSFED/UFED category. A third of them (N = 14) fulfilled OSFED criteria, whereas two thirds (N = 24) fell in the UFED category. The lifetime prevalence of OSFED/UFED was 1.5% (95% CI 1.1-2.0%), less than half of the prevalence of DSM-IV eating disorder not otherwise specified (EDNOS). The mean age of onset of OSFED/UFED was 18 years, median duration of symptoms was two years, and the 5-year probability of recovery was 60%. Over a third of women with OSFED/UFED suffered from comorbid psychiatric disorders. Both residual categories were clinically heterogeneous and included atypical forms of AN, BN, and BED. In conclusion, applying the DSM-5 criteria in a community sample of young women more than halved the occurrence of residual eating disorder diagnoses, but resulted in two instead of one clinically heterogeneous residual categories. Nevertheless, residual eating disorders were associated with considerable clinical severity.

9.4: Disordered Eating Pre- and Postpartum: From Epidemiological Research to Implementation of Internet-Based Psycho-Education

David Clinton, PhD, Karolinska Institutet, Stockholm, Stockholm, Cecilia Brundin Pettersson, BSc, Karolinska Institutet, Stockholm

Eating disorders pre- and postpartum can have considerable negative consequences, such as low birth weight, prematurity, miscarriage, increased risk for caesarean section, postnatal depression, feeding problems and interactional difficulties. The present project aimed to: 1) investigate symptoms of disordered eating pre- and postpartum using a standardised measure of eating disorder psychopathology; 2) design and implement an internet-based programme of psycho-education on disordered eating during and after pregnancy for relevant health care workers. A consecutive series of women attending either pre- (N = 426) or postnatal (N = 345) clinics in metropolitan Stockholm were assessed using the Eating Disorder Examination Questionnaire (EDE-Q). Assessments were conducted at either the first visit to prenatal clinics (10–12 weeks of pregnancy) or 6 to 8 months postpartum. Using an optimised version of the EDE-Q with 14 items and a cut-off score of ≥2.8, it was estimated that 5.3 % of prepartum and 12.8 % of postpartum mothers were suffering from clinical eating disorders. Seriously disordered eating behaviour during, and especially after, pregnancy may be more common than previously thought. In order to help frontline healthcare services to deal with the challenges of seriously disordered eating pre- and postpartum, an interactive internet-based programme of psycho-education was developed. The programme teaches participants about eating disorders, how to identify and talk about eating problems, as well as how to help women find appropriate specialist services. The programme uses text-based learning tools, video, illustrative fictional cases, quizzes and the use of a reflective diary to help participants learn and develop appropriate skills. Experiences of extending empirical research to service development is discussed, and preliminary results from the implementation of the programme are presented.

9.5: The Epidemiology of Eating Disorder Risk and DSM-5 Eating Disorders among Austrian Adolescents: Results from the Mental Health in Austrian Teenagers (MHAT) Study

Michael Zeiler, Mag., Medical University of Vienna, Department for Child and Adolescent Psychiatry, Vienna, Austria, Julia Philipp, Dr., Medical University of Vienna, Vienna, Austria, Stefanie Truttmann, Mag., Medical University of Vienna, Vienna, Austria, Gudrun Wagner, Dr., Medical University of Vienna, Vienna , Austria, Andreas Karwautz, FAED; Prof. Dr., Medical University of Vienna, Austria, Karin Waldherr, Prof.(FH) Dr., FernFH Distance Learning University of Applied Sciences, Wr. Neustadt, Lower Austria

