Paper Session II

Paper Session II
Saturday 2:45-4:15PM

Topic: Treatment of Eating Disorders (Adult) II
Club A, first floor
Jennifer Thomas, PhD, FAED & D. Catherine Walker, PhD

12.1: Predictors of Outcome in an Evidence-Based Intensive Outpatient Program for the Treatment of Eating Disorders Support Rapid Response and Treatment Duration

D. Catherine Walker, PhD, Union College, Schenectady, New York, Julia Brooks, Student, Union College, Schenectady, New York, Emily Ehrlich, Student, Union College, Schenectady, New York, Julie Morison, PhD, HPA/Livewell, Albany , New York, Drew Anderson, PhD, University at Albany, State University of New York, Albany, New York

Treatment outcome research is limited among transdiagnostic ED samples and in intensive outpatient programs (IOPs). Previous research within treatment settings for eating disorders (EDs) consistently indicates that rapid response (RR) to treatment is a significant predictor of overall treatment success (Macdonald et al., 2015). However, the RR phenomenon has not been replicated in a transdiagnostic ED IOP sample. The current study sought to examine previously reported outcome predictors (weight suppression, RR, severity of illness, ED diagnosis, number of comorbid disorders, comorbid substance use, and comorbid personality disorder), in a community IOP for EDs. The current sample consisted of 167 consecutive admissions to a community ED IOP from May 2013-June 2016. Patients were 90% female and 95% Caucasian, with mean illness duration of 8.08 years (SD = 7.95, range: .33-36) and mean of two treatment attempts (range: 0-6) prior to intake. Patients’ mean intake BMI was 21.58 kg/m2 (SD = 5.13, range: 14.53-45.09). Modal number of comorbid diagnoses was one (range: 0-3); 14 (8.2%) had a substance use disorder and 13 (7.6%) had a personality disorder diagnosis. Most were diagnosed with anorexia nervosa (n = 102; 61.1%), 38 (22.7%) with bulimia nervosa, 4 (2.4%) with binge eating disorder, 21 (12.6%) with other un/specified feeding or eating disorder, and 2 (1.2%) with avoidant/restrictive food intake disorder. Patients completed a mean of 13.73 weeks (SD = 12.37, range: 1-64) in the IOP. Only RR (F∆(1,76) = 10.96, p < .001) and total treatment duration (FΔ(1,74) = 5.51, p = .02) were significant outcome predictors, accounting for 12.6% and 6.1% of the variance in Eating Attitudes Test-26 change, respectively. The current study indicates that RR remains a consistent outcome predictor across treatment settings and in transdiagnostic samples. Future work should evaluate factors that mediate and moderate RR and incorporate RR findings into the design and implementation of outpatient ED treatment.

12.2: Clinical Characteristics of Eating Disorders patients who do not Respond to Cognitive Behavioral Therapy: A Six-Year Follow-up Study.

Giovanni Castellini, MD; PhD, University of Florence, Florence, APO AE, Valdo Ricca, MD, University of Florence, Florence, APO AE

Longitudinal studies indicated that a consistent rate of Eating Disorders (EDs) patients do not report any improvement when they undergo cognitive behavioral therapy (CBT). The present study attempts to identify the clinical characteristics of EDs who do not respond to CBT. Five hundred sixty four EDs patients (165 with Anorexia Nervosa [AN], 137 with Bulimia Nervosa [BN], 262 with Binge Eating Disorder [BED]) were evaluated by means of the Structured Clinical Interview for DSM-IV, and several self-reported questionnaires. The clinical assessment was conducted on the first day of admission and at further follow-up time points (end of individual CBT, 3, 6 years). A high rate of non-response to treatment rate was observed (AN: 20.0%; BN: 28.5%; BED: 14.1%). Duration of illness, the lack of weight gain at one year follow up, and severe EDs psychopathology were found to increase non-response risk in AN. Non-response was associated with impulsivity, diagnostic crossover, and substance abuse in BN patients, and with Unipolar Depression and Emotional Eating in BED patients. Non-responders showed mild behavioral improvement but no relevant change in EDs psychopathology. Therapeutic interventions should be targeted on specific psychopathology, taking into account potential non-response predictors.

12.3: Time to Eat! Less Regular Eating Patterns in Individuals who Binge and/or Purge

Kathryn Coniglio, BA, Massachusetts General Hospital, Boston, Massachusetts, Abigail Cooper, BA, Massachusetts General Hospital, Boston, Massachusetts, Kendra Becker, PhD, Massachusetts General Hospital, Boston, Massachusetts, Debra Franko, PhD; FAED, Northeastern University, Boston, Massachusetts, Lazaro Zayas, MD, Massachusetts General Hospital, Boston, Massachusetts, Kamryn Eddy, PhD; FAED, Massachusetts General Hospital, Boston, Massachusetts, Jennifer Thomas PhD; FAED, Massachusetts General Hospital, Boston, Massachusetts

A critical initial goal in enhanced cognitive behavioral therapy for eating disorders (CBT-E) is to help the patient establish a pattern of regular eating, comprising of three meals and two to three snacks at prescribed times throughout the day. A previous study showed that youth with anorexia nervosa restricting type (AN-R) follow a more regular pattern of eating than youth with binge eating/purging type (AN-BP). The current study aimed to test whether individuals who engage in any bingeing and/or purging follow a less regular pattern of eating (i.e., are more likely to skip meals and/or snacks) than individuals who do not binge or purge, regardless of diagnosis. Consecutive referrals to a residential eating disorders program (N=147, mean age=18.09) completed the Eating Disorders Examination (EDE), in which they described eating patterns prior to admission. We conducted a one-way ANOVA and found that women who binged and/or purged (n=92), compared to women who did not (n=55), more often skipped breakfast [F(1,145)=13.11, p<.001), mid-morning snack [F(1,145)=6.27, p<.05), lunch [F(1,145)=14.43, p<.001), mid-afternoon snack [F(1,145)=19.61, p<.001), dinner [F(1,145)=26.30, p<.001), and evening snack [F(1,145)=18.67, p<.001). The frequency of skipping any of these meals or snacks—except mid-morning snack—was correlated with the frequency of both bingeing (r=-.20 to -.39; all p’s <.05) and self-induced vomiting (r=-.26 to -.45; all p’s <.05). In particular, skipping dinner showed moderate-to-large correlations with bingeing (r=-.39, p<.001) and vomiting (r=-.45, p<.001). In sum, individuals who binge and/or purge may need additional support to establish a regular eating pattern. Furthermore, our results support the CBT-E recommendation that mid-morning snack could be optional for non-underweight patients whose primary goal is to reduce binge/purge frequency, as skipping mid-morning snack was not correlated with episodes of bingeing or purging.

12.4: Assertive Refeeding for Medically Compromised Adult Inpatients with Anorexia Nervosa: An Observational Study of Outcomes

Shane Jeffrey, RD; BSc, Royal Brisbane and Women's Hospital, Brisbane, Queensland, Kylie Matthews, M Nut and Diet, University of Queenland, St Lucia, Brisbane, Queensland, Warren Ward, MBBS, Royal Brisbane and Women's Hospital, Herston, Brisbane, Queensland, Amanda Davis, RD, Royal Brisbane and Women's Hospital, Herson, Brisbane , Queensland, Elsie Patterson, RD, Sunshine Coast University, Sippy Downs, Sunshine Coast, Queensland

Assertive refeeding for medically compromised adult inpatients with anorexia nervosa: An observational study of outcomes S. Jeffrey, K. Matthews, E. Patterson, A. Davis, W. Ward Refeeding syndrome (RS), a potentially life-threatening condition, has been defined as abnormal fluid and electrolyte shifts in response to the commencement of refeeding in malnourished patients. Due to the presence of significant malnutrition, patients with anorexia nervosa (AN) are typically considered high risk for developing RS. In adolescent AN inpatients, recent research has demonstrated that the traditional “start low, go slow” approach commonly recommended to minimise RS risk, is unnecessary. More assertive approaches to refeeding have not increased RS occurrence. Despite this, there is a paucity of evidence demonstrating the impact of this approach in adult inpatients with AN. In 2013, the Eating Disorders Outreach Service modified it’s refeeding guidelines for medically compromised adults with AN admitted to non-specialist medical wards across the state of Queensland, Australia. This protocol, supported by structured medical monitoring, commenced feeding at 1500cal/day (previously 1000cal/day) and increased by 500cal every second day until the goal of 3000cal/day was reached. Route of nutrition delivery was either via nasogastric feed or oral meal plan. Outcomes from a three year study sample of 163 patients compare pre and post data for incidence of RS, hypophosphatemia, hypoglycaemia, and time to medical stability. Preliminary data suggests that AN patients treated in a non-specialist environment are not medically disadvantaged by assertive nutritional rehabilitation with suitable levels of medical monitoring. Anecdotal feedback from health professionals and patients indicate the protocol is widely accepted.

12.5: Assessing the Effectiveness of Attention Training Therapy in the Treatment of Binge Eating within Bulimia Nervosa and Binge Eating Disorder - A Randomised Controlled Trial.

Nadine-Devaki Wright, BA; Student; Postgrad Dip in Ed, Univeristy of Dublin, Dublin, Ireland, Stephen Touyz, PhD; Professor, University of Sydney, Sydney, APO AE, Australia, Maree Abbott, DClinPsy; PhD; Associate Professor, University of Sydney, Sydney, APO AE, Australia, Evelyn Smith, PhD, University of Western Sydney, Sydney , APO AE, Australia, Elizabeth Hall, DClinPsy, University of Sydney, Sydney, APO AE, Australia

Binge eating is a core symptom of both Bulimia Nervosa (BN) and Binge Eating Disorder (BED). Information processing theories of eating disorders propose that selective attention towards food stimuli may be a significant factor maintaining binge eating within these disorders. The Attention Training Therapy (ATT) program used in this RCT was originally designed as a treatment for social phobia. ATT was found to be as effective as CBT but with better outcomes in reducing fear of negative evaluation and self-focused attention. This ATT program was modified to focus on binge eating by teaching individuals to shift their attention away from binge urges to the task at hand and towards thoughtful eating, which was hypothesized to reduce binge eating frequency. This study aimed to evaluate the efficacy of group-based manualised ATT for individuals with BN or BED compared to a waitlist control condition. 49 females met DSM-5 criteria (BN = 42; BED = 7) and were randomly allocated to either group-ATT treatment (N = 27) or waitlist control (N = 22). All participants were assessed using a clinician-administered semi-structured interview (EDE) and a battery of questionnaires at pre-treatment, post-treatment and follow-up.Results at post-treatment and follow-up found no significant reduction in binge eating frequency between the groups. However, the treatment group reported significantly increased impulse control, a reduced sense of ‘no control over their eating’ and they also endorsed fewer weight concerns compared to the waitlist group. This study provides a unique contribution to the eating disorders treatment literature by examining the impact of modifying attentional focus on binge eating symptoms, an area not previously researched.

12.6: Therapist Drift and Clinicians Experiences of Working with People who have Eating Disorders.

Sonja Skocic, BSc; MA; PhD, The University of Melbourne, Melbourne, Victoria

Several studies have suggested that clinicians are reluctant to commit to evidence-based treatment protocols (e.g., Simmons et al., 2008). The aim of this study was to explore the relationship between interpersonal experiences of working therapeutically with people that have eating disorders and therapist drift away from using evidence based treatment. Cross-sectional information was obtained from clinicians who treat eating disorders (N=165) via an online survey that included questions regarding countertransference, attitudes to evidence based protocols, and the personal experience of working therapeutically (including emotional avoidance, accommodating and enabling behaviours and rigid therapeutic interactions). Clinicians were also asked to record whether they had a lived experience of an eating disorder themselves. The results supported existing theories that describe therapist drift away from using evidence-based treatments (i.e., Waller, 2009) and highlight the impact that interpersonal factors between clinician and patient have on treatment selection and adherence to protocol. Moreover, the results suggest that problematic interpersonal factors between clinician and patient may be contributing to accommodating and enabling the eating disorder (as per Schmidt & Treasure, 2006; Treasure, 2011).

Topic: Child and Adolescence II
North Hall, second floor
Jocelyn Lebow, PhD & Annemarie van Elburg, MD, PhD, FAED

13.1: A Pilot Evaluation of Radically-Open Dialectical Behaviour Therapy for adolescents with Anorexia Nervosa

Mima Simic, MSc; MD; MRCPsych, South London and Maudsley NHS Foundation Trust, London, UK, Catherine Stewart, BA; DClinPsy; PhD, South London and Maudsley NHS Foundation Trust, London, UK, Katrina Hunt, BSc; DClinPsy, South London and Maudsley NHS Foundation Trust, London, UK, Samantha Bottrill, BSc; DClinPsy, South London and Maudsley NHS Foundation Trust, London, UK, Miriam Ziriat, BSc, South London and Maudsley NHS Foundation Trust, London, UK

Radically-Open Dialectical Behaviour Therapy (RO-DBT; Lynch et al., 2008) has been developed to treat conditions that are contributed to and maintained by ‘over-controlled’ personality features which manifest in over–constrained coping styles and a lack of social connection with others. Interpersonal difficulties, rigidity and under-expression of emotion have been well-documented in Anorexia Nervosa (AN) and can be seen as precipitating or maintaining factors in the illness. This presentation reports data from a pilot evaluation of RO-DBT with adolescents with AN. 57 young people (aged 12-17) have completed RO-DBT skills classes as a component of their intensive day patient treatment and 12 young people (aged 14-17) have completed RO-DBT skills classes and individual therapy as outpatients in Phase Three of FT-AN focused on individual development. Of these 4 young people received RO-DBT in both settings. This presentation will include data from underweight (% median weight for height mean = 83.78) and weight restored (% median weight for height mean = 93.55) young people. The classes are offered to all day patients, and to outpatients who identified as having features of over-control which prevent psychological recovery. Analysis of factors associated with over-control measured before and after RO-DBT treatment reveals significant improvement in self-reported social connectedness (t = 2.47, p<.05, n = 26) and experiences of anticipatory pleasure (t = 2.43, p<.05, n = 21), and consumatory pleasure (t = 4.68, p<.001, n = 25). Young people also reported changes in their perceptions of their parents’ acceptance of them (t = 4.84, p<.001, n = 14). Statistical trends are observed in changes in emotion regulation (t = 2.17, p 0 = .051., n = 13 ) and reductions of discomfort in attachment relationships (t = 2.12, p 0.06, n = 11). These findings will be discussed in relation to models of social, emotional and neurobiological features of AN, and the impact of change in these on recovery from AN in adolescence.

13.2: Compulsive Exercise in Adolescents with Eating Disorder: A Multi-site Longitudinal Study

Johanna Levallius, BSc; MSc, Karolinska Institute, Stockholm, Stockholm, Christina Collin, MD, Karolinska Institute, Stockholm, Stockholm, Andreas Birgegård, PhD, Karolinska Institute, Stockholm, Stockholm

Excessive exercise (EE) to control weight and/or shape has been proposed as significant in the etiology, development and maintenance of eating disorder (ED), resulting in more severe and enduring pathology. Few studies have investigated EE among adolescents with ED. This study aimed to investigate the clinical picture and prognosis of adolescents with ED and EE. Over 3000 girls and boys from a national ED database were investigated on EE prevalence and frequency in relation to psychiatric symptoms, associated features and outcome. Denial of illness was adjusted for. Adjusted EE prevalence in girls was 44%; with lowest prevalence in anorexia nervosa. Those with EE scored significantly higher than non-EE on total ED severity, restriction and negative perfectionism. There were only minor differences between EE and non-EE patients on emotional distress, hyperactivity, suicidality and self-esteem. Initial EE did not impact prognosis, yet cessation was associated with remission. Among boys, adjusted EE prevalence was 38%, and their clinical presentation mirrored very well that of girls. EE is a common clinical feature in adolescents with ED and cessation is associated with recovery. When controlling for denial of illness, EE had less detrimental impact on ED than predicted. We recommend development of a clear definition of EE and further exploration of treatment implications.

13.3: Establishing Goal Weights for Adolescent Eating Disorder Patients: What is the State of the Field?

Jocelyn Lebow, PhD, Mayo Clinic, Rochester, Minnesota, Leslie Sim, PhD, Mayo Clinic, Rochester, Minnesota, Erin C. Accurso, PhD, University of California, San Francisco, San Francisco, California

Eating disorder treatment success often hinges on achieving an “expected body weight” (EBW)—the weight at which physical and psychological symptoms recede, medical complications resolve, and functioning returns to baseline. This indicator is necessary for diagnostic accuracy, for tracking progress, and for justifying higher levels of care to third-party payers. Despite the importance of this marker, there is little consensus in the field about the optimal method of determining EBW. Calculation for adolescents becomes even more complicated since their EBW is a moving target. Despite several suggested methods (e.g., BMI, McLaren, Moore methods), it is unclear whether a practical “industry standard” exists for calculating or conceptualizing EBW. This study surveyed 112 child/adolescent eating disorder treatment providers (13.4% physicians, 38.4% psychologists, 36.7% masters level therapists, 9% dieticians). Results suggest that methods of determining EBW varied among practitioners. Although 40.7% use data from individual growth curves to determine EBW, the remaining use diverse approaches ranging from algorithms (e.g. the BMI method) to considering patient input. Group differences were considered with regards to 3 broad categories of EBW calculation methods: rule-based approaches (e.g., BMI, McLaren or Moore methods), individual difference-based approaches (e.g., growth curve, physiological, cognitive symptoms), or other approaches (e.g., patient preference). No group differences were found between provider types, years of experience, percentage practice with children/adolescents vs. adults, or practice setting. However, providers who used FBT were significantly more likely to use an individual approach vs. non-rule based approaches that were considered “other.” Results suggest that there is a lack of consensus of how best to define EBW. Clinical implications and future research directions will be discussed.

