International Journal of Eating Disorders最新文献

Objective: Generative artificial intelligence (AI) has the potential to be used in supporting people with eating disorders (EDs), but this also presents certain risks. This study aimed at comparing the psycholinguistic attributes (language markers of cognitive, emotional, and social processes) and lexico-semantic characteristics (patterns of word choice and meaning in text), and assessing potential harms of AI responses versus human responses in online communities (OCs).

Method: We collected pre-COVID data from Reddit communities on EDs, consisting of 3634 posts and 22,359 responses. For each post, responses were generated using four widely used state-of-the-art AI models (GPT, Gemini, Llama, and Mistral) with prompts tailored to peer support. The Linguistic Inquiry and Word Count (LIWC) lexicon was used to examine psycholinguistic features across eight dimensions, and a suite of lexico-semantic comparisons was conducted across the dimensions of linguistic structure, style, and semantics. Additionally, 100 AI-generated responses were qualitatively analyzed by clinicians to identify potential harm.

Results: Using OC responses as a comparison, AI responses were generally longer, more polite, yet more repetitive and less creative than human responses. Empathy scores varied among models. Qualitative analysis revealed themes of possible reinforcement of ED behaviors, implicit biases (e.g., favoring weight loss), and an inability to acknowledge contextual nuances-such as insensitivity to emotional cues and overgeneralized health advice. All AI chatbots produced responses containing harmful content, such as promoting ED behaviors or biases, to varying degrees.

Discussion: Findings highlight differences between AI and OC responses, with potential risks of harm when using AI in ED peer support. Ethical considerations include the need for safeguards to prevent reinforcement of harmful behaviors and biases. This research underscores the importance of cautious AI integration; further validation, and the development of guidelines are needed to ensure safe and effective support.

Objective: Emotional inertia, the tendency for emotions to persist over time, has received little attention in relation to eating disorders (ED). However, emotional inertia may reflect poor emotion regulation and unresponsiveness to environmental cues, and individuals may use ED behaviors to distract from or escape persistent emotions. We aimed to characterize emotional inertia and its relationship with ED behavior frequency among adults with EDs.

Method: Adults (N = 94) with bulimia nervosa (BN) or binge eating disorder (BED) spectrum EDs completed 7-14 days of ecological momentary assessment, reporting negative affect (NA), positive affect (PA), and ED behaviors. Inertia was computed using within-person autoregressive estimates in a multilevel model. We compared NA and PA inertia between diagnostic groups, and as predictors of ED behavior frequency.

Results: Average NA and PA inertia did not differ by diagnostic group. Higher NA inertia was cross-sectionally associated with greater overall frequency of compensatory behaviors in participants with BN-spectrum EDs. NA inertia was not cross-sectionally associated with binge eating but was positively associated with overall frequency of dietary restriction. The cross-sectional association of NA inertia with binge eating and dietary restriction was not moderated by diagnostic group. PA inertia was not cross-sectionally associated with frequency of any ED behaviors.

Discussion: Emotional inertia may be important for understanding the development and maintenance of ED behaviors, particularly dietary restriction. Future research should explore temporal relationships. Interventions promoting flexible emotional responding may help to reduce emotional inertia and its effect on ED behaviors.

Objective: As generative AI becomes more widely adopted for mental health support and information, concerns exist about its potential to depict psychiatric illnesses in stereotypical ways. This Spotlight investigates how ChatGPT-5 visually represents eating disorders and whether prompt-based manipulations can generate more inclusive outputs, with the goal of laying a roadmap for future research to ensure these technologies are applied in safe, ethical, and clinically responsive ways.

Methods: ChatGPT-5 was prompted to generate images depicting anorexia nervosa, bulimia nervosa, binge-eating disorder, and body image difficulties. A standard one-shot prompt asked the model to create an artistic, realistic, fictional image of the specified condition. The model was then re-tasked to generate multiple images per condition using two approaches: (1) standard prompts and (2) prompts explicitly instructing variation in age, gender, ethnicity, and body type. Outputs were visually compared to evaluate whether bias-reducing instructions produced more inclusive and diverse depictions.

Results: Standard prompts produced images that reinforced conventional stereotypes, depicting eating disorders largely through young, White, female figures with either emaciated (for anorexia nervosa) or overweight (for binge-eating disorder and body image difficulties) body types. Minimal change was observed with single bias-reducing prompts. In contrast, generating multiple images with explicit diversity instructions produced noticeably more inclusive representations, featuring variation in age, gender and ethnicity, although some stereotypical features persisted.

Conclusion: ChatGPT-5's depictions of eating disorders rely on stereotypical templates by default. However, prompting the model to generate multiple images with explicit demographic diversity instructions improved representational inclusivity.

Objective: Regular eating has never been rigorously evaluated as a standalone intervention for restriction. In this preliminary investigation, we evaluate changes in restriction observed during a focused regular eating module (NOURISH: Nutrition, Organization, and Understanding Restriction: Intervention for Sustainable Habits) in the context of evidence-based, personalized, modular eating disorder treatment.