The Mental Health in Austrian Teenagers (MHAT) study is the first large and representative epidemiological study on mental health problems and psychiatric disorders among adolescents in Austria. One aim was to assess the prevalence of disordered eating behaviors and full-syndrome eating disorders based on DSM 5 diagnostic criteria. A two-stage design was applied. In the first stage (screening stage) the eating disorders risk was assessed by the SCOFF questionnaire in a sample of 3610 students aged 10-18 years during school lessons. In a second stage (interview stage), adolescents with an elevated risk for mental health disorders as well as a sample of low-risk adolescents were interviewed by clinical psychologists to assess DSM 5 psychiatric disorders including eating and feeding disorders. Additional screening and interview data from unemployed adolescents and adolescents in inpatient care at Austrian child and adolescents psychiatric institutions were obtained. About one third of girls and 15% of boys showed an elevated risk for eating disorders. A high BMI, low socioeconomic status, burdensome events in the life course as well as chronic somatic and mental illnesses of parents and near relatives turned out as relevant risk factors. The point-prevalence of any eating or feeding disorder was 0.5%, the life-time prevalence was 2.6%. Correcting for adolescents not recruited via schools, the prevalence increased to 1.6% and 3.7% resp. Full-syndrome eating disorders were characterized by high comorbidity (57.9%) with other psychiatric disorders (especially depression, anxiety disorders and non-suicidal self-injury). The mental health service use of adolescents with a full-syndrome eating disorder was one of the lowest among all psychiatric disorders (18.8%). Although the prevalence of full-syndrome eating disorders was low, this study revealed a much higher proportion of subclinical symptoms and therefore the need for targeted prevention in this field.

9.6: Nationwide Survey on Patients with Eating Disorders in Hospitals in Japan.

Teysuya Aando, MD; PhD, National Institute of Mental Health, National Center of Neurology and Psychiatry, Kodaira, Tokyo, Hiroe Kikuchi, MD; PhD, National Institute of Mental Health, National Center of Neurology and Psychiatry, Kodaira, Tokyo, Norito Kawakami, MD; PhD, Graduate School of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo

The nationwide survey on patients with eating disorders (ED) in hospitals in Japan has not been conducted since 1998, which means a lack of epidemiological information on current status of ED patients in Japan. The present survey had two objectives. The first was to estimate the number of ED patients in hospitals. The second was to investigate the demographics and clinical characteristics of the patients. Targeted medical facilities were clinical departments of hospitals which were practicing psychiatry, psychosomatic medicine, internal medicine, pediatrics, and gynecology. From the relevant 11,766 facilities, 5220 were extracted by stratified sampling based on the scale of the hospital. The survey used a two-stage design. The first questionnaire asked the facilities to report the number of patient who appeared or were admitted between October 1, 2014 to September 30, 2015 of each sex and the DSM5 diagnosis. Then, the second questionnaire that inquires clinical information of the individual patient was sent to the facility which reported a patient in the first questionnaire. Effective responses were obtained from 2561 facilities (49.0%). The number of each diagnosis was estimated as follows: Anorexia nervosa, 12,667 (95%CI: 10,611-14,723); bulimia nervosa, 4,606 (3,133-6,078); binge eating disorder, 1,145 (833-1,457); other specified feeding or eating disorder, 2,447 (1,482-3,411); unknown diagnosis, 3,634 (2,278-4,990). The number of ED patients (total 24,498) estimated by this study was the same level as the number (total 23,200) that had been reported in the previous study in 1998. Because much fewer patients have been detected in clinical-based surveys, including the current study, than predicted by the school-based investigations, it is suggested that large number of ED patients are left untreated. The demographic and clinical information will be investigated in the second stage of the survey.

Topic: Prevention
North Hall, second floor
Melissa Atkinson, PhD & Susan Paxton, PhD, FAED

10.1: Advanced Data Integration for Epidemiologic Modeling to Evaluate Policy Approaches to Eating Disorder Prevention

Michael W. Long, ScD, George Washington University, Washington, DC, District of Columbia, Xindy C. Hu, MPH, Harvard Chan School of Public Health, Boston, Massachusetts, Davene R. Wright, PhD, Seattle Children's Hospital, Seattle, Washington, S. Bryn Austin, ScD, Boston Children's Hospital, Boston , Massachusetts