13.4: Caregiver Coping and Appraisal of the Caregiving Experience among Adolescent Eating Disorder Patients: A Comparative Study with Substance Use Disorder Patients and Healthy Adolescents.

Melissa Parks, MSc, Autonomous University of Madrid, Madrid, Spain, Dimitra Anastasiadou, PhD, Universitat Oberta de Catalunya, Barcelona, Madrid, Ana Rosa Sepulveda, PhD, Autonomous University of Madrid, Madrid, Madrid, Montserrat Graell, MD; PhD, Niño Jesus University Hospital, Madrid , Spain, Julio César Sánchez, MSc, Proyecto Hombre – Programa Soporte, Madrid, Spain, Tamara Alvarez, MSc, Autonomous University of Madrid, Madrid, Spain

Caring for a child with an eating disorder (ED has several mental and physical consequences. The stress-coping model suggests this relationship is moderated by the appraisal of the situation as positive or negative and the coping strategies possessed by the caregiver. Limited quantitative data exists on ED caregivers’ coping strategies or if they are associated with their appraisal of the caregiving experience. Furthermore, no studies employ comparison groups. This cross-sectional study involved 48 mothers and 44 fathers of ED patients, 46 mothers and 36 fathers of patients with a substance use disorder (SUD) and 63 mothers and 51 fathers of healthy adolescents (HC). Study aims included 1) to assess the differences in coping strategies (COPE-60) between groups, 2) to evaluate gender differences in use of coping strategies, and 3) to explore the relationship between coping strategies and appraisal of the caregiving experience (Experience of Caregiving; ECI). The results of the MANCOVA revealed an effect of caregiver group on coping strategies for mothers [F(8, 302) = 2.558, p = .010; Wilks' Λ = .877, partial η2 = .063], but not for fathers. Mothers of both patient groups used self-sufficient problem-focused coping (an adaptive strategy) more than HC mothers. Also, mothers used an adaptive strategy of social support coping significantly more than fathers in all three groups. Finally, among ED mothers, adaptive coping was associated with recognition of positive aspects of the caregiving experience whereas avoidant coping was associated with recognition of the negative aspects of the caregiving experience. The strategies used by ED caregivers do not differ greatly from the comparison groups. However, the relationship between coping strategies and appraisal underlines the importance of encouraging adaptive strategies and minimizing the use of maladaptive ones among ED caregivers. Furthermore, both mothers and fathers should be included in the treatment process.

13.5: A Comparative Study of ARFID and AN in Children under the Age of 13 in a Tertiary Care Hospital Setting: Prevalence, Severity, Chronicity and Co-morbidities.

Melissa Lieberman, PhD, Hospital for Sick Children, Toronto, Ontario, Melissa Houser, PhD, Hospital for Sick Children, Toronto

The purpose of this study was to examine eating disorders in children under age 13 in a tertiary care hospital setting, in light of the new DSM-5 diagnostic classification system. Children with ARFID and Anorexia (AN) were compared to determine the prevalence, severity, chronicity, and co-morbidity in these disorders. The study included patients ranging in age from 7 to 12, participating in inpatient or outpatient eating disorders treatment between January 2013 and July 2016. At time of entry into the Eating Disorders Program, patients and parents participated in a semi-structured diagnostic interview and completed a range of questionnaires. A total of 106 patients were assessed. Of these patients, 68 were diagnosed with AN, 25 with ARFID, 3 with OSFED, 2 with UFED, and 8 had no eating disorder. Therefore, 93 patients were retained for data analyses. The average age of patients was 11.27 (SD= .9), the average BMI was 15.28 (SD= 1.82), and the average percentage of ideal body weight (IBW) was 80.11 (SD= 7.62). Patients with ARFID and AN were compared descriptively and using one-way ANOVAS. 28% of the ARFID sample was male, while only 13% of the AN sample was male. Significant differences were found between groups for chronicity in months (F(90)= 12.22, p < .001). ARFID patients had higher rates of chronicity (M= 20.25, SD= 32.13) than patients with AN (M= 6.32, SD= 4.89). For illness severity, patients with ARFID (M= 81.00, SD= 7.90) and AN (M= 79.77, SD= 7.65) were similar in terms of percentage of ideal body weight (IBW), but patients with AN (M= 59.79, SD= 14.47) had significantly lower heart rates than patients with ARFID (M= 73.63, SD= 17.45). Related, 62% of patients with AN, and only 44% of patients with ARFID, were admitted as inpatients at the time of assessment. In terms of co-morbidities, 3% of patients with AN had a co-morbid developmental disorder, while 13% of those with ARFID did. High rates of anxiety disorders were found in both groups (38% for ARFID and 30% for AN). Co-morbid depressive disorders were low in patients with ARFID and AN. Findings have implications for assessment and treatment of eating disorders in children under the age of 13.

13.6: Program Evaluation of a Canadian Residential Treatment Model for Adolescents with Eating Disorders - Pilot Data from the First 2 Years of Operation

Leora Pinhas, MD, Ontario Shores Centre for Mental Health Sciences, Whitby, Ontario, Raluca Morariu, BSc, Ontario Shores Centre for Mental Health Sciences, Whitby, Ontario, Melanie Stuckey, PhD, Ontario Shores Centre for Mental Health Sciences, Whitby, Ontario, Sheila Bjarnason, MSW, Ontario Shores Centre for Mental Health Sciences, Whitby, Ontario

The Eating Disorder (ED) Residential Program (EDRP) is a government-funded program in Ontario, explicitly created to provide an alternative for adolescents unsuccessfully treated by traditional ED programs. It is an innovative model that incorporates concepts of transparency, patient directed care and co-design common in the general mental health discourse, but are relatively new to ED literature. Over the first two years of operation, quantitative data was collected using a questionnaire battery administered at admission and discharge and included demographics, primary and concurrent diagnoses, ED psychopathology (EDE-q), depressive symptoms (Child Depression Inventory; CDI2), anxiety symptoms (Multidimensional Anxiety Scale for Children; MASC2), heights and weights. Descriptive analyses, chi-square analyses, T-Tests and ANOVAs were performed in analyzing the data. Twenty two adolescents (61% of the total patient population; Female=21; Male=1), with a mean age of 16.1 years (SD=1.17) consented to be included in the study sample. Patients all had a minimum of 2 prior inpatient admissions and a mean length of stay of 7.7 (SD=5.13) months. Primary diagnoses were anorexia nervosa (91%), bulimia nervosa (5%) and ARFID (5%). Concurrent diagnoses were present in 68% of the sample, most commonly anxiety disorders (41%). The mean BMI on admission was 18.3 (SD=1.51) and at discharge was 19.8 (SD=1.50) (p<0.001). The EDE-q significantly improved from admission (M=3.93; SD=1.31) to discharge (M=1.55; SD=1.29) (p<0.005). A similar pattern was found on the MASC2 at admission (M=65.3; SD=16.57) and discharge (M=56.8; SD=20.02) (p<0.05) and the CDI at admission (M=69.83; SD=12.58) and discharge (M=55.42; SD=11.19) (p<0.001). Patients experienced both a clinically and statistically significant improvement during their stay at the EDRP. The results of this pilot study suggest that patients with chronic and/or complex EDs can benefit from an alternative innovative intensive treatment model.

Topic: Neuroscience II
Club E, first floor
Guido Frank, MD, FAED & Carrie McAdams, MD, PhD

14.1: Is Response in Inhibitory Control Circuitry Modulated by Eating in Women Remitted from Bulimia Nervosa?

Laura A. Berner, PhD, University of California, San Diego, San Diego, California, Amanda Bischoff-Grethe, PhD, University of California, San Diego, San Diego, California, Christina E. Wierenga, PhD, University of California, San Diego, San Diego, California, Alan Simmons, PhD, University of California, San Diego, San Diego , California, Ursula Bailer, MD, Medical University of Vienna, Vienna, Vienna, Alice V. Ely, PhD, University of California, San Diego, San Diego, California, Walter H. Kaye MD, University of California, San Diego, San Diego, California

Aberrant activation in inhibitory control circuitry is correlated with binge eating frequency in bulimia nervosa (BN). However, it is unclear whether this dysfunction is exaggerated 1) after restriction, potentially increasing vulnerability to binge eating initiation, or 2) after eating has started, potentially contributing to difficulty stopping eating. The purpose of the present study was to examine frontostriatal function in women remitted from BN (RBN) when both fasted and fed to help disentangle the impact of metabolic state on inhibitory control. RBN (n = 23) and control women (CW; n = 22) performed a parametric Stop Signal Task during fMRI on two counterbalanced visits—after a 16-hour fast or a standard meal. Regardless of metabolic state, groups performed similarly on the task, but, during inhibitory errors, the RBN group showed increased activation relative to CW in left dorsolateral prefrontal cortex (DLPFC) and right anterior cingulate (ACC). In contrast, activation during successful inhibition depended on metabolic state. After eating, when correctly inhibiting responses, RBN women showed greater activation than CW in left DLPFC. RBN also showed greater activation in this region during response inhibition when fed compared to when they were fasted. Increased error-related ACC activation has been observed in women ill with BN and predicts BN onset in adolescents. Therefore, our results in RBN suggest that increased error-related ACC response may represent a trait biomarker for the disorder. Findings during successful inhibition suggest a potential neural mechanism underlying fast-binge-purge cycles: Individuals with BN may require fewer prefrontal resources to inhibit responses when fasted, but greater effort to maintain the same level of control when fed. Since stopping may require greater neural effort after eating in BN, focusing on stimulus control and inhibitory training in the post-meal period may be an effective target for future interventions.

14.2: Structural Covariance Networks in Anorexia Nervosa: A Graph Theoretical Analysis

Enrico Collantoni, MD, University of Padua, Padua, Veneto, Enrico Collantoni, MD, University of Padua, Padua, Veneto, Davide Gallicchio, MD, University of Padua, Padua, Veneto, Maria Antonietta Vergine, MD, University of Padua, Padua , Veneto, Elisa Bonello, PsyD, University of Padua, Padua, Veneto, Elena Tenconi, PhD; PsyD, University of Padua, Padua, Veneto, Daniela Degortes PhD; PsyD, University of Padua, Padua, Veneto, Paolo Santonastaso, MD, University of Padua, Padua, Veneto, Angela Favaro, MD; PhD, University of Padua, Padua, Veneto

In this study we used graph theory to describe cortical network organization in Anorexia Nervosa (AN). We used a MRI morpho-structural covariance analysis based on three parameters: cortical thickness, gyrification and fractal dimension. A total of 38 patients with acute AN and 38 healthy controls were included in this study. A group of 20 patients in full remission from AN were included to test the state/trait nature of any MRI finding. Data was collected on a Philips Achieva 1.5T scanner equipped for echo-planar imaging. Surface extraction, local gyrification index and cortical thickness estimation were completed using the FreeSurfer package. Graph analysis was performed using the Graph Analysis Toolbox. The differences between the analysed parameters likely depend on their different morpho-functional meanings: cortical thickness is more influenced by situational factors, like malnutrition, while gyrification and fractal dimensionality show more stable features over neurodevelopment. In patients with acute AN, the covariance analysis among cortical thickness values showed a significant increase of some parameters (clustering coefficient, local efficiency, modularity) indicating a more segregated network. We also observed a reduced global efficiency coefficient, which is a measure of global network integration. In the recovered patients group, we noticed a similar global trend (increased network segregation and lower global integration) without statistically significant differences for any single parameter. According to gyrification indexes, the covariance network showed a trend towards high segregation, both in acute and recovered patients. We did not, on the contrary, observe any significant difference in the covariance networks in the analysis of fractal dimension. The presence of increased segregation properties in cortical covariance networks in AN may be determined by a retardation of neurodevelopmental trajectories, or by an energy saving adaptive response.

14.3: Medial Prefrontal Cortex Engagement During Self and Other Evaluations is Related to Body Shape and Anxiety in Adolescent Anorexia Nervosa

Carrie McAdams, MD; PhD, UT Southwestern Medical Center, Dallas, Texas, Jie Xu, MD; PhD, UT Southwestern Medical Center, Dallas, Texas, Jessica Harper, BA, UT Southwestern Medical Center, Dallas, Texas, Erin Van Enkevort, PhD, UT Southwestern Medical Center, Dallas , Texas, Kelsey Latimer, PsyD, UT Southwestern Medical Center, Dallas, Texas, Urszula Kelley, MD, UT Southwestern Medical Center, Dallas, Texas

Anorexia nervosa (AN) is an illness that frequently begins during adolescence and involves weight loss motivated by a desire to change one’s appearance. Two groups of adolescent girls (AN-A, weight-recovered following AN, n = 24) and (HC-A, healthy comparison, n = 18) completed a functional magnetic resonance imaging task involving social self-evaluations. Depression, anxiety, eating behaviors and body shape were correlated with neural activity in a priori regions of interest, established from prior work examining adults with AN. A cluster in medial prefrontal cortex and the dorsal anterior cingulate correlated with the body shape questionnaire (r = -0.44, p = 0.004); subjects with more body shape concerns used this area less during self than friend evaluations in both groups. A cluster in medial prefrontal cortex and the cingulate correlated with anxiety in the AN-A group (r = -0.51, p = 0.01), whereby more anxiety was associated with engagement when disagreeing rather than agreeing with social terms during self-evaluations. One year clinical follow-up was obtained for the AN-A group, leading to recovered (AN-AR) and ill (AN-AI) groups. The AN-AR group used the posterior cingulate and precuneus more when considering another person’s perspective during self-evaluations than the AN-AI group (r = 0.62, p = 0.005). Differences in the utilization of frontal brain regions for social evaluations may contribute to both anxiety and body shape concerns in adolescents, potentially mediating pathology in AN. As clinical outcomes were associated with use of the posterior cingulate and precuneus, neural differences related to social evaluations may provide clinical predictive value. Activation of these neural regions appropriately for social evaluations may be a key component for achieving sustained weight-recovery following AN.

14.4: Neural Correlates of Explicit Regulation of Negative Emotions in Patients with Anorexia Nervosa

Stefan Ehrlich, MD, Faculty of Medicine, TU Dresden, Dresden, Saxony, Joseph A. King, PhD, Faculty of Medicine, TU Dresden, Dresden, Saxony, Franziska Ritschel, DiplPsych, Faculty of Medicine, TU Dresden, Dresden, Saxony, Ilka Boehm, DiplPsych, Faculty of Medicine, TU Dresden, Dresden , Saxony, Daniel Geisler, Dipl.Ing., Faculty of Medicine, TU Dresden, Dresden, Saxony, Fabio Bernardoni, PhD, Faculty of Medicine, TU Dresden, Dresden, Saxony, Kersten Diers DiplPsych, Department of Psychology, TU Dresden, Dresden, Saxony, Alexander Strobel, PhD, Department of Psychology, TU Dresden, Dresden, Saxony, Thomas Goschke, PhD, Department of Psychology, TU Dresden, Dresden, Saxony, Henrik Walter, MD; PhD, Department of Psychology, Charité Berlin, Berlin, Berlin, Maria Seidel, MSc, Faculty of Medicine, TU Dresden, Dresden, Saxony

Theoretical models and treatment development efforts have increasingly focused on the role of emotion recognition and regulation difficulties in the etiology and maintenance of anorexia nervosa (AN). However, up to now research has mainly included self-report data to undermine this theory. The current study aims at testing the ability of AN patients to willingly downregulate negative emotions by means of reappraisal. Such strategies are regarded as adaptive and have been shown to successfully reduce amygdala activity and recruit prefrontal areas associated with cognitive control processes. Emotion regulation ability is investigated via behavioral assessment (arousal ratings) and neural activity (fMRT) by using a standard emotion regulation paradigm comparing simply watching emotional stimuli and regulating them. Our sample consists of 35 acute AN patients and an pairwise age-matched control group. Behavioral and imaging analyses suggest a successful reduction of arousal and amygdala activity during the regulation condition for both patients and controls. However, compared with controls, individuals with AN showed relatively increased activation in the amygdala as well as in the bilateral dorsolateral prefrontal cortex during the passive viewing of aversive compared with neutral pictures. Our data do not support the notion of a general emotion regulation deficit in AN. However, the findings support previous theories of increased emotional reactivity to negative events in AN. Further, we found increased task-independent recruitment of brain regions implicated in cognitive control which could be interpreted within the framework of habitual control processes in AN. The current results might also suggests that reappraisal may represent a successful emotion regulation strategy for individuals with AN.

14.5: Threat detection and attentional bias to threat in women recovered from anorexia nervosa: Neural alterations in extrastriate and medial prefrontal cortices.

Lasse Bang, MA; Student, Regional Department for Eating Disorders, Oslo University Hospital, Oslo, Oslo, Øyvind Rø, MD; PhD, Regional Department for Eating Disorders, Oslo University Hospital, Oslo, Oslo, Tor Endestad, PhD, Department of Psychology, University of Oslo, Oslo, Oslo

Behavioral studies have shown that anorexia nervosa (AN) is associated with attentional bias to food- and body-related cues. Similar attentional bias to general threat cues have been reported, but findings are inconsistent. The aim of this study was to investigate the neural responses associated with threat-detection and attentional bias to threat in women recovered from AN. We used functional magnetic resonance imaging to measure neural responses to a dot-probe task, involving pairs of angry and neutral face stimuli, in 22 adult women recovered from AN (age 27.32 ± 5.14 years) and 21 comparison women (age 26.00 ± 4.71 years). Women recovered from AN were behaviorally unimpaired, and did not exhibit a behavioral attentional bias to threat. In response to angry faces, women recovered from AN showed significant hypoactivation in the extrastriate cortex. During attentional bias to angry faces, recovered AN women showed significant hyperactivation in the medial prefrontal cortex. Our results suggest that women recovered from AN are characterized by altered neural responses to general threat cues, and display differential neurocognitive attentional processes to such stimuli, which may reflect compensatory mechanisms. This could account for some of the discrepant findings of behavioral attentional bias to general threat in AN.

14.6: Lower levels of Glutamate in the Brains of those with Anorexia Nervosa: A Magnetic Resonance Spectroscopy Study at 7 Tesla

Alexandra Pike, BA; MSc, University of Oxford, Oxford, Oxfordshire, Beata Godlewska, MBBS; PhD, University of Oxford, Oxford, UK, Ann Sharpley, BSc; PhD, University of OIxford, Oxford, UK, Rebecca Park, BSc; MB; MRCPsych; PhD, University of Oxford, Oxford , UK, Agnes Ayton, MD; MRCPsych; MSc; MMedSc, University of Oxford, Oxford, UK, Uzay Emir, BSc; MSc; PhD, University of Oxford, Oxford, UK, Philip Cowen BS; BSc; MB; MD; MRCPsych, University of Oxford, Oxford, UK

The purpose of this study was to use Magnetic Resonance Spectroscopy (MRS) to examine brain glutamate concentration in patients with Anorexia Nervosa (AN) compared to healthy controls, with the hope of improving our neurobiological understanding of this disorder. Recently it has become possible to perform MRS at 7T, which enables researchers to differentiate between glutamate and its precursor and metabolite glutamine. The higher field strength removes the confounding necessity found at lower field strengths to derive a combined measure of these two neurochemicals (known as Glx). We applied this technique to 13 patients with anorexia nervosa (AN) and 12 healthy controls, and looked at three voxels: the anterior cingulate cortex, the occipital cortex, and the putamen. We corrected for total CSF in the voxel and for water content, and analysed the data using LCModel. We found that in patients with acute AN, brain glutamate levels were significantly lowered compared to healthy controls in all three voxels we studied. A repeated measures ANOVA for glutamate showed a main effect of diagnosis (F (1,19) = 13.7, p=0.002) but no interaction between diagnosis and voxel region (F (2,19) = 0.053, p=0.95). There was no change in glutamine concentration in AN (all p values > 0.1). The overall reduction in glutamate across these three voxels was about 8%. Some previous investigations have found lowered Glx in AN; our study suggests this is attributable to lowered glutamate concentrations rather than alterations in glutamine. The lowered glutamate levels in AN might be a consequence of starvation or could play a maintaining role in the disorder. If the latter is the case there may be a place for pharmacological manipulation of glutamate in AN treatment.

Topic: Body Image II
Club H, first floor
Jennifer Lundgren, PhD, FAED & Kendra Becker, PhD

15.1: Body Dissatisfaction Predicts Engagement in Non-Compensatory Purging among Eating Disorder Patients who have a Greater Tendency to be Intolerant of Negative Emotions

Kendra Becker, BA; MS; PhD, Massachusetts General Hospital/Harvard Medical School, Boston, Massachusetts, Nassim Tabri, PhD, Carleton University, Ottawa, Kathryn Coniglio, BA, Massachusetts General Hospital, Boston, Massachusetts, Sarah Fischer, PhD; FAED, George Mason University, Fairfax , Virginia, Franziska Plessow, PhD, Massachusetts General Hospital/Harvard Medical School, Boston, Massachusetts, Debra Franko, FAED; PhD, Northeastern University, Boston, Massachusetts, Kamryn Eddy PhD; FAED, Massachusetts General Hospital/Harvard Medical School, Boston, Massachusetts, Jennifer Thomas, PhD; FAED, Massachusetts General Hospital/Harvard Medical School, Boston, Massachusetts

Overvaluation of weight/shape or body dissatisfaction (BD) is identified as the core psychopathology underlying extreme weight-control behaviors in the cognitive behavioral therapy (CBT) model of eating disorders (ED). However, while CBT emphasizes reducing binge frequency as a way to reduce compensatory purging, fewer strategies are described for addressing non-compensatory purging (e.g., vomiting or laxative use not associated with a specific binge episode). Further, non-compensatory purging may be complicated by poor emotion regulation. We examined the utility of BD and low distress tolerance for predicting engagement in non-compensatory purging (combined laxative use and self-induced vomiting) among individuals with ED. We hypothesized that BD would predict engagement in non-compensatory purging, particularly among participants who have a predisposition towards negative mood intolerance and emotional reactivity. In Study 1 (N = 143), we examined whether the Behavioral Inhibition Scale (BIS: measure of anxiety sensitivity) moderated the relationship between BD and non-compensatory pursing in a transdiagnostic residential sample. In Study 2 (N = 193), we examined whether negative urgency (NU: tendency to act rashly during negative moods) moderated the relationship between BD and non-compensatory purging in a transdiagnostic outpatient sample. As expected, in Study 1, BD was a predictor of non-compensatory purging among participants who scored high on BIS (B = .01; p =.02). Among participants who scored low on BIS, BD was not related to non-compensatory purging. We observed the same pattern of results in Study 2 with NU as moderator (B = .06; p = .02). These findings suggest that addressing negative mood intolerance and emotional reactivity by including an emphasis on distress tolerance, mindfulness, and emotion regulation may be instrumental for addressing non-compensatory purging, particularly among patients with high BD.

15.2: Disentangling Body Image: The Relative Clinical Significance of Weight/Shape Overvaluation, Dissatisfaction, and Preoccupation in Australian Girls and Boys

Deborah Mitchison, MPsych; MSc; PhD, Macquarie University, Sydney, NSW, Phillipa Hay, DPhil; FAED, Western Sydney University, Sydney, ACT, Scott Griffiths, PhD; BPsych(Hons), University of Canberra, Canberra, ACT, Stuart B Murray, DClinPsy; PhD; BPsych(Hons), University of California San Francisco, San Francisco , California, Caroline Bentley, BSc, Australian National University, Canberra, ACT, Kassandra Gratwick-Sarll, BA; DClinPsy, Australian National University, Canberra, ACT, Carmel Harrison PhD; BPsych(Hons), Australian National University, Canberra, ACT, Jonathan Mond, MPH; PhD, Western Sydney University, Sydney, NSW

The distinctiveness, and relative clinical significance of overvaluation, dissatisfaction, and preoccupation with body weight/shape remains inconclusive. This study sought to add to the evidence by testing associations between these three body image constructs and indicators of clinical significance. Male and female secondary students (N = 1666) aged 12-18 years completed a survey that included measures of dissatisfaction with, overvaluation of, and preoccupation with weight/shape, psychological distress, eating disorder behaviors, and basic demographic information. Conditional process analysis was employed to test the independent and mediating effects of overvaluation, dissatisfaction, and preoccupation on distress, dietary restraint, and objective binge eating. In girls, preoccupation demonstrated the strongest independent and mediating effects on distress, dietary restraint, and binge eating; whereas neither the direct or indirect effects of dissatisfaction on distress and overvaluation on binge eating were significant. Among boys however, the direct and indirect effects of overvaluation, dissatisfaction, and preoccupation on distress and eating disorder behaviors were relatively equal. These findings indicate that preoccupation with weight/shape may be particularly clinically significant in girls, whereas all constructs of body image disturbance may be equally significant in boys. The findings are consistent with the view that these constructs, while closely related, are distinct.

15.3: Weight and Shape Overvaluation as a Core Symptom in Eating Disorder Psychopathology: A Transdiagnostic and Disorder-Specific Network Analysis

Russell DuBois, MS, Northeastern University, Boston, Massachusetts, Rachel Rodgers, PhD, Northeastern University, Boston, Massachusetts, Debra Franko, PhD, Northeastern University, Boston, Massachusetts, Kamryn Eddy, PhD, Massachusetts General Hospital, Boston, Massachusetts, Jennifer Thomas, PhD, Massachusetts General Hospital, Boston, Massachusetts

The purpose of the current study is to test the enhanced cognitive behavioral model of eating disorders by investigating whether weight and shape overvaluation is a core symptom in eating disorder psychopathology using a network analysis. Males and females (n = 194; age M = 25.5, SD = 11.7) who were seeking outpatient treatment for an eating disorder and who received an eating disorder diagnosis were included in the study. We used the Eating Disorder Examination Questionnaire and the Eating Pathology Symptoms Inventory to assess participants’ eating disorder symptoms. We conducted a network analysis to create transdiagnostic and disorder-specific (anorexia nervosa, bulimia nervosa, and binge eating disorder) eating disorder symptom networks whereby each node in the networks represents one eating disorder symptom and each edge in the networks represents the L1 (lasso) regularized covariance between two symptoms. We calculated the centrality of each symptom to identify core symptoms that have the greatest correlation with all other symptoms in the network. We assessed network stability with subset bootstrapping. Results indicated that weight and shape overvaluation had the largest symptom centrality among all of the eating disorder symptoms (MStrength = 1.12). Weight and shape overvaluation also exhibited the highest symptom centrality when calculated separately for each diagnostic group, including binge eating disorder. Our results provide robust support for the enhanced cognitive behavioral model of eating disorders by highlighting the powerful influence of overvaluation of weight and shape on nearly all other eating disorder symptoms, regardless of eating disorder diagnosis. Our findings identify weight and shape overvaluation as both a critical transdiagnostic treatment target and a potentially useful severity specifier for binge eating disorder.

15.4: When the Fit-Ideal Gets Scary: Body Image Ideals and Associated Eating and Body Dysmorphic Disorder Symptoms in Women in Weight Training

Andrea Hartmann, BSc; MPsych; PhD, Osnabrück University, Osnabrück, Lower Saxony, Leon Lange, BSc, Osnabrück University, Osnabrück, Schleswig-Holstein, Anna Spree, MPsych, Christian-Albrechts-Universität zu Kiel, Kiel, Schleswig-Holstein, Florian Steenbergen, BSc, Osnabrück University, Osnabrück , Lower Saxony, Manuel Waldorf, PhD, Osnabrück University, Osnabrück, Lower Saxony

To examine the differential relationship of thin, lean, and muscular body image ideals with eating and body dysmorphic disorder pathology, exercise dependence, and the use of appearance- and performance-enhancing drugs and supplements in a sample of women in weight training. We recruited 158 female weight trainers with a mean age of 26.6 years. In an online survey, participants completed Drive for Thinness (DT), Leanness (DT), and Muscularity (DM) scales, Eating Disorder Examination-Questionnaire, Muscle Dysmorphia Inventory (MDI), and Body Dysmorphic Disorder Questionnaire. Additionally, 36 different body silhouettes assessed in a subsample the participants’ perceived current and ideal mix of fat and muscular body mass. A total of 29.5% participants (subsample of n=51) chose the two greatest muscle options on the “should” silhouettes. DT and DM correlated significantly with ED (r = .85 and .36) and BDD (r = .66 and .31; all ps < .01) pathology. DL was only related to ED pathology (r = .22, p < .05). While exercise dependence (MDI) is highly associated with all drives (all r > .22, all ps < .01), the scales supplement use and pharmacology are only correlated with DM and DL (r = .49 and .18), and DM and DT (r = .24 and .28; all ps < .01), respectively. This study is the first to focus on DM in relation to DT and DL in females and examines a sample of women in weight training. Findings suggest that both DT and DM might increase the risk for ED and BDD pathology. Women in fitness sports, therefore, need to be a target for screening and further investigation, in particular given the potential harmful use of appearance- and performance-enhancing drugs and supplements. Furthermore, the newly developed fat-muscle silhouettes allow for a more face-valid assessment of fat/muscle body image ideals.

15.5: Examination of the Mechanisms Accounting for the Protective Effect of Media Literacy on Body Dissatisfaction

Rachel Rodgers, PhD, Northeastern University, Boston, Massachusetts, Siân McLean, PhD, La Trobe University, Melbourne, Victoria, Susan Paxton, PhD; FAED, La Trobe University, Melbourne, Victoria

Media literacy has been highlighted as a critical protective factor, conferring resilience to media pressures that increase eating disorder risk. However, to date, the mechanisms underlying the protective effect of media literacy are poorly understood. The aim of the present study was to bridge this gap, by exploring the moderating role of three dimensions of media literacy on the relationship between media exposure and body dissatisfaction, mediated by thin-ideal internalization and appearance comparison among early adolescent girls. A sample of 284 early adolescent girls, (Mean age = 13.15 years), completed self-report measures of media exposure, thin-ideal internalization, appearance comparison, body dissatisfaction, and three dimensions of media literacy including: realism scepticism, similarity scepticism, and critical thinking. Moderated mediation analyses were conducted. Findings revealed different patterns of relationships for the different dimensions of media literacy. Specifically, similarity scepticism moderated the mediated relationship between media exposure and body dissatisfaction via both thin-ideal internalization (p = .003) and appearance comparison (p = .004). In contrast, realism scepticism and critical thinking were not found to moderate the mediated pathways between media exposure and body dissatisfaction. Findings suggest that similarity scepticism, the scepticism that one’s body could resemble unrealistic images in media, may be a critical component of media literacy in mitigating the effects of media exposure on body dissatisfaction. Further longitudinal and experimental research should aim to confirm these findings and clarify the role of other dimensions of media literacy with a view to informing prevention efforts.

15.6: Effectiveness of a Dissonance-Based Intervention for Girl Guides and Girl Scouts on Body Image and Disordered Eating: Results from a Global Pragmatic Controlled Trial

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

The World Association of Girl Guides and Girl Scouts (WAGGGS), the world’s largest youth organization for girls, partnered with the Dove Self-Esteem Project to implement a global body image intervention, Free Being Me (FBM). Using a community participatory approach, FBM was adapted from the dissonance-based eating disorder (ED) prevention intervention The Body Project. A train-the-trainer dissemination model was created, with global trainers delivering national trainings, after which trainees cascaded training down to local groups. Since 2013, FBM has been disseminated in over 120 countries to 3 million young people. The purpose of this study was to conduct a pragmatic controlled trial to explore the effectiveness of Free Being Me on body image, disordered eating, and related risk factor outcomes among universal samples of girl guides and girl scouts in nine countries (UK, Argentina, Germany, Hong Kong, India, Japan, Netherlands, Taiwan, Thailand). Girls (N = 371; Mage = 13.9) in each country were assigned to receive the intervention delivered over four weeks by a community leader, or, a waitlist control group. Standardised self-report measures of body image, disordered eating, and related risk factors were administered at pre- and post-intervention. After controlling for baseline scores, girls who received Free Being Me reported significant reductions in body dissatisfaction, eating disorder symptoms, behavioural avoidance due to body image concerns, negative affect, and thin-ideal internalisation compared to waitlist controls. The pattern of results varied somewhat between countries. These findings provide preliminary evidence to support the scaling up of body image and eating disorder prevention interventions. Future research with larger sample sizes and longer-term follow-up, particularly among middle- and low-income countries often ignored in ED prevention research, will be beneficial to further understanding of the effectiveness of global dissemination efforts for evidence-based body image and ED prevention interventions.

Topic: Gender, Ethnicity, Culture
Meeting Hall 1A, first floor
Lindsay Bodell, PhD & Kendrin Sonneville, ScD, RD

16.1: Eating Disorder Examination Questionnaire: Norms for Transgender Youth

Claire Peterson, PhD, Cincinnati Childen's Hospital Medical Center, Cincinnati, Ohio, Michael Toland, PhD, University of Kentucky, Lexington, Kentucky, Abigail Matthews, PhD, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, Samantha Mathews, BS, Xavier University, Cincinnati, Ohio, Lee Ann Conard, DO, RPh, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio

The purpose of this study was to contribute new data about the factor structure of the Eating Disorder Examination Questionnaire (EDE-Q), and for the first time, provide normative data on the EDE-Q for transgender (TG) youth. This sample consisted of 186 male-to-female (MTF; n = 51) and female-to-male (FTM; n = 135) TG youth age 12-23 (M = 17.01, SD = 2.89) with a mean BMI of 26.31 (SD = 7.6). The EDE-Q was best represented by a bifactor solution with 3 orthogonal factors (restraint, eating concern, shape/weight concern), c2(209) = 396.353, p < .001, RMSEA = .078, CFI = .973, and WRMR = 0.831. Furthermore, 81% of the common variability running through all EDE-Q items was due to a common dimension (eating disorder) and that use of a one-factor solution will lead to biased parameter estimates based on the multidimensional data (ECV = 0.81, PUC = 0.63). Multiple group analyses did not show latent mean differences (estimate = 0.06, z = 0.35, p = .73) between MTF and FTM on the general eating disorder dimension. Plus, no differences were found between observed means (estimate = 0.09) based on MTF status, z = -0.40, p = .59, d = 0.07 (MTF: M = 1.45, SD = 1.33, FTM: M = 1.54, SD = 1.29). As a whole group, observed mean EDE-Q Global score was 1.52 (SD = 1.3). In the last month 20% endorsed objective binge eating episodes while only 3% endorsed self-induced vomiting. Results indicate no differences between MTF and FTM transgender youth on EDE-Q Global scores. There were no gender differences observed for self-induced vomiting. However, MTF youth were more likely to endorse binge eating than FTM youth (z = 2.16, p = .03, estimate = 0.88). Overall, transgender norms on the EDE-Q are similar to cisgender samples. Further, the bifactor model of the EDE-Q is recommended for modeling purposes. For individual assessment scoring purposes the Global scale can be interpreted as a unidimensional reflection of eating disorder symptoms. However, raw subscale scores are not recommended.