Scaling up eating disorders (ED) prevention strategies for population impact will require a new and concerted focus on policy translation research. Evaluating policies supporting public health prevention requires accurate and up-to-date epidemiologic modeling of the disease course, which currently does not exist. We develop and calibrate an Eating Disorder Microsimulation (EDM) model to conduct economic evaluations of prevention strategies. Each individual in the model is followed over a 25-year period as they transition among several health states annually: ED-free, anorexia nervosa, bulimia nervosa, binge eating disorder, other specified feeding or eating disorder, and dead. Data on diagnoses, outpatient, and inpatient charges were obtained from 10,782 patients and 67,119 patient visits provided by six US pediatric hospitals in the PEDSnet research collaborative, one of the largest in the nation. We model duration and expense of outpatient treatment, risk for hospitalization by treatment and patient characteristics, and transition between ED types. We calibrate the model to gold-standard data such as the Global Burden of Disease 2013, the US National Comorbidity Study-Adolescent Supplement and the US Healthcare Cost and Utilization Project Nationwide Inpatient Sample. After calibration, cumulative 25-year incidence and prevalence projections were similar between the EDM and the most recent epidemiologic evidence. One early application of our modeling approach has shown that secondary ED prevention would be as cost-effective as other well-accepted screening programs at willingness-to-pay thresholds of $50,000 and $100,000 per quality-adjusted life year gained. The same approach will be applied to several other prevention strategies in a comparative cost-effectiveness analysis framework. The presentation has broad applicability to the development of country-specific prediction models to evaluate the cost-effectiveness of proposed ED prevention strategies.

10.2: A Preliminary Investigation of REbeL: A Dissonance-based Program to Promote Positive Body Image, Healthy Eating, and Empowerment in Teens

Laura Eickman, PsyD, REbeL, Inc., Overland Park, Kansas, Jessica Betts, MS; RD, REbeL, Inc., Overland Park, Kansas, Lauren Pollack, MA, University of Missouri-Kansas City, Kansas City, Missouri, Frances Bozsik, MS, University of Missouri-Kansas City, Kansas City , Missouri, Brooke Guiot, Student, University of Missouri-Kansas City and REbeL, Inc., Kansas City, Missouri, Marshall Beauchamp, MS, University of Missouri-Kansas City, Kansas City, Missouri, Jenny Lundgren PhD; FAED, University of Missouri-Kansas City, Kansas City, Missouri

Short-term outcomes associated with participation in REbeL, a peer-led dissonance-based eating disorder prevention program for high school students, were evaluated. REbeL promotes healthy body image and eating behaviors/attitudes through social media and school-based activities that target body acceptance, mindful eating and exercise, healthy self-esteem and empowerment, and reduction of weight stigma. Four Kansas City regional high schools were randomized to REbeL or an assessment-only wait list control condition during the 2015-2016 academic year. One control school withdrew participation after learning of their randomization as a control. Seventy-one students enrolled in the study (REbeL schools N = 48; Control N = 23) and were assessed on measures of eating attitudes/behaviors, body image, weight bias, self-esteem, empowerment, and mood at the beginning of the school year; 37 REbeL students and 20 control students completed assessments at the end of the school year. Mixed effects GLM, controlling for baseline scores and nested across school and time, compared groups on outcomes at the end of the academic year. Pairwise comparisons with Bonferroni correction demonstrated that, in comparison to controls, students in both REbeL schools had statistically significantly lower scores on the Body Checking Questionnaire (adjusted means: 44.3 and 51.5 [REbeL] vs. 68.5 [control]) and the Eating Disorder Examination-Questionnaire (adjusted global mean: 1.1 and 1.2 [REbeL] vs. 2.5 [control]); all ps

10.3: Perceived Barriers and Facilitators Towards Seeking help for Eating Disorders: A Systematic Review

Laura Hart, PhD, La Trobe University, Melbourne, Victoria, Susan Paxton, PhD, La Trobe University, Melbourne, Victoria, Agus Salim, PhD, La Trobe University, Melbourne, Victoria