16.2: Prospective Associations between Childhood Gender Conformity, Bullying Victimization, and Adolescent Disordered Eating Behaviors in a UK Cohort Study

Jerel Calzo, PhD; MPH, San Diego State University, San Diego, California, S. Bryn Austin, ScD; FAED, Boston Children's Hospital and Harvard Medical School, Boston, Massachusetts, Stuart Murray, DClinPsy; PhD, University of California, San Francisco, San Francisco, California, Allegra Gordon, MPH; ScD, Boston Children's Hospital and Harvard Medical School, Boston , Massachusetts, Nadia Micali, MRCPsych; PhD; MD; FAED, Icahn School of Medicine at Mount Sinai, New York, New Mexico

The internalization of cultural appearance ideals may be one dominant pathway by which gender conformity (i.e., adherence to masculinity and femininity norms) in childhood influences disordered eating behaviors (DEB). The processes by which conformity and nonconformity to gender norms are policed, which could include bullying and social exclusion, may create an additional pathway to DEB. We examined associations among childhood gender conformity, bullying victimization, and adolescent DEB in a representative cohort study of UK youth (Avon Longitudinal Study of Parents and Children; 2,367 males, 2,681 females). Childhood conformity to masculinity and femininity were assessed at age 8 years; past-year victimization (bullying [teasing; being hit]; social exclusion) at age 10; and frequency of past-year purging, overeating, and binge eating at age 16 via self-report questionnaire. Gender stratified, prospective path analyses (adjusting for sexual orientation, weight status, socioeconomic status, body dissatisfaction) revealed significant indirect paths from childhood gender conformity, bullying victimization, and adolescent binge eating. In females, higher conformity to masculinity (β=0.02) and femininity (β=0.03) at age 8 were associated with greater victimization at age 10 (p’s <0.01), and greater victimization predicted greater frequency of binge eating (β=0.17). The indirect paths from gender expression to binge eating via victimization were significant (p’s <0.01). In males, higher conformity to masculinity (β=0.03) and femininity (β=0.02) were associated with greater victimization at age 10 (p’s <0.001), and greater victimization was linked to greater frequency of binge eating in adolescence (β=0.13, p<0.001). The indirect paths from gender expression to binge eating via victimization were significant (p’s <0.01). Results indicate that victimization experiences may represent an additional intermediate mechanism linking gender expression to eating disorder risk.

16.3: Black Beauty: Exploring Body Ideals Among Black Women

Alice S. Lowy, MA, Northeastern University, Boston, Massachusetts, Elizabeth S. Cook, MS, Northeastern University, Boston, Massachusetts, Debra L. Franko, PhD; FAED, Northeastern University, Boston, Massachusetts, Rachel F. Rodgers, PhD, Northeastern University, Boston , Massachusetts

Available instruments measuring body dissatisfaction often reflect White Western ideals that may not capture nuances of other racial and ethnic perceptions of attractiveness. It has been suggested that Black women in particular may not endorse the thin ideal and instead attribute importance to culturally specific body features. This may protect them from body dissatisfaction and eating disorder symptoms. The purpose of this study was to examine the associations between Black women’s endorsement of mainstream and cultural specific beauty ideals and eating disorder symptoms. The Black Body Ideals Scale (BBIS) was developed to assess beliefs about what Black women consider to be ideal body features, as well as personal investment in those beliefs. A sample of 49 Black female college students, mean age = 22.9 (5.5) years, completed an online questionnaire, including the BBIS, Eating Disorder Diagnostic Scale-5 (EDDS-5) and the Female Drive for Muscularity Scale (FDMS). Preliminary analyses revealed that culturally specific ideals were associated with drive for muscularity (r = .37, p = .02). Interestingly, small but inverse associations were found between eating disorder symptoms and culturally specific ideals (r = -.16) and Eurocentric ideals (r = .14). These results suggest that endorsement of culturally specific features might protect against eating disorders among young Black women, whereas endorsement of Eurocentric features may increase risk. Such findings support using an intersectional approach in future prevention efforts that addresses cultural body ideals in relation to body image and eating pathology.

16.4: Reducing the Stigma of Eating Disorders: A Meta-Analysis and Narrative Synthesis.

Joanna Doley, Student; BPsyc(Hons), La Trobe University, Melbourne, VIC, Laura Hart, BA; BSc; PhD, La Trobe University The University of Melbourne, Melbourne, VIC, Arthur Stukas, BA; PhD, La Trobe University, Melbourne, VIC, Katja Petrovic, MPsych, La Trobe University, Melbourne , VIC, Ayoub Bouguettaya, BSc in Psychology (Hons), Deakin University (Burwood campus), Melbourne, VIC, Susan Paxton,; BA (Hons); MPsych; PhD; FAED, La Trobe University, Melbourne, VIC

Stigma is recognized as a problem for individuals with eating disorders (EDs), as it forms barriers to disclosure and help-seeking. Interventions to reduce ED stigma may assist in removing these barriers; however it is not yet known which types of intervention are effective. This systematic review examined the effectiveness of intervention types, and identified gaps in the literature. We identified eligible studies through four databases, and relevant LISTSERVs. Two independent raters performed screening, data extraction, and quality assessment. We conducted meta-analyses on etiological explanations of ED (comparing biological, multifactorial, and sociocultural) and stigma reduction strategy (education and contact). Other study characteristics of interest were examined in a critical narrative synthesis. Eighteen papers were eligible for narrative synthesis, with ten also eligible for inclusion in meta-analysis. Interventions emphasizing biological etiology of ED resulted in lower stigma than those emphasizing multifactorial or sociocultural etiology, most notably in attitudinal stigma (biological versus sociocultural explanations; N=5, g =.47, p <.001; biological versus multifactorial explanations N=3, g =.48, p <.001). Education-only interventions had larger reductions in stigma (N=2, g =.31, p =.45) than those including contact (N=4, g =.17, p =.18), but these results were less reliable than findings for etiological interventions. Most studies examined stigma related to Anorexia Nervosa or EDs as a general diagnostic category and had mostly female, undergraduate participants. There was inconclusive evidence on iatrogenic effects of interventions. Despite their effectiveness, research is needed to verify that biological explanations do not cause unintentional harm, such as biological essentialism. Future research should examine the role of in vivo contact, directly compare education and contact strategies, and aim to generalize findings to community populations.

16.5: Developmental Trajectories of Eating Disorder Symptoms in Black and White Girls

Lindsay Bodell, PhD, The University of Chicago, Chicago, Illinois, Jennifer Wildes, PhD, The University of Chicago, Chicago, Illinois, Yu Cheng, PhD, University of Pittsburgh, Pittsburgh, Pennsylvania, Andrea Goldschmidt, PhD, Warren Alpert Medical School of Brown University/The Miriam Hospital, Providence , Rhode Island, Kate Keenan, PhD, The University of Chicago, Chicago, Illinois, Alison Hipwell, PhD, University of Pittsburgh, Pittsburgh, Pennsylvania, Stephanie Stepp PhD, University of Pittsburgh, Pittsburgh, Pennsylvania

Epidemiological research suggests racial differences in the temporal course and presentation of eating disorder symptoms. However, no studies have examined associations between race and eating disorder symptom trajectories across youth and adolescence, which is necessary to inform culturally sensitive prevention interventions. The purpose of the current study was to examine the trajectories of eating disorder symptoms from childhood to young adulthood and to examine whether race is associated with trajectory group membership. Data were drawn from 2,305 Black and White girls who participated in a community-based longitudinal cohort study examining the development of psychopathology (The Pittsburgh Girls Study). The child and adult versions of the Eating Attitudes Test assessed self-reported eating disorder symptoms at six time points between ages 9 and 21 years. Group-based trajectory modelling was used to examine developmental trajectories of dieting frequency, binge eating, vomiting, and overall eating disorder psychopathology. Three or four distinct developmental patterns were found for each disordered eating construct, including none, increasing, decreasing, or stable trajectories. White race was associated with a greater likelihood of being in the increasing and stable dieting trajectories and the increasing eating disorder psychopathology trajectory group. Black race was associated with a greater likelihood of being in the increasing or decreasing vomiting trajectories and the decreasing eating disorder psychopathology trajectory group. These results highlight the importance of examining eating disorder symptoms by racial background and the potential need for differences in the timing and focus of prevention interventions in these groups.

16.6: Community Based Nutrition Education to Reduce Geophagic Practices and Improve Nutrition in Women of Reproductive Age in Nakuru Municipality, Kenya: A Pilot Study

Sharon Iron-Segev, DSc, R.D., The Hebrew University of Jerusalem, Rehovot, IL, Janerose Nasimiyu Lusweti, MSc, The Hebrew University of Jerusalem, Rehovot, Kenya, Elizabeth Kamau-Mbuthia, PhD, Egerton University, Nakuru, Kenya, Aliza Stark, PhD, R.D., The Hebrew University of Jerusalem, Rehovot, IL

Geophagia, the deliberate consumption of earth, soil, stones or clay, is considered a type of pica. It is prevalent among pregnant and breastfeeding women, and is widespread in sub-Saharan Africa. In Nairobi Kenya, reported rates are at an alarming level of 74%. Health risks from geophagia include increased exposure to heavy metals and parasites and micronutrient deficiencies. Poor nutritional status coupled with geophagic practices are known to contribute to poor pregnancy outcomes. This study aimed to reduce geophagy and improve diet through nutrition education. A cross sectional study was conducted in 135 women of reproductive age (15-49 y), in low socioeconomic areas of Nakuru, Kenya. Questionnaires including dietary recall and a focus group discussion were used to determine knowledge, attitude and practices in geophagia. This was followed by a nutrition education intervention and program evaluation. Geophagia cut across all ages with 91(67.4%) consuming ≥100 g/day. Strong cravings were reported by 81.5% as the primary reason for geophagic practices. Sociodemographic factors were not associated with the amount of geophagic material consumed or dietary practices. The most common negative effect reported by participants was constipation (16.3%). Nutrition education focusing on geophagia significantly elicited a decrease in amount of geophagic material consumed per day in 82 (77%) of participants (Z= -7.914, p<.000). After intervention the proportion of participants who chose to make half their plate vegetables using the healthy plate model increased significantly from 12% to 76.4%. A significant increase in dietary diversity was also observed (Z= -3.058, p<.002). Nutrition education was shown to be an effective approach for reducing geophagic practices in rural Kenya and improving overall nutrient intake. This pilot intervention showed that it is important to identify women at risk and provide nutrition education to combat this eating disorder.

Topic: Innovative Uses of Technology
Meeting Hall 1B, first floor
Cheri Levinson, PhD & Markus Moessner, PhD

17.1: Preliminary Findings from a Randomized-Controlled Trial of BodiMojo: A Mobile App for Positive Body Image

Rachel Rodgers, PhD, Northeastern University, Boston, Massachusetts, Elizabeth Donovan, PhD, Bodimojo, Boston, Massachusetts, Tara Cousineau, PhD, Bodimojo, Boston, Massachusetts, Elizabeth Cook, MA, Northeastern University, Boston , Massachusetts, Kayla Yates, BA, Northeastern University, Boston, Massachusetts, Alice Lowy, MA, Northeastern University, Boston, Massachusetts, Debra Franko PhD; FAED, Northeastern University, Boston, Massachusetts

Mobile technology presents a high potential as a means of delivery for interventions aiming to develop positive body image. Furthermore, self-compassion has emerged as a useful framework for body image interventions. The aim of the present study was to examine the efficacy of a randomized-controlled evaluation of BodiMojo, a mobile app based intervention grounded in self-compassion aiming to promote positive body image. A sample of 273 adolescents and young adults, mean (SD) age = 18.36 (1.34), 75% female, participated in the study. Participants were randomly allocated to either an assessment-only control group or the intervention group, in which they were asked to use BodiMojo for 6 weeks. The intervention consisted of participants receiving two daily text messages containing a quiz, an audio mediation, an affirmation, or a behavioral tip. In addition, participants were asked to record their mood in the app once a day and to keep a daily gratitude journal. At baseline and post-test, participants completed measures of body esteem, social comparison, and dimensions of self-compassion including self-judgment. Findings revealed significant time (pre vs. post) and group (intervention vs. control) interactions for body esteem (p = .044) and self-judgment (p = .047) in the expected directions, such that body esteem showed a greater increase in the intervention group as compared with the control group, and self-judgment revealed a greater decrease in the intervention group relative to the control group. No significant group effects were found for social comparison. These findings provide preliminary support for BodiMojo as a cost-effective mobile app intervention for promoting positive body image. Interventions grounded in self-compassion theory may have the potential to promote positive body image by targeting negative self-evaluations.

17.2: Differences in Affective Trajectories in Ecological Momentary Assessment Studies of Binge Eating

Brittany Stevenson, MS; Student, University of Central Florida, Orlando, Florida, Robert Dvorak, PhD, University of Central Florida, Orlando, Florida, Stephen Wonderlich, PhD, University of North Dakota, Grand Forks, North Dakota, Ross Crosby, PhD, Neuropsychiatric Research Institute, Fargo , North Dakota, Kathryn Gordon, PhD, North Dakota State University, Fargo, North Dakota

The affect regulation model of binge eating suggests that negative mood increases leading up to a binge-eating episode, and then dissipates following the episode, resulting in a negatively reinforced behavior. However, a recent meta-analysis suggested that negative mood actually increases following a binge-eating episode. It has been suggested that the findings of this meta-analysis were due to differences in the timing of pre-binge versus post-binge assessments; pre-binge assessments are usually not anchored to the binge event itself, whereas post-binge assessments are often initiated because of the binge episode (i.e., they are event-contingent assessments). However, in the current study, these assessment schedules were reconciled; the study utilized only random assessments, so that assessments would be evenly spaced from each binge event. The current study also analyzed loss of control (LOC) eating in lieu of binge eating, due to evidence that loss of control eating is the most clinically meaningful concept. When comparing pre-LOC affect to post-LOC affect, we found significant differences in guilt, anger, anxiety, and emotional instability. In support of the affect regulation model, emotional instability and anger appeared to increase prior to, and decrease following, LOC eating. Anxiety exhibited a marked decrease after LOC eating, and guilt rapidly increased prior to LOC eating, with a further increase and slight decrease afterward. In order to control for time-of-day effects, we also compared LOC days to non-LOC days, matching LOC days to non-LOC days using each individuals’ average LOC time. We found that guilt, sadness, and emotional instability increased leading up to the LOC episode, but no moods decreased significantly more on LOC days than non-LOC days, contrary to within-days analysis results. When controlling for time-of-day effects and utilizing random-only assessments, results do not support a negative reinforcement model of LOC eating.

17.3: Mealtime Eating Disorder Cognitions Predict Eating Disorder Behaviors: A Mobile Technology Based Ecological Momentary Assessment Study

Cheri Levinson, PhD, University of Louisville, Louisville, Kentucky, Laura Fewell, BA, McCallum Place, St. Louis, Missouri, Leigh Brosof, BA, University of Louisville, Louisville, Kentucky, Lauren Fournier, Student, Washington University in St. Louis, St. Louis , Missouri, Eric Lenze, MD, Washington University in St. Louis, St. Louis, Missouri

Individuals with eating disorders struggle to eat during meals, which leads to significant weight loss and relapse. However, little is known about what contributes to difficulty eating meals such as how eating disorder (ED) cognitions that occur during a meal contribute to ED behaviors. We used mobile technology to test which mealtime ED cognitions predicted daily ED behaviors. Using the technique of ecological momentary assessment (EMA), we tested the ED cognitions of perfectionism (concerns about making mistakes during a meal), anxiety during a meal, avoiding emotions during a meal, ruminating about a meal, and worrying about weight gain during a meal as predictors of ED behaviors (e.g., restriction, binge eating, vomiting, excessive exercise, and body checking). Participants (N=61 individuals with an ED) completed 28 at-home ecological self-assessments across one week utilizing a mobile application. They then completed an assessment of ED symptoms at one-month follow up. Concerns about making mistakes during a meal (b=.16, p=.018) significantly predicted daily restriction. Rumination about a meal (b=.17, p=.004) predicted later binge eating. Worrying about weight gain during a meal (b=.07, p=.048) predicted later vomiting. Avoiding emotions during a meal predicted later excessive exercise (b=.08, p=.010). Finally, avoiding emotions during the meal (b= .14, p=.008), worrying about weight gain during a meal (b=.17, p<.001) and rumination about a meal (b=.14, p=.006) predicted later body checking. At one month follow up, fears of gaining weight (Wald=7.78, p=.005) and rumination about a meal (Wald=5.68, p=.017) predicted drive for thinness. This study pinpoints specific ED cognitions occurring during mealtimes that predict ED behaviors. We conclude that mobile technology-based EMA is ideal for detecting ED cognitions. Interventions could be developed to target these cognitions (e.g., purposively making mistakes; approaching emotions) to reduce ED behaviors. These interventions have the potential to promote healthy mealtime experiences and prevent ED relapse.

17.4: Risk for Eight Specific Eating Disorders, Obtained by a Virtual Assistant ("NUTMIN"), for the Support of Diagnosis Process of Eating Disorders: A Validation Study

Ana Olivia Caballero-Lambert, MPC, Universidad Iberoamericana Leon, Leon, Guanajuato, Marcelo Funes-Gallanzi, MPhil; PhD, AVNTK, SC, Guadalajara, Guanajuato, Ana Olivia Caballero-Lambert, MPC, Universidad Iberoamericana Leon, Leon, Guanajuato, César Horacio Torres-Montañez, ISC, AVNTK, SC, Guadalajara , Jalisco, Daniela Irazu Lopez-Gomez, LN, Universidad Iberoamericana Leon, Leon, Guanajuato, Krisein Alejandra Martinez-Fuentes, LNCA, Universidad Iberoamericana Leon, Leon, Guanajuato, Martha Leticia Guevara-Sangines PhD, Universidad de Guanajuato, Guanajuato, Guanajuato

Artificial intelligence has allowed several advances in health areas. “Nutmin” was developed as an Android app, using technology already in use by a general cognitive engine called Rachael Repp. This conversational assistant is capable of six logic operations (equivalence, similarity, induction, deduction, abduction, and metaphor); it is also able to learn free unstructured text and it is capable of improving itself. In a combined effort between Universidad Iberoamericana León and AVNTK, SC., it was possible for "Nutmin" to identify the user's risk for developing eight different eating disorders (ED´s) - anorexia nervosa, bulimia nervosa, binge eating disorder, avoidance/restrictive disorder, pica, rumination, purging disorder and night eating syndrome. To achieve this, a qualitative study was conducted involving patients with eating disorders and health experts in these diseases. A questionnaire was used to find typical eating disorders phrases and confirmatory questions were added. Currently, a study is being conducted in a 484 adolecent sample, 15-19 y, both sexes, to validate the risk for these eight eating disorders determined by "Nutmin". The risk for these ED´s is compared to body mass index, food intake obtained by a validated 24-hour recall, abnormal eating behaviors using Unikel´s validated Mexican brief questionnaire to measure the risk of abnormal eating behaviors, food selection using a Mexican questionnaire to identify food selection motives, and self-image using a validated scale. The aim of this research is to obtain a valid instrument that could be used by any physician worldwide as a diagnostic support to increase ED´s detection, opportune referral and treatment. At the conference, preeliminary data will be presented. "Nutmin" is being translated to 10 different languages including English, Spanish, Italian, French and Russian. It is important to point out that after obtaining a percentage of risk for any of the eight ED´s, "Nutmin" directs the user to a health and nutrition orientation section. Finally, the user can be referred to a specialized clinic or an independent professional. As said before, this app has the objective to be of diagnostic support, at no time it seeks to take the place of a well-trained eating disorders professional.

17.5: Between- and Within-Subjects Analysis of an Individualized Internet-Based Program for Prevention and Early Intervention: Associations Between Eating Disorder Symptoms and Program Utilization.

Sally Kindermann, DiplPsych, Center for Psychotherapy Research, University Hospital Heidelberg, Heidelberg, Baden-Wurttemberg, Markus Moessner, DPhil, Center for Psychotherapy Research, University Hospital Heidelberg, Heidelberg, Baden-Wurtemberg, Fikret Ozer, BSc, Center for Psychotherapy Research, University Hospital Heidelberg, Heidelberg, Baden-Wurtemberg, Stephanie Bauer, DSc, Center for Psychotherapy Research, University Hospital Heidelberg, Heidelberg , Baden-Wurttemberg

Internet-based interventions have proven their potential for the prevention of eating disorders (ED). Yet, it is challenging to engage individuals in prevention programs, especially if they do not experience significant psychological impairment. Research has shown that individuals with substantial ED symptoms are more likely to engage in publically available online programs than at-risk populations. Instead of providing a structured intervention that offers the same amount of support to all participants, a more flexible individualized approach as offered by the online program ProYouth allows to adjust the intensity of support to participants’ individual needs and thus might be suitable for a broad spectrum of participants. The aim of the present study was to investigate associations between ED related impairment and program usage both on a between-subjects and a within-subjects level. The sample consists of N = 396 German ProYouth participants who used the program between 2 and 52 months. Generalized estimated equations (GEEs) were used to examine associations between self-reported ED symptoms monitored throughout participation and program utilization assessed via server logs. The results show that participants with high levels of weight and shape concerns and those with high frequencies of dieting had significantly higher numbers of page visits. Within-subjects dieting showed the strongest associations with the intensity of program utilization. Results indicate that both between- subjects and within-subjects intensity of program utilization is associated with the level of impairment. The flexible approach of ProYouth yields a good fit of individual needs and the intensity of support. Especially for prevention and early intervention an individualized, needs-based approach appears promising.

17.6: ProYouth OZ: An Online Peer-to-Peer Support Prevention and Early Intervention Program for Young People at Risk of Eating Disorders

Kathina Ali, BSc; MSc, The Australian National University, Canberra, ACT, Louise Farrer, PhD, The Australian National University, Canberra, ACT, Daniel Fassnacht, DiplPsych; PhD, The Australian National University, Canberra, ACT, Elizabeth Rieger, BA; MPsych; PhD, The Australian National University, Canberra , ACT, Markus Moessner, DiplPsych; PhD, University Hospital Heidelberg, Heidelberg, Baden-Wuerrtemberg, Kathleen M Griffiths, BSc; PhD, The Australian National University, Canberra, ACT, Stephanie Bauer DiplPsych; PhD, University Hospital Heidelberg, Heidelberg, Baden-Wuerrtemberg

Evidence suggests that Internet-based approaches are effective for the prevention and treatment of eating disorders and peer-to-peer support is commonly used as a component of these interventions. However, the effectiveness of online peer-to-peer support for eating disorders has not yet been examined. The current study will evaluate ProYouth OZ, an Internet-based prevention and early intervention program. The program consists of a number of modules including psycho-education, a supportive monitoring and feedback system, and peer support delivered through synchronous weekly chat sessions. Young people at risk of eating disorders were randomly assigned to one of the three study conditions: (1) ProYouth OZ including peer-to-peer support (participation in weekly moderated peer support group chats), (2) ProYouth OZ without peer-to-peer support, and (3) a waitlist control group. Assessments were conducted at baseline, post-intervention (6 weeks), at 3- and 6-month follow-up to examine disordered eating behaviors, help-seeking barriers (e.g., stigma, eating disorder literacy), and help-seeking intentions. Recruitment commenced October 2016 and preliminary data from the first recruitment wave will be presented. Internet-based interventions offer a promising approach to preventing and treating eating disorders. This is one of the first studies to examine the additional effect of synchronous peer-to-peer support in an Internet-based prevention and early intervention program for eating disorders. This study will also provide a better understanding of whether peer support can successfully reduce barriers to care and increase help-seeking in young people.

Topic: Diagnosis, Classification and Measurement
Panorama, first floor
Ross Crosby, PhD, FAED & Pamela Keel, PhD, FAED

18.1: Eating, Food, and Substance Versus Behavioral Addiction: Investigating the Validity of the Eating Addiction Questionnaire

Kristin von Ranson, PhD; FAED, University of Calgary, Calgary, Alberta, Melissa King-Hope, BA, University of Calgary, Calgary, Alberta

Food addiction (FA) is theorized to contribute to rising obesity rates. Although controversial, FA theory holds that, like alcohol and other substances of abuse, hyper-palatable foods are addictive. Alternatively, it was recently proposed that people may become addicted to the act of eating, rather than to food as a substance; i.e., eating behaviours may be experienced as behavioural rather than substance addictions. This critical distinction may impact addiction-influenced treatment approaches used for eating disorders. The purpose of this study was to evaluate the convergent and divergent validity of the first measure of eating addiction, the Eating Addiction Questionnaire (EAQ). Female and male undergraduates (N = 576) completed a first version of the EAQ, and after items were revised according to detailed feedback from 8 eating disorders and obesity experts, 428 participants completed a second version. Participants also completed self-report measures of similar (FA, eating psychopathology, food craving and grazing), and dissimilar constructs (alcohol and illicit substance abuse). Principal components analyses of EAQ items revealed 1 factor. EAQ scores were moderately correlated (r = .50) with FA scores, indicating shared and unique variance. Stronger correlations were observed with eating-related constructs such as binge eating (r = .70), food cravings (r = .66), and grazing (r = .58), and weaker correlations were seen with non-eating-related substance addictions of drug abuse (r = .10) and alcohol abuse (r = .12). This pattern of findings suggests eating addiction is distinct from FA as well as several other types of eating and addictive symptoms, and provides preliminary support for the convergent and divergent validity of the EAQ. Additional research is needed to help disentangle the complex issue of whether and how food and eating are addictive, and how FA and eating addiction may relate to existing constructs such as binge eating.

18.2: Investigating the DSM-5 Severity Specifiers Based on Body Mass Index for Anorexia Nervosa

Oyvind Ro, MD, Regional Department for Eating Disorders, Oslo University Hospital, Norway, Oslo, Norway, Deborah L. Reas, PhD, Regional Department for Eating Disorders, Oslo University Hospital, Norway, Oslo

The DSM-5 includes severity indictor for anorexia nervosa (AN) based on current body mass index (BMI). This study investigated differences in global ED pathology, eating disorder behaviors, and eating disorder-related impairment across DSM-5 severity levels in a clinical sample of AN. A treatment-seeking sample of 146 AN patients (6 men, 140 women) were categorized as 34 (23.3%) mild (>=17.0 BMI), 35 (24.0%) as moderate (16-16.99 BMI), 32 (21.9%) as severe (15-15.99 BMI), and 45 (30.8%) as extreme (.100 except EC = .07). Similarly, there were no significant subgroup differences on laxative use, vomiting, binge eating, or excessive exercise (all p’s > .100). Likewise, ED-specific impairment as measured by the Clinical Impairment Assessment global score showed no significant differences between sub-groups (.279). Bivariate correlational analyses for the total sample showed significant and positive associations between BMI and restriction (p = .022) and the global EDE-Q (p = .04). We found little empirical evidence to support the DSM-5 severity grouping based upon on current BMI in adults, although dimensional analyses did indicate a significant bivariate relationship between increasing BMI and higher ED pathology, especially increased restraint.

18.3: Are Eating Disorders and Related Symptoms Risk Factors for Suicidal Thoughts and Behaviors? A Meta-Analysis

April Smith, PhD, Miami University, Oxford, Ohio, Elizabeth Velkoff, BA, Miami University, Oxford, Ohio, Jessica Ribeiro, PhD, Florida State University, Tallahassee, Florida, Joseph Franklin, PhD, Florida State University, Tallahassee , Florida

The purpose of the present meta-analysis was to examine whether eating disorders (EDs) are longitudinal predictors of later suicidal behavior. EDs are widely believed to be risk factors (i.e. longitudinal predictors) for suicidal thoughts and behaviors. Although prior cross-sectional research and data from standardized mortality ratio studies provide indirect evidence for this belief, longitudinal research is mixed. Thus, questions remain about whether ED factors are risk factors for subsequent suicidal thoughts and behaviors. Papers published through January 1st 2015 were identified through literature searches using PubMed, PsycInfo, and Google Scholar. We identified a total of 2,541 longitudinal studies. Inclusion required that studies include at least one longitudinal analysis predicting a discrete suicide outcome (i.e., suicide ideation, suicide attempt, or suicide death) using any ED-relevant factor. A total of 11 studies (35 statistical tests) met inclusion criteria. A random effects model was used for meta-analyses; meta-regression analyses used a random-effects model with unrestricted maximum likelihood. Results indicated that EDs and their symptoms were significant but weak predictors of suicide attempts but not death. The strongest predictor identified in this meta-analysis was ED diagnosis predicting suicide attempt, and was associated with an odds ratio of 2.19. Effects remained weak when moderators such as sample age, sample severity, and length of follow-up were considered. These findings suggest that a reconsideration of the relationship between EDs and suicide is needed. Further, by reviewing the methodological limitations of previous research, these results highlight avenues for future research that can advance our understanding of the relationship between EDs and suicidal thoughts and behaviors.

18.4: The Predictive Validity of Purging Disorder: A Comparison to Bulimia Nervosa at 10-Year Follow-Up

K. Jean Forney, MS, Florida State University, Tallahassee, Florida, Pamela Keel, PhD; FAED, Florida State University, Tallahassee, Florida

Purging disorder (PD) has been included in the DSM-5 as an Other Specified Eating Disorder and is distinguished from bulimia nervosa (BN) by the absence of objectively large binge eating episodes. Limited data exist describing the long-term outcome of PD and the predictive validity of the PD-BN distinction. 218 women (87 PD; 131 BN) were invited to complete follow-up assessments approximately 10 years after baseline. Thus far, 64.4% of women with baseline PD (n=56) and 64.1% of women with baseline BN (n=78) have completed diagnostic interviews and questionnaires at mean (SD) 10.00 (3.99) years following baseline assessments. Participants had a mean (SD) age of 33.43( 7.57) years at follow-up. Of those with PD at baseline, 33.33% (n=17) had achieved full or partial remission; in contrast, 46.8% (n=36) of those with BN at baseline had achieved full or partial remission. PD and BN did not differ in likelihood of remission, p=.15. Baseline diagnosis was associated with follow-up diagnosis, such that stability was more likely than cross-over for both PD (59% stability) and BN (74% stability) (p=.02). Individuals with PD reported a smaller decrease in purging frequency over time relative to those with BN at a trend level, p=.056. Women with PD at baseline also reported a smaller decrease in shape concerns than those with BN at baseline, p=.01. Results support the longitudinal stability of PD’s clinical presentation. Distinguishing between PD and BN provides information about course, as PD appears to be associated with a more chronic course. Data collection will be completed in spring 2017.

18.5: Primary Care Assessment and Triage of Adolescent Patients with Anorexia Nervosa

Kendra Homan, PhD, Mayo Clinic, Rochester, Minnesota, Leslie Sim, PhD, Mayo Clinic, Rochester, Minnesota, Lisa Kransdorf, MD; MPH, UCLA Division of Internal Medicine and Pediatrics, Los Angeles, California, Susan Crowley, PhD, Utah State University, Logan , Utah, Jocelyn Lebow, PhD, Mayo Clinic, Rochester, Minnesota, Margo Scott, Student, Creighton University, Omaha, Nebraska

Medical complications in adolescent eating disorders are common and often life threatening. In acute situations, hospitalization may be required to achieve medical, nutritional and/or psychiatric stability. The American Academy of Pediatrics has published specific criteria outlining when hospitalization is advised for adolescents with anorexia nervosa (AN). However, many adolescents may not be receiving appropriate care. Barriers include lack of physician awareness and poor adherence to these guidelines. The purpose of the present study was to examine the assessment and triage that adolescents with AN receive in primary care. A retrospective cohort review of all adolescent patients (ages 10-18) who were diagnosed with AN at Mayo Clinic between 2010 and 2016 was conducted. The cohort consisted of 69 patients (M age=13.9 years, SD=2.1; 85.5% female; 94.2% White). Criteria for hospitalization and associated treatment recommendations were abstracted from patients’ primary care appointment in which eating and/or weight concerns were first identified. Results indicated that during the initial episode of care criteria for hospitalization were inconsistently assessed. Although blood pressure was collected for the majority of patients (n=59, 85.5%), less than half received measurements of heart rate (n=29, 42.0%), temperature (n=29, 42.0%), or assessment for cardiac arrhythmia (e.g., electrocardiogram; n=23, 35.9%). Even fewer patients were assessed for percentage weight loss (n= 9, 13.0%), orthostatic changes (n=8, 11.6%), or body fat percentage (n= 0). Though most patients received inadequate assessment, 34.8% (n= 24) met at least one criterion for hospitalization. Of these, only eight patients (33.3%) were hospitalized. Results suggest that adolescent patients are not receiving medical assessment and triage consistent with practice guidelines. Implications for medical education are discussed.