Parents provide an important environment in which young children develop body image attitudes and eating patterns. However, there are few evaluated resources design to to support parents. Confident Body, Confident Child (CBCC) is a new resource to assist parents in fostering body confidence and healthy eating patterns in 2-6 year-old children. The booklets, website, poster and information session were developed from research on child risk factors for body dissatisfaction and disordered eating and a Delphi expert consensus study. CBCC was evaluated using a four arm Randomized Controlled Trial; A) CBCC resource + face-to-face information session, B) CBCC resource only, C) Nutrition resource only and D) waitlist control. 345 parent participants completed online self-report measures of parenting variables (knowledge, behavioural intentions, family meals and parental feeding practices) at baseline, 6-weeks post-intervention, six and twelve months later. ANCOVA analyses using multiple imputation for missing data revealed that, after controlling for baseline scores, there were significant differences between groups at 6-months and that many of these positive changes were still apparent at twelve month follow-up. The family meal time scales, child BMIz and amount of resource read, showed no significant differences between groups at six months. Parental knowledge, positive and negative parenting intentions, and the parental feeding practices of instrumental feeding and weight-based restriction, showed sustained improvement in the two CBCC groups at six and twelve months. Parents receiving the information session and resources appeared to perform better than parents receiving the resource only, suggesting face-to-face sessions are a superior method for dissemination. Although some intervention effects were no longer apparent at 12-months, this evaluation revealed that CBCC had a positive and enduring impact on the knowledge, intentions and feeding practices of parents in our sample. Plans for the future dissemination of CBCC and how to access the resources are outlined.

10.4: A Systematic Review of the Existing Models of Disordered Eating: Do They Inform the Development of Effective Interventions?

Jamie-Lee Pennesi, BA, Flinders University, Adelaide, South Australia, Tracey, D. Wade, PhD, Flinders University, Adelaide

Despite significant advances in the development of prevention and treatment interventions for eating disorders and disordered eating over the last decade, there still remains a need to develop more effective interventions. In line with the 2008 Medical Research Council (MRC) evaluation framework from the United Kingdom for the development and evaluation of complex interventions, the development of sound theory is a necessary precursor to the development of effective interventions. Thus the aim of the current review is to identify the existing models for disordered eating and identify those models which have helped inform the development of prevention and/or treatment interventions for disordered eating. A literature search was conducted by using the PsycINFO database (OvidSP). Keywords anorexia nervosa (Title) OR bulimia nervosa (Title) OR disordered eating (Title) OR eating disorders (Title) OR bulimic (Title) OR eating (Title) AND model (Title) OR theory (Title) were used to locate pertinent publications in all journals using an advanced search. Publications were then inspected for studies meeting the clearly defined inclusion criteria. While an extensive range of theoretical models for the development of disordered eating were identified, only a few models have led to the development of effective interventions. Of the fifty-four models described in the literature, only ten (18.5%) had progressed beyond mere description and on to the development of interventions that have been evaluated. This review will add important insights to the eating disorder prevention literature and help to inform the development of effective approaches to prevention.

10.5: Efficacy of an Internet-Based Prevention Program in a Female Population with Subclinical Anorexia Nervosa

Kristian Hütter, DiplPsych, Technische Universität, Dresden, Sachsen, Corinna Jacobi, DiplPsych; Dr. rer. biol. hum., Technische Universität Dresden, Dresden, Sachsen, Bianka Vollert, DiplPsych, Technische Universität Dresden, Dresden, Sachsen, Paula von Bloh, DiplPsych, Technische Universität Dresden, Dresden , Sachsen, Nadine Eiterich, DiplPsych, Technische Universität Dresden, Dresden, Sachsen, Denise Wilfley, PhD, Washington University School of Medicine, St. Louis, Missouri, C. Barr Taylor MD, Stanford School of Medicine, Stanford, California