18.6: Getting Shredded: Development and Validation of a Disordered Eating Measure for Increasing Muscularity and Leanness

Marita Cooper, MPsych, Australian National University, Centre for Integrative Health, Brisbane, Queensland, Kate Pollard, Bachelor of Nut & Diet (Hons), Centre for Integrative Health, Brisbane, Australian Capital Territory, Kathleen Griffiths, DPhil, Australian National University, Canberra, Australian Capital Territory

Male societal ideals are increasingly promoting unrealistic mesomorphic body types. These body types endorse a combination of high muscularity and leanness equally unattainable as thin ideals commonly idealised in individuals with Anorexia and Bulimia Nervosa. This study aimed to develop a valid and reliable scale to measure disordered eating behaviours that promote muscularity and leanness. The Eating for Muscularity Scale (EMS) was developed through review of relevant literature as well as consultation with experts in the areas of Muscle Dysmorphia, eating disorders and dietetics. This process culminated in the development of a broad item pool that was then evaluated in a sample of community participants and gymnasium attendees. Internal reliability was adequate and convergent validity was assessed with the Drive for Muscularity Scale. These findings provide preliminary evidence that the EMS is a reliable and valid measure of dysfunctional eating behaviours towards muscularity as well as leanness. Additional research is needed to evaluate the clinical utility and discriminant validity of the EMS.

Topic: Treatment of Eating Disorders (Child and Adolescent)
Meeting Hall IV, second floor
Kelly Bhatnager, PhD & Dasha Nicolls, MBBS, MD, FAED

19.1: Factors Associated with the Strength of Early Therapeutic Alliance in a Group of Teenagers Treated for ED

Dominique Meilleur, PhD, Université de Montréal, Montréal, Québec, Guillaume Morin, BSc, Université de Montréal, Montréal, Québec, Olivier DiPietrantonio, BSc, Université de Montréal, Montréal, Québec, Marilyne Laverdière, BSc, Université de Montréal, Montréal, Québec, Geneviève Porlier, BSc, Université de Montréal, Montréal, Québec, Danielle Taddeo, MD, Centre Hospitalier Universitaire Sainte-Justine mère-enfant, Montréal, Québec, Jean-Yves Frappier, MSc MD; Centre Hospitalier Universitaire Sainte-Justine mère-enfant, Montréal, Québec

Studies had suggested that therapeutic alliance is a promising factor associated with the outcome of the intervention. The aim of this study was to examine the relationship between individual variables in a group of teenagers treated for ED and the strength of the therapeutic alliance at the beginning of an inpatient treatment. The sample was composed of 94 teenagers with ED (ANR=82%, ANB=7%, B= 6% and EDNOS = 4%) aged between 11 and 18 y.o. and hospitalized on an inpatient unit ED program in a University Children’s Health Centre in Québec, Canada. At the beginning of the inpatient treatment (T1), participants completed self-report questionnaires assessing Motivation to change (Anorexia Nervosa Stage of Change Questionnaire), Self-efficacy (Self- Efficacy Questionnaire, SEQ), Eating disorder risk scales and Psychological scales (Eating Disorder Inventory-3). The participants also answered a questionnaire on the strength of the therapeutic alliance with their assigned therapist. Therapeutic alliance was assessed with the short version of the Working Alliance Inventory-S (WAI-S) and was completed after at least 3 meetings with their therapist. Results showed that Motivation to change (r= .332, p < .001) and Self- efficacy (total score) were positively correlated (r= .348, p < .001) with the strength of the therapeutic alliance at T1. Two subscales of the SEQ were strongly correlated with the TA : the Social Self-efficacy (r= .372, p < .001) and the Emotional Self-efficacy (r=.298, p < .01) scales. The scores on the scales of Drive for Thinness (r= - .215, p < .05) and Body Dissatisfaction (r= - .207, p < .05) of the EDI3 were negatively correlated with the therapeutic alliance. The scores on Personal Alienation (r= - .246, p < .05), Emotional Disregulation (r= - .238, p < .05) and Asceticism (r= - .237, p < .05), some of the psychological scales of the EDI3, were negatively correlated with the strength of the TA. Analysis of regression showed that scores on Social Self-efficacy and Emotional Self-efficacy (subscales of the Self-efficacy Questionnaire) measured at T1 are the best predictors (model of two factors) of the strength of TA at T1, accounting for 20.6% of the variance. These results suggest that self-efficacy is an important factor to consider in the establishment of the therapeutic alliance with teenagers treated for ED.

19.2: How Much is Enough? Dose, Phase Completion, and Outcome in Family Based treatment for Adolescent Anorexia Nervosa

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

This aim of this research was to investigate Family Based Treatment (FBT) dose and phase completion on outcomes at 12-month follow up. Participants were 69 medically unstable adolescents with DSMIV diagnosed Anorexia Nervosa with an EBW 78.27 (6.39) percent and Global EDE 3.03 (1.13) at admission. These participants were a subset of a randomised control trial that investigated length of hospital admission prior to a 20 sessions FBT protocol. These participants completed the 20 session protocol but if not remitted at session 20 could opt to continue FBT sessions or move to other treatment options. Assessments occurred at admission, session 20 and 12 months post the 20th session. The analysis first compared those who met a strict remission criteria (Weight >95% EBW and EDE 1 SD of community norms) at session 20 with those who continued with FBT and those who choose to change to another treatment. The second analysis compared those who completed phase 3 of treatment with those who ceased treatment at phase 2. Fifty three participants did not meet remission criteria at session 20 with 39 continuing with FBT sessions. At 12 month follow up these participants had higher weights, and twice as many met remission criteria than those who opted for a change in treatment at session 20. Additionally those that continued in FBT had significantly less hospital readmissions and less admission days at 12 months follow up.Thirty nine particpants commenced phase 3 and at 12 month follow up had a significantly higher remission rate (49%) than those who ceased sessions in Phase 2 (23%). Those that finished in phase 3 had significantly lower global EDE and hospital readmission days than those who completed 20 or more sessions but clinically were in phase 2. These results indicate that continuing beyond the recommended 20 sessions can increase remission rate for some adolescents. It also suggests completing all treatment phases leads to improved outcomes, and less psychosocial disruption through reduced hospitalisation. These results have implications for routine clinical practice where patients often present with multiple problems and need to be cared for regardless of how quickly they initially respond to treatment. Results also point to the importance of addressing adolescent issues after the initial focus on weight recovery.

19.3: Home-Based Interventions in Acute Restrictive Eating Disorders: A Mixed Methods Study

David Clinton, PhD, Karolinska Institutet, Stockholm, Stockholm, Carl Estenfeld, BSc, Stockholms centrum för ätstörningar (SCÄ), Stockholm

Restrictive eating disorders are severe, life-threatening conditions that can affect patients at a young and formative age. In order to avoid negative course of illness it is imperative that these disorders are diagnosed and treated early. To meet these challenges The Stockholm Centre for Eating Disorders (SCÄ) has developed a mobile family unit (MOF) that uses intensive short-term interventions in the home environment to normalise eating, support parenting skills and improve self-image. The present study examined the effectiveness of these home-based interventions. A mixed-methods design using quantitative and qualitative components was used to compare MOF patients aged 10-14 years (N=40) with a control group matched on age and diagnosis receiving specialist ED treatment without home-based components (N=40). Qualitative interviews were conducted of mothers (N=7), fathers (N=6) and patients (N=6) treated at the MOF unit. One year after initial assessment MOF patients and controls showed significant increases in BMI, as well as significant reductions in self-rated ED psychopathology and negative aspects of self-image. No significant differences were found between groups. Experiences of home-based interventions were characterised by four major themes (and component themes): Importance of being at home (Emotional comfort, Practicality, Ease of disclosure); Getting to the heart of the matter (Quick interventions, Confrontation, Regaining the initiative, Building relationships); Feeling lost (Humbled by the situation, Communication and knowledge, Being taken seriously); and Being part of a whole (Family unit, Wholeness, Time with therapists). Home-based interventions present unique opportunities to strengthen the working alliance, confront shame, and engage families in effective treatment.

19.4: Implementation and Efficacy of Multi-Family Therapy for Anorexia Nervosa at the Eating Disorder Service, The Children’s Hospital, Westmead, Australia

Elaine Tay, DPsych, Eating Disorders Service, The Children's Hospital Westmead, Sydney, NSW, Andrew Wallis, LCSW-C, Eating Disorder Service, The Children's Hospital Westmead, Sydney, NSW, Julian Baudinet, DPsych, Eating Disorder Service, The Children's Hospital Westmead, Sydney, NSW, Lisa Dawson, DPsych; PhD, Eating Disorder Service, The Children's Hospital Westmead, Sydney, NSW

The Eating Disorder Service at The Children’s Hospital Westmead is the first service in Australia to integrate Multi-Family Therapy (MFT) as a novel enhancement to standard Family Based Treatment (FBT) for AN. The aim of this study was to pilot the implementation of MFT, and assess efficacy of MFT augmenting standard outpatient treatment for AN. Participants were 57 female and 3 male adolescents (age range 10.73 - 17.2) with DSMV diagnosis of AN and their families. This is a case series of the first 2 years of implementation. Patients were selected for MFT if progress was poor in outpatient therapy, mostly determined by poor weight gain, residual issues with food variety, high levels of distress or other comorbid issues impeding progress. The program ran 6 times per year with up to eight families per group. Each group of families worked together in an initial 4-day workshop, followed by 6 one-day workshops over a nine-month period. Measures of outcome include weight change, measures of parental confidence and efficacy, solidarity and hope, as well as retention and satisfaction measures. %EBW increased over the initial 4-day workshop, alongside a statistically significant increase in %EBW from one month prior (M=91.94) to 4-6 months post (M=96.8) initial 4-day workshop. Families reported benefiting from the solidarity and support of the group (up to 94%), noting this as integral to improvements in efficacy and confidence in making change, and building hope. Measures also indicated high levels of satisfaction (up to 92%). Evidence suggests MFT targets families that do not progress in standard FBT for AN by increasing treatment intensity, building solidarity and increasing parental efficacy. The group context of MFT and the opportunity to receive therapeutic input from multiple sources appears to increase a sense of agency, mobilizing family resources to continue to progress in outpatient care.

19.5: Outcomes of Aggressive Oral Refeeding for Adolescents with Anorexia Nervosa

Elizabeth de Klerk, BSc; MD, BC Children's Hospital, Vancouver, British Columbia, Ana Sofia Lopez, BSc; MSc; MD, BC Children's Hospital, Vancouver, British Columbia, Peiyoong Lam, MD; FRCPC, FRACP, MBBS, BC Children's Hospital, Vancouver, British Columbia

There is large variability in refeeding practices of adolescents with Anorexia Nervosa (AN) requiring hospitalization. This study aims to demonstrate that adolescents with AN have a low incidence of medical complications even when started on oral meal plans of over 1300 calories and reviews the impact on vital signs, weight gain and mid-upper arm circumference (MUAC) change. REB approval was obtained for a retrospective chart review of female patients 10-18 years of age with AN admitted to BC Children’s Hospital for medical stabilization and nutritional rehabilitation from December 1st 2014 to December 1st 2015. Data collected included demographics, vital signs, weight, MUAC, ECG changes, caloric intake/day and length of stay. Electrolytes (including phosphate, magnesium, ionized calcium, glucose, venous pH) and liver function tests were also assessed. To date, 21 out of 42 charts have been reviewed. Mean age was 13.9 years (SD+/-1.4). Initial caloric intake ranged from 1300-4050 kcal/day with median starting caloric intake of 2000. Average weight gain was 0.73 kg/week (SEM +/- 0.06). Average gain in MUAC was 0.28 cm/week (SEM +/- 0.01). As per protocol, all patients were initiated on oral phosphate supplementation at 500mg BID. No patients developed refeeding syndrome. One patient demonstrated hypophosphatemia (defined by <0.80mmol/L) and one had hypoglycemia (defined by glucose <3.0mmol/L). Of note, 95% were acidotic (venous pH <7.35) on admission, with 44% recovering by discharge. Additionally, 67% had elevated liver enzymes (AST/ALT >30U/L). 76% had orthostatic changes in HR (defined by >30bpm change), with 62.5% recovering by discharge. No patients had prolonged QT interval. The low incidence of complications with higher calorie meal plans provides additional support to recent literature advocating aggressive refeeding in AN patients. The presence of acidosis and transaminitis on admission deserves further research.

19.6: Cognitive Remediation Therapy for Children and Adolescents with Complex and Severe Eating Disorders: Outcomes from Individual and Group Formats

Amy Harrison, MPsy; DClinPsy; PhD, University College London Ellern Mede Service for Eating Disorders, London, UK, Pamela Stavri, MSc; BSc, Ellern Mede Service for Eating Disorders, London, UK, Lynn Ormond, BSc; MSc, Ellern Mede Service for Eating Disorders, London, UK, Francine McEneny, BSc, Ellern Mede Service for Eating Disorders, London , UK, Dilan Akyol, BSc; MSc, Ellern Mede Service for Eating Disorders, London, UK, Annum Quershi, BSc; MSc, Ellern Mede Service for Eating Disorders, London, UK, Hind Al-Khairulla MD; MRCPsych, Ellern Mede Service for Eating Disorders, London, UK

Cognitive remediation therapy (CRT) is a low intensity treatment adjunct adapted for individuals with severe and complex eating disorders (EDs) who have difficulties with globally-oriented and flexible thinking. CRT focuses on the process rather than the content of thinking and aims to support recovery through the development of cognitive strategies. The evidence base for adults with EDs is supported by four randomised controlled trails (RCTs), but needs strengthening before conducting an RCT involving young people. In this study, an uncontrolled trial was conducted to explore whether CRT delivered in individual and group formats would be a feasible, acceptable and beneficial treatment for n=125 adolescent inpatients with severe and complex EDs. Seventy patients (mean age=15.22, SD=1.44) received 10 sessions of individual CRT and 55 patients (mean age=14.89, SD=1.74) received 10 sessions of group CRT. In individual CRT, n=1 patient (1.43%) dropped out and intention-to-treat analyses revealed medium-sized improvements in central coherence (Rey-Osterrieth Complex Figure Task (RCFT); d=0.5), small-sized improvements in set-shifting (Trail-Making Test; d=0.41), and small to large-sized improvements in switching-related initiation and inhibition skills (Colour-Word Interference Test; d=0.41 and Hayling Test; d=0.72). There was a large-sized improvement in motivation to recover (Motivational Stages of Change for Adolescents Recovering from an ED d=0.86). Group format CRT had a higher drop-out (9.09%; n=5) and intention-to-treat analyses indicated that after group CRT, patients showed small-sized improvements in global information processing (RCFT; d=0.25) and a medium-sized improvement in self-reported flexibility (Cognitive Flexibility Scale; d=0.72). Patients found the group acceptable, with a mean satisfaction rating of 4.15/5 (0.62). The materials were easily adapted for the group setting and these data suggest an RCT investigating CRT for young people with EDs is warranted.

Topic: Emotions and Emotion Regulation
Meeting Hall V, second floor
Jason Lavender, PhD & Tiffany Brown, PhD

20.1: Do People With Eating Disorders Have Difficulties Recognizing Emotions In Others?

Marcela Marin Dapelo, PhD, King's College London, London, UK, Kate Tchanturia, PhD; DClinPsy; FAED, King's College London, UK

People with Eating Disorders (ED) often exhibit difficulties in the socio-emotional domain. The ability to recognize emotions in others has been explored in people with ED, but findings have been mixed. Moreover, most studies have used prototypical displays of emotions, which are less comparable to more ambiguous real-life facial expressions. This study aimed to investigate the ability to recognize anger, disgust, fear, happiness, and sadness in ambiguous facial expressions in individuals with AN and BN. 103 women aged 18 to 55 years participated in the study, 35 had a diagnosis of Anorexia Nervosa (AN), 26 had Bulimia Nervosa (BN), and 42 were free from ED symptoms and served as healthy controls (HC). Participants completed an emotion recognition task in which they viewed pictures of faces portraying blended emotions and were requested to select the emotion that best described the facial expression shown in a forced-choice paradigm. Pictures were taken from the Facial Expression of Emotions: Stimuli and Test set of morphed facial expressions. Results indicated that participants with AN and with BN exhibited poorer recognition of disgust, compared to HC (Median (Mdn)AN=79.17; MdnBN=89.58; MdnHC=95.83; pANvs.HC<0.01; pBNvsHC<0.01), and often misinterpreted it as anger. Even though participants with AN showed the poorest performance, differences with BN were not significant (pANvsBN=0.24). In addition, both AN and BN participants showed a higher preference to interpret non-angry faces as anger, compared to HC (MdnAN=3.13; MdnBN=2.08; MdnHC=1.04; pANvs.HC<0.01; pBNvsHC<0.01; pANvsBN=0.30). Participants with ED did not differ from HC in the recognition of other emotions. The findings suggest that people with ED do not exhibit major difficulties in emotion recognition, but problems may be specific to the ability to identify and interpret disgust and anger. The study findings may relate to findings on attention and interpretation bias towards anger in people with ED.

20.2: Unique Associations of Affect and Maladaptive Perfectionism with Eating Disorder Symptoms in Women with Bulimia Nervosa

Jason Lavender, PhD, Neuropsychiatric Research Institute, Fargo, North Dakota, Kathryn Smith, PhD, Neuropsychiatric Research Institute, Fargo, North Dakota, Tyler Mason, PhD, Neuropsychiatric Research Institute, Fargo, North Dakota, Stephen Wonderlich, PhD, Neuropsychiatric Research Institute, Fargo , North Dakota, Ross Crosby, PhD, Neuropsychiatric Research Institute, Fargo, North Dakota, James Mitchell, MD, Neuropsychiatric Research Institute, Fargo, North Dakota, Anna Bardone-Cone PhD, University of North Carolina, Chapel Hill, Chapel Hill, North Carolina, Thomas Joiner, PhD, Florida State University, Tallahassee, Florida, Marjorie Klein, PhD, University of Wisconsin-Madison, Madison, Wisconsin, Daniel Le Grange, PhD, University of California San Francisco School of Medicine, San Francisco, California, Scott Crow, MD, University of Minnesota, Minneapolis, Minnesota , Carol Peterson, PhD, University of Minnesota, Minneapolis, Minnesota

This study examined theoretically relevant affect (variability and intensity) and personality constructs in relation to a range of eating disorder (ED) symptoms in bulimia nervosa (BN). Adult women with full or subthreshold DSM-IV BN (N=198) completed affect, personality, and ED measures including the Eating Disorder Examination-Questionnaire (EDE-Q), Frost Multidimensional Perfectionism Scale (FMPS), Dimensional Assessment of Personality Pathology (DAPP), and Inventory of Depressive Symptomatology (IDS). A series of multiple regression analyses were conducted with various ED symptoms as dependent variables (i.e., dietary restraint, eating concern, shape concern, weight concern). Covariates included body mass index and age. Predictor variables were maladaptive perfectionism (calculated as the mean of the z-scored FMPS concern over mistakes and doubts about action subscales), depressive symptoms (IDS total score), and affect lability (DAPP affective lability score). Age (β=-.15, p<.05) and depressive symptoms (β=.25, p<.05) were uniquely associated with dietary restraint (R-squared=.14), whereas depressive symptoms (β=.26, p<.01) and maladaptive perfectionism (β=.38, p<.001) were uniquely associated with eating concern (R-squared=.30). For shape concern (R-squared=.31), body mass index (β=.21, p<.001) and maladaptive perfectionism (β=.37, p<.001) were uniquely associated, whereas body mass index (β=.20, p<.01), depressive symptoms (β=.23, p<.05), and maladaptive perfectionism (β=.30, p<.001) were uniquely associated with weight concern (R-squared=.31). Of note, affect lability was not uniquely associated with any ED symptoms in this sample. This pattern of findings suggests potentially distinct roles for maladaptive perfectionism and affect variables in relation to different ED symptoms in BN. These results may have clinical implications in terms of providing guidance for possible treatment targets to address the various ED symptoms experienced by women with BN.

20.3: Deliberate Expression of Emotions: A Study In Individuals With Anorexia Nervosa And Those Who Have Recovered

Marcela Marin Dapelo, PhD, King's College London, London, UK, Sergio Bodas, CPsychol, King's College London, London, UK, Kate Tchanturia, PhD; DClinPsy; FAED, King's College London, London, UK

The literature in emotion expressivity indicates that people with Anorexia Nervosa (AN) have difficulties expressing emotions nonverbally. Past studies have looked at spontaneous expressions of emotion, but the deliberate attempt to convey emotions has remained unexplored. Moreover, it has been proposed that the ability to imitate facial expressions contributes to developing control over facial displays, and as such, might relevant for facial emotion expressivity. This study aimed to assess the ability to deliberately pose and imitate facial expressions of emotions in women with acute AN and those who have recovered from the disorder. Participants were 36 women with AN, 16 who had recovered from AN (REC), and 42 who had no history of eating disorders and served as healthy controls (HC). Participants were instructed to pose and to imitate facial expressions of anger, disgust, fear, happiness, and sadness. Facial expressions for the imitation task were taken from the set of Pictures of Facial Affect. The participants’ facial expression was recorded and a blind rater evaluated their accuracy. Results indicated that the AN group had the poorest performance both posing and imitating facial expressions of emotions (Posed emotions: Mean (M)AN=3.34; MHC=4.12; pANvsHC<0.01; d=0.78; Imitation MAN=3.74; MHC=4.31; pANvsHC<0.01; d=0.87). The REC group showed an intermediate profile with scores that were closer to the HC group (Posed emotions: MREC=3.79; pRECvsHC=0.55; d=0.34; Imitation: MREC= 4.23; pRECvsHC=0.97; d=0.11), but differences between REC and AN did not reach statistical significance (Posed emotions: pRECvsAN=0.33; d= 0.52; Imitation: pRECvsAN=0.06; d=0.72). The study findings are consistent with the literature on spontaneous facial expression of emotions, showing that people with AN have difficulties when attempting deliberately to express emotions through the face, and that REC show an intermediate profile that tends to be more similar to healthy individuals.

20.4: Alexithymia Predicts Greater Improvement in Emotion Regulation After Dialectical Behavior Therapy

Tiffany Brown, PhD, University of California, San Diego, San Diego, California, Jade Avery, BA, Dartmouth College, Hanover, New Hampshire, Michelle Jones, PhD, University of California, San Diego, San Diego, California, Anne Cusack, PsyD, University of California, San Diego, San Diego , California, Julie Trim, PhD, University of California, San Diego, San Diego, California, Leslie Anderson, PhD, University of California, San Diego, San Diego, California, Christina Wierenga PhD, University of California, San Diego, San Diego, California, Walter Kaye, MD, University of California, San Diego, San Diego, California

Previous research has demonstrated that eating disorder patients with high alexithymia have less favorable treatment outcomes, which has been attributed to difficulties with emotion regulation. While dialectical behavior therapy (DBT) is an effective treatment for improving emotion regulation, no studies have examined how alexithymia levels may impact patients’ ability to improve emotion regulation over treatment. Thus, the purpose of the present study was to examine alexithymia as a predictor of change in emotional regulation strategies in adults with eating disorders following DBT. Participants were 173 adults (M[SD] age = 26[9] years) who completed assessments at intake and discharge from treatment at the UCSD Eating Disorders Partial Hospitalization Program. Contrary to expectations, results demonstrated that alexithymic patients demonstrated greater improvement in total Difficulties in Emotion Regulation Scale scores (DERS; p-values < .001), DERS emotional clarity (p-values < .002), and DERS emotional awareness scores (p-values < .03) compared to patients with possible alexithymia and non-alexithymia, who did not differ from one another (all p-values >.82). Individuals with alexithymia demonstrated improvements of medium to large effect sizes across measures (d range = 0.67-1.14), while individuals with possible and non-alexithymia demonstrated small effects (d range = 0.22-0.34). Results suggest that, contrary to previous research suggesting that alexithymia may be a negative prognostic factor for treatment, alexithymic patients demonstrated greater improvements in emotional clarity, awareness, and overall emotion regulation after DBT, suggesting that these deficits are malleable. Results also provide further support that DBT results in improvements in emotion regulation and that DBT may be an effective form of therapy for eating disorder patients who are alexithymic and may have difficulty with other treatments.

20.5: Shame and Eating Behavior in Sample of Russian Women with Eating Disorders

Svetlana Ilina-Bronnikova, MSc; PhD, Center for intuitive eating "IntuEat", Moscow, Vladislav Bukhtoyarov, MD; PhD; PhD ABD, Center for intuitive eating IntuEat, Saint Petersburg, Angelina Chekalina, PhD; MPsych, Center for intuitive eating IntuEat, Moscow, Ksenia Syrokvashina, MD; PhD, Center for intuitive eating IntuEat, Moscow , Zoya Zvyagintseva, MPsych, Center for intuitive eating IntuEat, Moscow

The implication of emotional experience on eating behaviour is widely discussed among professionals in the field. Shame is especially important due to cultural and micro social influences on the development of eating disorders and due to an invalidating effect of the environment. The aim of this study was to reveal an impact of shame on eating habits in a sample of 142 Russian female patients with eating disorders (binge eating disorder, bulimia nervosa, unspecified eating disorder). Methods used for assessment of the sample included NVM (Dutch personality inventory, a variation of mini-MMPI), DEBQ (Dutch Eating Behaviour Questionnaire), EAT-26 (The Eating Attitudes Test), IES-2 (The Intuitive Eating Scale-2). Statistical analysis (the Kruskal-Wallis criterion) shows that patients with high level of shame have high disposition to external (χ2 = 12,791; p = 0,012) and emotional (χ2 = 8,020; p = 0,091) eating. They also have highest scores of occupation with food (χ2 = 13,856; p = 0,008). Finally, they have lower ability to follow physical hunger and satiety cues and a higher tendency to eat due to emotional reasons (χ2 = 13,212; p = 0,010). Summarizing results of the statistical analysis we can say that shame is an important factor of eating behavior. Shame defines emotional aspects of disordered eating and also affects awareness of physical sensations. Shame is supposed to be one of the key elements of a vicious cycle of dieting. Restrictive eating behavior often leads to uncontrolled overeating and blaming thoughts like “I am lazy and weak, unable to control what I eat”. These thoughts and shame act as a trigger for secondary emotions like guilt, anger, anxiety. Lack of emotion regulation skills leads to the use of food as a way to cope with emotions. In conclusion, our data demonstrate that there are in our sample two ways how shame is connected with disordered eating behavior - it leads to emotional dysregulation and lowers ability to follow bodily signals.

20.6: Affective Instability in Bulimia Nervosa: Temporal Associations between Volatile Emotion and Dysregulated Eating Behavior

Laura A. Berner, PhD, University of California, San Diego, San Diego, California, Ross D. Crosby, PhD, Neuropsychiatric Research Institute; UND School of Medicine and Health Sciences, Fargo, North Dakota, Li Cao, MS, Neuropsychiatric Research Institute, Fargo, North Dakota, Scott G. Engel, PhD, Neuropsychiatric Research Institute; UND School of Medicine and Health Sciences, Fargo , North Dakota, Jason M. Lavender, PhD, Neuropsychiatric Research Institute; UND School of Medicine and Health Sciences, Fargo, North Dakota, James E. Mitchell, MD, Neuropsychiatric Research Institute; UND School of Medicine and Health Sciences, Fargo, North Dakota, Stephen A. Wonderlich PhD, Neuropsychiatric Research Institute; UND School of Medicine and Health Sciences, Fargo, North Dakota

Prior bulimia nervosa (BN) research suggests that increasing negative affect (NA) precedes binge eating and purging (B/P), and that NA decreases following B/P. Despite this evidence supporting a role for NA intensity in BN, fluctuations in affective state, or “affective instability,” may align more closely with the construct of “emotion dysregulation” thought to drive BN symptoms. No study to date has used ecological momentary assessment (EMA) to examine whether NA instability temporally relates to B/P, and therefore potentially precipitates or plays a role in reinforcing these behaviors. In the current study, women with BN (n = 133) logged multiple daily affect ratings and eating disorder behaviors over 2 weeks using portable digital devices. Two state-of-the-art indices quantified affective instability: Probability of Acute Change, which represents the likelihood of extreme increases in NA, and Mean Squared Successive Difference, which represents average NA variability. GEE models compared instability before and after B/P episodes and on B/P days versus non-B/P days. On B/P days, extreme NA increases were less likely after B/P episodes than before them. However, average NA instability was greater on B/P days than non-B/P days, greater after B/P episodes than during the same time period on non-B/P days, and greater after B/P episodes than before them (all ps <0.01). Results lend support to the notion that bulimic behaviors are negatively reinforcing (i.e., via post-behavior reduced likelihood of an acute increase in NA), but indicate that these behaviors are ineffective in promoting ongoing emotional stability. In fact, our findings suggest that B/P ultimately may worsen average NA volatility, potentially precipitating subsequent maladaptive behaviors. Interventions for BN that promote steady affective state maintenance and specifically focus on implementing skills during the post-B/P time period may be particularly helpful.

Topic: Relapse Prevention and Recovery
Terrace 1, second floor
Bruno Nazar, MD, PhD & Greta Noordenbos, PhD

21.1: Insights in Recovery: Harnessing Narratives of Lived Experience to Engage Patients in Recovery. Common Themes from four Qualitative Studies of the Experience of Recovery from Eating Disorders

Lesley Cook, MA, The Butterfly Foundation, Sydney, New South Wales

Qualitative inquiry is offering valuable insights into aspects of recovery from eating disorders. In 2016, the Insights in Recovery research project and three independent qualitative studies in Australia, explored the lived experience of recovery for people with eating disorders using phenomenological interpretive approaches. Participants experienced inconsistencies between their actual experience of recovery and the expectations of recovery they received from their treatment professionals. They understood recovery to be a more complex process than treatment of symptoms; one that moved them towards a satisfying life. Eating disorders occur in the context of a person’s life and their understanding of what is happening to them provides the framework for their experience of treatment and recovery. People want to receive treatment that is based on an integrated understanding of their lives. Person-centred care is the most effective way to treat someone with an eating disorder. Understanding recovery from the person’s perspective is fundamental to being able to develop treatment interventions that better support the transition to recovery. This session will present three key themes that emerged across all four studies: connectedness, including a connected life journey and social connection; dealing with feelings, including fear and shame and mental health issues such as depression; and control, including the commitment to recover, having choices and developing life skills. Participant’s experiences support arguments for integrated care that extends beyond the remission of symptoms to equip people for the longer journey of recovery. Implications for treatment approaches will be discussed. Together, these studies engaged 148 people with lived experience of eating disorders. Opportunities for international multi-agency research to further investigate the commonalities in recovering from an eating disorder will also be discussed.

21.2: Rate, Timing and Predictors of Relapse in Patients with Anorexia Nervosa following a Relapse Prevention Program: A Cohort Study.

Tamara Berends, MSc; RN, Altrecht Eating Disorders, Zeist, Utrecht, Berno van Meijel, PhD, InHolland University of Applied Sciences, Amsterdam, Zuid Holland, Willem Nugteren, MSc, Parnassia Psychiatric Institute, The Hague, Zuid Holland, Mathijs Deen, MSc, Parnassia Psychiatric Institute, The Hague , Zuid Holland, Unna Danner, PhD, Altrecht Eating Disorders, Zeist, Utrecht, Hans Hoek, MD, Parnassia Academy, The Hague, Zuid Holland, Annemarie van Elburg MD, Altrecht Eating Disorders, Zeist, Utrecht

Purpose of the study: Relapse is common among recovered anorexia nervosa (AN) patients. In leading guidelines there is general consensus that relapse prevention in patients treated for AN is a matter of essence. However, lack of methodological support hinders the practical implementation of relapse prevention strategies in clinical practice. For this reason we developed the Guideline Relapse Prevention Anorexia Nervosa. In this study we examine the rate, timing and predictors of relapse when using this guideline. Sample and methods: Cohort study with 83 AN patients who were enrolled in a relapse prevention program for anorexia nervosa with 18 months follow-up. Data were analyzed using Kaplan-Meijer survival analyses and Cox regression. Summary data and results: Eleven percent of the participants experienced a full relapse, 19% a partial relapse, 70% did not relapse. Survival analyses indicated that in the first four months of the program no full relapses occurred. The highest risk of full relapse was between months 4 and 16. None of the variables remained a significant predictor of relapse in the multivariate Cox regression analysis.The guideline offers structured procedures for relapse prevention. In this study the relapse rates were relatively low compared to relapse rates in previous studies. We recommend that all patients with AN set up a personalized relapse prevention plan at the end of their treatment and be monitored at least 18 months after discharge. It may significantly contribute to the reduction of relapse rates.

21.3: Definition of Recovery from Multiple Perspectives: Qualitative Study of Patients with Eating Disorders, their Parents, and Multi-Disciplinary Clinicians

Tracy Richmond, MD; MPH, Boston Children's Hospital, Boston, Massachusetts, Alice Woolverton, Student, Boston Children's Hospital, Boston, Massachusetts, Kathy Mammel, MD, University of Michigan Mott Children's Hospital, Ann Arbor, Michigan, Allegra Spalding, Student, Boston Children's Hospital, Boston , Massachusetts, Rollyn Ornstein, MD, Penn State University, Hershey, Pennsylvania, Amanda Bryson, BA, Penn State University, Hershey, Pennsylvania, Grace Kennedy BA, Florida State University, Talahassee, Florida, Ellen Rome, MD, Cleveland Clinic, Cleveland, Ohio, Elizabeth Woods, MD; MPH, Boston Children's Hospital, Boston, Massachusetts, Sara Forman, MD, Boston Children's Hospital, Boston, Massachusetts

In order to understand the definition of recovery from an eating disorder from the perspective of patients, their parents, as well as clinicians, we recruited a convenience sample of patients with eating disorders diverse in age, gender, and eating disorder diagnosis (n=24, age 12-23, anorexia nervosa n=16, bulimia nervosa n=5, binge-eating disorder n=3, avoidant/restrictive food intake disorder n=3), their parents (n=20), as well as dietitians (n=11), and mental health (n=14) and primary care providers (n=9) from three sites: Boston Children’s Hospital, University of Michigan C.S. Mott Children’s Hospital, and Penn State Hershey Children’s Hospital. In-depth, semi-structured, qualitative interviews with participants focused on several domains, including recovery (“Most people look at weight restoration but we think that recovery is much more complex than that. How would you define recovery?”). Interviews were performed by uniformly trained research assistants following a standardized interview guide; recordings were transcribed and analyzed using thematic analysis. Only clinicians reported resumption of menses as key to recovery (n=3); 4/6 respondents who identified non-weight biological signs (e.g., heart rate) as important were clinicians. The most commonly reported aspect of recovery was self-acceptance, reported by 8 each of patients (“just a healthy perspective on how we look and how you feel”), parents (“It’s the whole person. Not just the weight but how they perceive themselves and their self-esteem”), and providers. Other commonly reported definitions were living without an unhealthy focus on food and overall happiness/quality of life. Our study demonstrates the difference in perspectives between patients, parents and clinicians when defining eating disorder recovery. Clinicians should incorporate markers of recovery beyond weight restoration and resumption of menses, including quality of life or positive self-image, when treating patients with eating disorders.