Despite the urgent need for early interventions, no targeted (indicated) prevention programs for women at risk for anorexia nervosa are available. We developed a internet-based prevention program (Student Bodies-AN) specifically targeting this risk group. Following a pilot study with promising results, the objective of this randomized controlled trial was to determine the efficacy of Student Bodies-AN. The internet-based CBT-prevention program Student Bodies-AN consists of 10 weekly sessions and a booster session and includes psycho-educative and interactive components moderated by trained clinical psychologists. Woman aged 18 and above with either low body weight and high weight and shape concerns or normal body weight, high weight and shape concerns and high restrained eating were recruited at universities in three German cities as well as via an online version of the screening questionnaire. A total of N=168 participants were randomized to the intervention or a waitlist control condition. Primary outcomes for all participants are clinically significant changes in attitudes and behaviors of disordered eating and changes in BMI in the underweight group of participants at 12-month follow-up. In addition, specific symptoms of disordered eating (e.g., binge eating and compensatory behaviors) and associated psychopathology (e.g., depression) were assessed as secondary outcomes. At 12-month follow-up ITT analysis (mixed models) have shown a BMI increase for underweight participants (large effect size) and small to moderate effects for specific attitudes and behaviours of disordered eating (e.g. drive for thinness, restrained eating). Further analysis (e.g. binary outcomes of clinical relevance, moderators / mediators) are in progress.

10.6: Acceptability of an Online Mindfulness Intervention for Reducing Eating Disorder Risk Factors: Results from a Randomized Controlled Pilot Study in Young Adult Women

Melissa Atkinson, PhD, University of the West of England, Bristol, Avon, Philippa Diedrichs, PhD, University of the West of England, Bristol, Avon, Nichola Rumsey, PhD, University of the West of England, Bristol, Avon, Tracey Wade, PhD, Flinders University, Adelaide , South Australia

The application of mindfulness in the prevention of eating disorders has received preliminary support with regard to efficacy in a face-to-face format, however large scale dissemination is impeded by limited voluntary uptake by young women and reliance on expert facilitators. Offering interventions online may help to overcome these barriers to implementation. This study assessed the acceptability of an online mindfulness intervention for improving body image and other risk factors for eating disorders in young adult women. British undergraduate women (N= 174, Mage = 20.34, SD = 1.67; Mbmi = 23.78, SD = 4.97) were allocated to a self-guided online mindfulness intervention (3 x 30min modules delivered over 3 weeks) or control condition (online brochure outlining general tips for improving body image). Compliance and acceptability of the online mindfulness intervention were assessed at post-intervention. Of the 87 allocated to receive online mindfulness, 40% completed the first module, 28% completed the second, and 25% completed all three modules. Moderate acceptability was indicated across aspects of understanding (M=4.25), effectiveness (M=3.61), enjoyment (M=3.58), ease of use (M=3.88), and likelihood of continued use (M=3.56). Themes from qualitative written feedback for failing to engage with the programme included lack of time, being too busy or forgetting, too much written content and reading, and not being personally useful. Despite positive feedback from some participants, these findings indicate limited feasibility of the online mindfulness intervention in this format. Future implementation and evaluation will require additional effort to maintain engagement, with the inclusion of video content in place of written instruction, additional reminders, use of commitment devices, and option to connect with other participants likely to be beneficial.

Topic: Biology and Medical Complications
Terrace 1, second floor
Suzanne Dooley-Hash, MD & Mimi Israel, MD, FRCP, FAED

11.1: A Longitudinal Investigation of the effect of Eating Disorder Diagnoses and Eating Disorder Behaviours on Bone Mineral Density in Adult Women

Lauren Robinson, BSc, MPhil, UCL, London, UK, Victoria Aldridge, BSc; MSc; PhD, UCL, London, UK, Emma Clark, BSc; MSc; PhD, Bristol University, Bristol, UK, Nadia Micali, MRCPsych; MD; PhD; FAED, UCL, London, UK

This study aims to investigate the prospective association between lifetime ED and ED behaviors and Bone Mineral Density (BMD) in mid-life in women recruited as part of the Avon Longitudinal Study of Parents and Children (ALSPAC). A total of 5,658 women enrolled in ALSPAC participated in a lifetime prevalence study, lifetime ED behaviors and ED diagnoses were obtained. BMD was measured using Dual X-Ray Absorptiometry (DXA) at a mean age of 49 years. Linear regression methods investigated the association between ED behaviors and ED diagnosis and total body, hip, arm and leg BMD. Restrictive AN (AN-R), but not Binge-Purging AN (AN-BP) diagnosis was associated with a lower total body less head (TBLH) BMD when controlling for ethnicity, education, height2 and TBLH Bone Area (BA) z=0.37 (-0.62; -0.12); p=0.004 and z=-0.13 (-0.46; -0.19); p=0.43 respectively. Fasting and Restricting behaviors were individually predictive of low BMD at legs, arms, hip and total body after controlling for confounders (p<0.01), however the effect of purging behaviors on BMD (vomiting and misuse of medication) did not remain significant when those engaging in fasting behaviors were removed from the analysis. This is the first study of its kind to investigate ED behaviors and ED diagnosis prospectively in a community sample. ED diagnoses and ED behaviors were both individually predictive of low BMD in adult women across four anatomical sites. This study suggests that individuals who present ED behaviors without meeting the criteria for an ED diagnosis may still be vulnerable to low BMD and bone fractures throughout adulthood.

11.2: Genome-Wide Association Study Reveals First Locus for Anorexia Nervosa & Metabolic Correlations

Cynthia Bulik, PhD; FAED, University of North Carolina and Karolinska Institutet, Chapel Hill and Stockholm, North Carolina, Laramie Duncan, PhD, Stanford University, Palo Alto, California, Zeynep Yilmaz, PhD, UNC Chapel Hill, Chapel Hill, North Carolina, Eating Disorders Working Group /Psychiatric Genomics Consortium, Consortium, UNC Chapel Hill, Chapel Hill , North Carolina

The Eating Disorders Working Group of the Psychiatric Genomics Consortium conducted the largest genome-wide association study (GWAS) of anorexia nervosa to date, by combining existing samples worldwide. With 3,495 AN cases and 10,982 controls, following uniform quality control and imputation using the 1000 Genomes Project (phase 3), we performed a GWAS of 10,641,224 common variants (minor allele frequency > 1%). One region on chromosome 12 reached genome-wide significance (top hit rs4622308, p=4.3x10-9). This region spans over six genes and contains variants that previously yielded significant GWAS findings for Type I diabetes, rheumatoid arthritis, asthma, polycystic ovary syndrome, and height. The heritability estimate (h2) for AN calculated using the variants on the genotyping chip was 0.20 (SE=0.02), which is comparable to chip-based h2 results for other psychiatric disorders. This estimate is expectedly lower than twin-based heritability, as this analysis only reflects common variants, whereas twin analysis captures the effects of all types of genetic variation. Using LD Score Regression techniques applied to publicly available summary statistics from a range of phenotypes, we identified significant positive genetic correlations between AN and schizophrenia, neuroticism, educational attainment, and HDL cholesterol along with significant negative genetic correlations with body mass index, insulin, glucose, and lipid phenotypes. This is the first report of a genome-wide significant finding for AN and the addition of incoming samples (~17,000 queued for genotyping) will yield additional significant loci. The observed patterns of genetic correlations support the reconceptualization of AN as a disorder with both psychiatric and metabolic components.

11.3: Initial Medical Findings in 1,000 Consecutive Inpatient Eating Disordered Patients

Philip Mehler, MD; FACP, CEDS; FAED, Eating Recovery Center, Denver, Colorado, Keegan Walden, BA; PhD, Eating Recovery Center, Denver, Colorado, Jennifer McBride, BS; MS; MD, Eating Recovery Center, Denver, Colorado, Simrat Kaur, BS, COPIC Medical Foundation student, Denver , Colorado, Jennifer Watts, BS; MD, Eating Recovery Center, Denver, Colorado, Kristine Walsh, MPH; MD, Eating Recovery Center, Denver, Colorado