21.4: Early Response in the Treatment of Eating Disorders: A Systematic Review and Diagnostic Test Accuracy Meta Analysis

Bruno Nazar, MSc; MD; PhD, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, Louise Gregor, BA; MSc, King's College London, London, UK, Gaia Albano, BA, King's College London, London, UK, Valentina Cardi, CPsychol; PhD, King's College London, London, UK, Janet Treasure, MD; PhD; FAED, King's College London, London, UK

The early response to Eating Disorders treatment is thought to predict a later favourable outcome. We have reviewed the literature and used a diagnostic test accuracy meta analysis to examine the robustness of this concept. We followed PRISMA guidelines and summarized the criteria used to define early and late response across studies. We used a diagnostic test accuracy meta-analysis to estimate the size of the effect. We synthesised results from 29 studies. Fifteen studies were used in the meta-analysis. In anorexia nervosa (AN), the diagnostic odds ratio (DOR) of early responders to predict remission was 4.85 (95%CI: 2.94-8.01) and the summary Area Under the Curve (AUC)=.77(SE=.03). For bulimia nervosa (BN), DOR was 2.75 (95% CI:1.24-6.09) and the AUC=.67(SE=.04), while for binge eating disorder (BED), DOR was 5.01 (95% CI: 3.38-7.42) and the AUC=.71 (SE=.03). Early behaviour change predict later symptom remission for AN and BED but there is less predictive accuracy for BN treatment as heterogeneity of studies is high and other accuracy measures such as senitivity and specificity are low and presented significant bias.

21.5: The “Face” and “Place” of Eating Disorder Recovery: A Critical Discourse Analysis of Eating Disorder Treatment Center Promotional Materials

Andrea LaMarre, MSc; Student, University of Guelph, Guelph, Ontario, Erin Harrop, MSW, University of Washington, Seattle, Washington

Residential eating disorder treatment centers have experienced rapid growth in the U.S. in the past decade with the advent of the Affordable Care Act and increases in behavioral health coverage. Often these centers have extensive marketing strategies, involving online and community promotional materials. Marketing targets clinicians, patients, families, and communities, exhibiting the center, their approach to treatment, and their conceptualization of recovery. In addition to providing valuable treatment services, these centers necessarily shape the discourse around who gets, who can benefit from treatment for, and who recovers from eating disorders. In this exploratory qualitative study, we examine promotional materials (i.e. brochures, websites, and promotional merchandise) of 32 leading eating disorder treatment centers in North America. The purpose of this study is: 1) to better understand the breadth of conceptualizations of eating disorders 2) to discover what archetypes of eating disorders are promoted within materials and 3) to examine how centers differentiate themselves and promote their specific approach to eating disorder treatment. We draw on critical feminist and critical race theories as a framework and employ critical discourse analysis of words and images. Our analysis suggests that there are several issues with the representation of eating disorders in promotional materials, including the ways in which eating disorders are framed as the purview of young, female, cisgender, white, thin, and middle/upper-class people. Exceptions exist, particularly amongst treatment centers that differentiate themselves as more diversity-welcoming places. Promotional materials also provide information about the consequences of eating disorders and their approaches to surmounting the challenges associated with eating disorders but remain tethered to dominant constructions of health that suggest that there is one “best” way to be a healthy citizen in North America. The implication of this narrow representation of eating disorders may be that those who do not fit the stereotype may not feel welcome in treatment settings and may struggle to find services that adequately meet their needs. Further, these narrow representations may reinforce damaging stereotypes present in society at large.

21.6: Identifying Fundamental Criteria for Eating Disorder Recovery: A Systematic Review and Qualitative Meta-Analysis

Jan Alexander de Vos, MSc, Human Concern foundation, center for eating disorders, Amsterdam, Noord Holland, Andrea LaMarre, MSc, Guelph University, Guelph, Overijssel, Mirjam Radstaak, PhD, Twente University, Enschede, Overijssel, Ernst Bohlmeijer, PhD, Twente University, Enschede , Overijssel, Gerben Westerhof, PhD, Twente University, Enschede, Overijssel,

Outcome studies for eating disorder recovery regularly measure only pathology change as an outcome. Researchers, patients and recovered individuals highlight the importance of using additional criteria for measuring eating disorder recovery. There is, however, no clear consensus on which criteria to use. The aim of this study was to find fundamental criteria for eating disorder recovery according to individuals who were considered recovered. A systematic review and a qualitative meta-analytic approach were used. Eighteen studies with recovered patients and meeting various quality criteria were included. Results of the included studies were analyzed using a meta-summery technique where the frequency of the found criteria was examined. Several dimensions of psychological well-being and self-adaptability were found to be fundamental criteria for eating disorder recovery, besides the absence of pathology. The most frequently mentioned criteria were: self-acceptance, positive relationships, personal growth, decrease in eating disorder behavior/cognitions, self-adaptability/resilience and autonomy. Recovered patients rate the presence of aspects of psychological of psychological well-being as important aspects of recovery in addition to the absence of pathology. Supplementary criteria are needed to understand and measure recovery. We recommend to include instruments measuring psychological well-being and self-adaptability in monitors and guidelines for the measurement of eating disorder recovery.

Topic: Treatment of Eating Disorders (Adult) III
Terrace Room 2, second floor
Anthea Fursland, PhD, FAED & Heather Thompson-Brenner, PhD, FAED

22.1: Emotional Expression Predicts Treatment Outcome in Focal Psychodynamic and Cognitive Behavioural Therapy for Anorexia Nervosa: Findings from the ANTOP study

Timo Brockmeyer, DiplPsych; PhD, University Hospital Heidelberg, Heidelberg, Baden-Württemberg, Hans-Christoph Friederich, MD, Heinrich Heine University, Düsseldorf, Baden-Württemberg, Wolfgang Herzog, MD, University Hospital Heidelberg, Heidelberg, Baden-Württemberg

This study adopted a computational psychotherapy research approach to examine the potential impact of in-session emotional expression by patients with anorexia nervosa (AN) on treatment outcome in outpatient psychotherapy. Verbal emotional expression was assessed by means of computerised quantitative text analysis. Verbatim transcripts of audio recorded therapy sessions from the initial, early, middle, and late phase of treatment from n = 89 AN outpatients were obtained from a large randomised controlled trial that investigated manualised Focal Psychodynamic Therapy and Cognitive Behavioural Therapy for AN. Greater verbal expression of negative but not positive emotions by patients in the middle phase of treatment predicted favourable treatment outcomes, i.e. a higher body mass index at the end of treatment and lower observer-rated eating disorder psychopathology at the end of treatment and at 12-months follow-up. The effects were independent of treatment condition, BMI at baseline, AN subtype, illness duration, and completer status, and specific to the middle phase of treatment. The main finding suggests that, irrespective of treatment condition, high levels of negative emotional expression in the middle phase of treatment constitutes an important active ingredient of successful psychotherapy for AN.

22.2: Acceptance in Eating Disorder Treatment at the Higher Levels of Care: A Naturalistic Effectiveness Study

Keegan Walden, BA; MS; PhD, Eating Recovery Center, Denver, Colorado, Jamie Manwaring, PhD, Eating Recovery Center, Denver, Colorado, Emmett Bishop, MD; FAED, Eating Recovery Center, Denver, Colorado, Alan Duffy, MS, Eating Recovery Center, Denver , Colorado, Gabriela Hurtado, PhD, Eating Recovery Center, Austin, Texas, Craig Johnson, PhD; FAED, Eating Recovery Center, Denver, Colorado

The treatment of severe and enduring eating disorders has received increased attention in the last decade, with many studies examining treatment outcomes and predictors of outcome. More recently, Dingemans and colleagues called for more naturalistic outcome studies with less stringent inclusionary and exclusionary criteria, and greater ecological validity and generalizeability. Accordingly, using a transdiagnostic sample of adult participants (N = 1,135) and no systematic exclusions, this study 1) evaluates the effectiveness of Acceptance and Commitment Therapy (ACT)-based eating disorder treatment at higher levels of care (i.e., partial hospital, residential, and inpatient) in terms of individual and group level psychological change, and 2) evaluates changes in acceptance, a central concept underlying ACT, as a predictor of outcome. Participants completed measures of eating disorder related psychopathology, depression, personality, and acceptance upon admission and discharge. Indices of clinically significant change and effect size were computed to evaluate individual and group level change, respectively. Regression models evaluated predictors of reduced eating disorder psychopathology in the context of personality variables. The majority of participants (58.2%) were treatment responders on at least one measure of eating disorder psychopathology, with participants with bulimia nervosa showing a higher treatment response rate than participants with anorexia nervosa (restricting type). In addition, treatment responders and non-responders did not differ in terms of treatment history, which suggests that prior treatment at a higher level of care facility does not reduce the likelihood of a successful outcome in the future. Effect sizes of all outcome measures ranged from d = .29 to d = 1.29. In addition, increased acceptance over the course of treatment both differentiated treatment responders and non-responders, and robustly predicted decreased eating disorder risk, regardless of diagnostic category.

22.3: Weight Change over the Course of Binge Eating Disorder Treatment: Relationship to Eating Behavior and Psychological Factors

Carly Pacanowski, PhD; RD, University of Delaware, Newark, Delaware, Tyler Mason, PhD, The Neuropsychiatric Research Institute, Fargo, North Dakota, Ross Crosby, PhD, The Neuropsychiatric Research Institute, Fargo, North Dakota, James Mitchell, MD, The Neuropsychiatric Research Institute, Fargo , North Dakota, Scott Crow, MD, The University of Minnesota, Minneapolis, Minnesota, Steve Wonderlich, PhD, The Neuropsychiatric Research Institute, Fargo, North Dakota, Carol Peterson PhD, The University of Minnesota, Minneapolis, Minnesota

Treatment for Binge Eating Disorder (BED) has not typically produced a significant change in weight despite reducing frequency of binge eating; both excess adiposity and psychological distress and impairment associated with binge eating increase risk for serious chronic health conditions. Individual variability in weight change may help to explain overall nonsignificant weight changes during treatment. Participants were 189 adults with DSM-IV BED who participated in a randomized clinical trial evaluating the efficacy of five months of cognitive-behavioral therapy for BED. Measures included anthropometric assessments of height and weight at baseline, mid-treatment, and end of treatment (EOT) and the Eating Disorder Examination Interview at baseline and EOT. Data were analyzed using multilevel models and bivariate correlations. During treatment, there was a mean weight gain of 1.9 ± 14.5 pounds. Twenty-five percent of the sample lost ³ 5 pounds and 25% of the sample gained ³ 8 pounds. Results showed that baseline objective binge episodes (OBE) moderated the trajectory of weight over the course of treatment. Individuals with greater baseline OBEs gained weight over the course of treatment, whereas those with lower OBEs maintained or lost weight. Higher baseline levels of objective overeating episodes (OOE) and restraint were significantly associated with baseline weight, although, they did not influence weight trajectory. Individuals engaging in more OOEs weighed more. Conversely, individuals higher in restraint weighed less. Changes in OBEs from baseline to EOT was associated with change in weight from baseline to EOT such that increases in OBEs were associated with more weight loss. Given this unexpected finding that greater reductions in OBEs were associated with more weight gain, further investigation of eating behavior during BED treatment to understand the energetic contributions to weight change or stability is warranted.

22.4: Impact of Residential Treatment on Eating Disorder Symptoms

Scott Crow, MD, University of Minnesota and The Emily Program, Minneapolis, Minnesota, Jillian Lampert, MPH; PhD, The Emily Program, St. Paul, Minnesota, Nicole Siegfried, PhD, Castlewood Treatment Centers, Birmingham, Alabama, Kourtney Gordon, RD, Fairwinds Treatment Center, Lutz , Florida, Aimee Arikian, PhD, The Emily Program, St. Paul, Minnesota, Craig Johnson, PhD, Eating Recovery Center, Denver, Colorado, Ann Erickson BA, Neuropsychiatric Research Institute, Fargo, North Dakota, Ross Crosby, PhD, Neuropsychiatric Institute, Fargo, North Dakota

The purpose of this study was to examine the impact of residential eating disorder treatment on eating disorder (ED) symptoms, mood, anxiety, and quality of life. Participants were 393 consecutive adults or adolescents with anorexia nervosa, bulimia nervosa, or another ED entering residential eating disorder treatment at 17 centers. Participants completed the EDE-Q, PHQ-9, Spielberger State/Trait Anxiety Inventory, and the Eating Disorders Quality of Life Scale at admission and discharge. Statistically significant improvements were seen in nearly all measures from admission to discharge, and effect sizes were large. Average weight restoration in the anorexia nervosa group was 1.53 lb. per week. In this large sample, clinically and statistically significant improvements were seen with treatment. These results support the acute benefit of residential ED treatment. Although the results are encouraging, more research on the long-term outcome is needed.

22.5: Evaluation of Hospitalisation for Anorexia Nervosa : the EVHAN Study

Malaïka Lasfar, MD, University hospital of Rouen, Rouen, France, Sylvie Berthoz, MD; PhD, CESP, Inserm, university Paris-Sud, UVSQ, University paris-Saclay, Villejuif, France, Christophe Lalanne, MD; PhD, CESP, Inserm, Univ. Paris-Sud, UVSQ, Université Paris-Saclay, Villejuif, France, Priscille Gérardin, MD; PhD, University Hospital of Rouen, Rouen , France, Nathalie Godart, MD; PhD, CESP, Inserm, Univ. Paris-Sud, UVSQ, Université Paris-Saclay, Villejuif, France, Evhan Group, MD, CESP, Inserm, Univ. Paris-Sud, UVSQ, Université Paris-Saclay, Villejuif, France

Anorexia Nervosa is complicated by psychological, bodily or social difficulties that can be life-threatening for the individuals concerned. In the most severe cases one or several long periods of hospitalisation may be required. International guidelines define inpatients treatment main aims. However treatment modalities are sometimes different from one centre to an other and not enough is known about the way in which the health of these individuals evolves during and after hospitalisation, nor about factors that might contribute to favourable or unfavourable outcomes. Likewise little is known about the reasons why almost half of the hospitalisations for Anorexia Nervosa are terminated prematurely (dropout) nor about the factors that could explain the varying length of hospitalisations. We will describe here a French multicentre research project the EVHAN study (Eudract number: 2007-A01110-53, Clinical trials). This study was conducted between March 2009 and December 2012. The aims were to assess 1-the different modalities of treatment in 11 French Eating Disorders impatient unit and 2-their impact on outcome at discharge and 12 months later 3- predictive factors of dropout and length of stay.The primary hypothesis is that the different treatment types impact outcome of patients at discharge from inpatient treatment and at one year follow-up, even after adjustment for confounding factors (age, length of illness, number of previous hospitalizations, clinical state at intake). We will summarize the first results of the study: description of the organisation of the care offered in the 11 centers and how there varies in terms of modalities; the impact of inpatient treatment (at discharge and one year later). We will also describe how this study contributed to open exchanges and partnerships among the eleven centres specialised in the care of eating disorders, and has enabled the establishment of a collaborative network between researchers and clinicians working on Anorexia Nervosa, both nationally and internationally.

22.6: Demystifying the Refeeding Process: Implementation of an Aggressive Weight Restoration Protocol

Heather Gallivan, PsyD, Park Nicollet Melrose Center, St. Louis Park, Minnesota, Deborah Mangham, MD, Park Nicollet Melrose Center, St. Louis Park, Minnesota, Alicia Phillips, RD, Park Nicollet Melrose Center, St. Louis Park, Minnesota, Dawn Taylor, PsyD, Park Nicollet Melrose Center, St. Louis Park, Minnesota

Concern about the refeeding syndrome has long dictated the rate of weight restoration in eating disorder treatment. Experience after World War II and in other situations of extreme malnutrition showed that rapid calorie repletion could result in serious fluid and electrolyte complications, including heart failure and death. As a result the eating disorder community has for decades adopted the mantra “Start low and go slow”. This practice however is not evidence based, and furthermore prolongs the duration of treatment, increases the expense of treatment and possibly delays the patients’ response to therapy. We embarked upon a quality improvement project to test the viability of a rapid refeeding protocol. Our study included all patients admitted to an inpatient eating disorders unit with a diagnosis of AN or OSFED and with a BMI between 14 and 18. There were a total of 51 patients. All patients were given a refeeding meal plan designed to attain a weight gain of 4 pounds per week. Patients were monitored daily for signs of refeeding complications. They received psychological support to manage anxiety about weight gain and to ensure program compliance. Four months of data suggest that the average weight restoration of these 51 patients was 4.0 lbs per week. Medical complications were minor and easily corrected. There were no episodes of hypophosphatemia. We also found that between admission and discharge, 31% of the patients had improvement by at least one score category in their PHQ-9 score, 41% had improvement in their EDE-Q score and 28% had improvement in their GAD-7 score. Our experience has shown that a more aggressive approach to refeeding is both safe and doable. Studies like ours will likely help change the landscape of eating disorder treatment, from the slow and costly to a more rapid, more effective and less expensive approach to refeeding.