The purpose of this study is to define the prevalence of common diagnostic medical test findings in the largest-ever sample of consecutive patients with anorexia nervosa (AN-R and AN-BP) and bulimia. This is a 3-year retrospective study of medical record-derived data from 1,000 consecutive patients treated in a large eating disorder health care system. Prognostic factors assorted with medical complications of eating disorders will also be discussed. Average percent ideal body weight in the AN-R patients was 72%, and 77% in the AN-BP patients. 42% of AN-R patients had osteoporosis and 19.7% of AN-BP patients (p<.003). Only 3% of all newly admitted AN-R patients had hypophosphatemia, but 22% had hypokalemia; in AN-BP, 34% had hypokalemia (p<.001). Hyponatremia was present in 11% of all AN patients, 41.6% had bradycardia and 2.6% of AN-R patients had QT prolongation, but it was present in 5.3% of those with AN-BP (p – NS). Vitamin D deficiency was noted in 2.3% of AN-R patients, but it was never noted in any AN-BP or bulimic patients. Thirty percent of all patients with AN and 15% of patients with bulimia had anemia, 7% of patients with AN had thrombocytopenia and 11.1% had an elevated MCV, which was the most common red cell indices abnormality. The most frequent blood chemistry abnormality was a metabolic alkalosis, with over 50% of all patients with AN having bicarbonate levels greater than 27 mm/hg, and 41.8% of all patients with bulimia. 36% of patients with AN-R had abnormally high ALT/AST levels, but only 20.9% of those with AN-R (p<.001). These results demonstrate a high prevalence of significant medical findings in newly admitted patients with eating disorders, but important differences based on the type of eating disorder.

11.4: Assessment of Sex Differences in Fracture Risk among Patients with Anorexia Nervosa: A Population-Based Cohort Study using The Health Improvement Network

Neville Golden, MD; FAED, Stanford University School of Medicine, Palo Alto, California, Jason Nagata, MSc; MD, Stanford University School of Medicine, Palo Alto, California, Mary Leonard, MSc; MD, Stanford University School of Medicine, Palo Alto, California, Lawrence Copelovitch, MA; MD, The Children's Hospital of Philadelphia, Perelman School of Medicine, Philadelphia, Pennsylvania, Michelle Denburg, MD; MSc, The Children's Hospital, Philadelphia, Perelman School of Medicine, Philadelphia, Pennsylvania

Though previous studies have demonstrated an increased fracture risk in females with anorexia nervosa (AN), fracture risk in males is not well characterized. The objective of this study was to examine sex differences in fracture risk and site-specific fracture incidence in AN. We performed a population-based retrospective cohort study using The Health Improvement Network, which includes data from 553 general practices shown to be representative of the United Kingdom. The median calendar year for the start of the observation period was 2004-5. We identified 9,239 females and 556 males 40 years of age (HR 2.54, 95% CI 1.32-4.90; p = 0.005) but not among males ≤40 years. Females with AN had a higher risk of fracture at nearly all anatomic sites. The greatest increased fracture risk was noted at the hip/femur (HR 5.59; 95% CI, 3.44-9.09) and pelvis (HR 4.54; 95% CI 2.42-8.50) in females. Males with AN had a significantly increased risk of vertebral fracture (HR 7.25; 95% CI, 1.21-43.45) compared to males without AN. Our results demonstrate that AN was associated with higher incident fracture risk in females across all age groups and in males >40 years old. Sites of highest fracture risk include the hip/femur and pelvis in females and vertebrae in males with AN.

11.5: Reduced Coronary Blood Flow in Adolescents with Anorexia Nervosa.

Nogah Kerem, MD, Bnai- Zion Medical Center, Haifa, Northern Israel, Jenny Garkaby, MD, Bnai- Zion Medical Center, Haifa, Northern Israel, Liat Gelerneter- Yaniv, MD, Bnai - Zion Medical Center, HAifa, Northern Israel, Yasmine Sharif, Student, Sakkler School of Medicine, Tel Aviv University, Tel- Aviv , Central Israel, Isaac Srugo, MD, Bnai- Zion Medical Center, Haifa, Northern Israel, Dawod Sharif, MD, Bnai- Zion Medical Center, Haifa, Northern Israel

Cardiovascular complications in patients with anorexia nervosa (AN) may be significant and life threatening. Reduction in the Coronary Blood Flow (CBF), which supplies the oxygen demand of the myocardium, can potentially lead to ischemia, cardiomyopathy, and arrhythmias. The purpose of our study was to characterize the CBF in adolescents with AN. Our study evaluated CBF by measuring peak diastolic velocity calculated by echocardiography in adolescents with AN, who were hospitalized to an Adolescent Medicine Unit at their malnourished stage. A sample of healthy controls was used to establish the expected normal curve of CBF. 40 adolescents with AN were examined: 11/29 males/females, mean age of 15.3±2.4 years, mean weight loss of 22.3±11% percent body weight, mean BMI on admission 16.8±2.9 Kg/m2, percent of median BMI for age and gender 84.5±13.3%, minimal nocturnal heart rate of 38±6 beats per minute (bpm). 14 healthy controls with a functional heart murmur were examined and evaluated for their CBF curve. Their measurements were compared to those of the AN group. Peak Diastolic Velocity was 23.7±7.5 cm/sec in the AN group versus 33.9±6.3 cm/sec in the controls, p<0.00001. Reduced CBF was found to be correlated to lower percent of median BMI for age and gender (p=0.009), nocturnal bradycardia (p=0.039), diurnal bradycardia (p=0.003), decreased peak systolic velocity (p<0.0001), and decreased cardiac flow index (p<0.0001). Males were significantly more medically unstable than the females in the AN group (nocturnal bradycardia of 34.3±4.7 Vs 39.1±5.7 bpm, hospitalization days for reaching minimal heart rate of 45 16.1±5.2 Vs 10±5 days), yet no statistically significant difference was found in CBF between the genders. Our study indicates a significant decrease in CBF in adolescents with AN, and to the best of our knowledge it is the first to investigate this potentially hazardous complication, which may explain papillary muscle dysfunction (and thus valves regurgitation), and potential future cardiomyopathy.

11.6: A Longitudinal, Epigenomewide Study of DNA Methylation in Women with Anorexia Nervosa: Results in Actively ill, Long-Term Recovered, and Normal-Eater Control Women.

Howard Steiger, PhD; FAED, Douglas Mental Health University Institute, Montreal, Quebec, Linda Booij, PhD, Concordia University, Montreal, Quebec, Aurélie Labbe, PhD, McGill University, Montreal, Quebec, Luis Agellon, PhD, McGill University, Montreal , Quebec, Mimi Israël, PhD; FAED, Douglas Mental Health University Institute, Montreal, Quebec, Moshe Szyf, PhD, McGill University, Montreal, Quebec, Lea Thaler PhD, Douglas Mental Health University Institute, Montreal, Quebec, Ridha Joober, MD; PhD, Douglas Mental Health University Institute, Montreal, Quebec, Esther Kahan, BSc, Douglas Mental Health University Institute, Montreal, Quebec, Danaëlle Cottier, Student, Douglas Mental Health University Institute, Montreal, Quebec, Erika Rossi, BA, Douglas Mental Health University Institute, Montreal, Quebec, Kevin McGregor, MSc, McGill University, Montreal, Quebec

This ongoing study investigates genomewide methylation profiles in women with and without Anorexia Nervosa (AN) using 450K Illumina bead arrays. At present, we have obtained pre-treatment data for 95 AN patients (41 with AN-restrictive subtype and 54 with AN-Binge/Purge subtype), 52 AN patients at both pre- and post-treatment (28 with AN-restrictive subtype and 24 with AN-Binge/Purge subtype), 22 women in recovery from AN (for at least 12 months), and 37 normal-weight, normal-eater women. Ongoing data collection will enlarge each of the samples noted. Interim analyses using False Discovery Rate-corrected comparisons identify numerous probes that differentiate women with and without AN, touching genes associated with main neurotransmitter functions, neurogenesis, protein coding/transcription, lipid/glucose metabolism, and health of blood, bone and teeth. Preliminary pathway analyses implicate gene pathways affecting cell function, transcription, endocrine and cardiac status, and carbohydrate metabolism. Longitudinal data indicate associations between changes in BMI and changes in methylation at genes implicated in main neurotransmitter functions, immunity, glucose and lipid metabolism, and smell and taste. Findings are consistent with altered methylation in active AN, and normalization after nutritional rehabilitation, at genomic regions that are relevant to common behavioral, emotional and physiological complications in AN. If, as suggested by these data, alterations are reversible with nutritional rehabilitation, then these data have numerous clinical implications.