Method: We analyzed pooled data from two eating disorder treatment trials ranging from 13 to 20 weeks in duration. Using ecological momentary assessment data collected throughout the entire modular treatment, we evaluated within-group changes in restriction and restraint from the beginning to the end of the 3-4 session regular eating (i.e., NOURISH) module. Point estimates and 95% confidence intervals, as well as Cohen's d, were calculated for paired-sample t-tests.

Results: Our complete sample included seven participants. Because our sample size was small, our interpretation focuses on effect sizes rather than statistical significance. Results suggested non-significant, yet moderate-sized reductions in both restriction (d = 0.72) and restraint (d = 0.76) during the NOURISH module, with the 95% confidence intervals for these effects including zero.

Discussion: Although continued research in larger samples is warranted, our preliminary findings indicate that NOURISH holds promise as an evidence-based module to address restriction and restraint in the context of personalized eating disorder treatment.

Objective: Program-led and focused models may overcome structural barriers to accessing ED care, such as limited availability, for youth with EDs by prioritizing strategic, evidence-based care delivered through a structured approach. The Rapid Intervention to Support Eating Issues (RISE) pilot aimed to promote weight restoration and prevent hospitalization among malnourished adolescents at risk for hospitalization. We used a "home hospital" approach, integrating medical oversight, family-based treatment principles, and nutritional support via structured outpatient care.

Methods: Participants completed 4-5 visits with adolescent medicine and nutrition over 8 weeks. They received psychoeducation and support in implementing home hospital. Vital signs, anthropometrics, dietary intake, ED behaviors, and cognitions were assessed.

Results: A total of 27 patients participated. Patients experienced low hospitalization rates (n = 1 throughout; 3.7%) and significant weight gain (M_{end of treatment} = +2.7 kg from baseline, 95% CI: 2.6-4.7). There were statistically significant increases in calorie intake (M_baseline = 43.3% of recommendation; M_{end of treatment} = 76.0% of recommendation; dz = 0.98, 95% CI: 0.45-1.50) and decreases in the report of disordered weight control behaviors (n reporting at baseline = 10 [37%], n reporting at end of treatment = 3 [11.1%]; paired RD = -1.00, 95% CI: -1.00--0.33).

Discussion: This program-led and focused intervention produced meaningful outcomes and circumvented hospitalization for youth at high risk in a short time frame. This approach offers promise for scalable, early ED care that leverages programmatic expertise, consistent with evolving models of mental health service delivery.

Objective: To estimate the prevalence of probable eating disorders and self-reported eating disorder diagnoses among East Asian young adults aged 18-25 years across US colleges.

Method: Using data from the 2020 to 2023 Healthy Minds Study, a repeated cross-sectional survey of US college students, we analyzed data from East Asian and White participants aged 18-25 years (N = 160,740). Eating disorder risk was assessed using the SCOFF questionnaire and eating disorder diagnoses were self-reported. Using multivariable logistic models, we generated odds ratios (OR) and confidence intervals (CI) to estimate inequities in the prevalence of a probable eating disorder and eating disorder diagnoses between East Asian and White young adults, adjusting for gender identity, age, international student status, sexual orientation, financial stress, and study year.

Results: We found no statistically significant differences in the prevalence of a probable eating disorder among East Asian young adults compared to White young adults (OR: 1.04; 95% CI: 0.97-1.11) after adjustment. Among those with a probable eating disorder (n = 68,651), East Asian young adults had nearly half the odds (OR: 0.55; 95% CI: 0.47-0.65) of having a self-reported diagnosed eating disorder compared to White young adults.

Discussion: While the prevalence of having a probable eating disorder was similar among East Asian and White young adults, East Asians had almost half the odds of self-reporting an eating disorder diagnosis compared to White young adults. Future research is warranted to better understand barriers to eating disorder diagnosis among East Asian young adults in the US.

Objective: Negative body image is thought to play an important role in the onset and maintenance of anorexia nervosa and bulimia nervosa. The intervention Expand Your Horizon (EYH), which is focused on increasing functionality appreciation, is being investigated as a potential approach for improving body image. This study investigated the themes that were identified when adolescent girls with an eating disorder are asked to describe their body functionality within the context of EYH.

Method: Fifty-eight girls with an eating disorder received the EYH intervention and wrote about everything their body can do and why these functions are meaningful to them. The qualitative data were analyzed via Thematic Analysis.

Results: We identified five themes from the data: body functions as a means to experience and regulate emotions; body functions as a vehicle to form social connections; body functions as a means to experience independence and identity; how body functions are intertwined with the eating disorder; and the aesthetic body.

Discussion: This study highlights the importance of body functionality in the experiences and perceptions of adolescent girls with eating disorders. The findings support the idea that adolescents with eating disorders are able to appreciate their bodies for what they can do rather than focusing solely on their appearance.

Objective: This study aimed to explore potential disease-specific and weight-related neurostructural alterations in patients with acute anorexia nervosa (AN).

Method: Employing a novel BMI-matched design, structural MRI data were collected from 36 females with AN, 35 normal-weight healthy controls (NHC), and 29 underweight healthy controls (UHC). Cortical (thickness, surface area) and subcortical (volume) morphometry measures were computed via FreeSurfer. Group differences were tested using generalized linear models, with associations examined for BMI, symptom severity, and weight suppression (lifetime highest minus current weight).

Results: Compared with UHC, AN patients exhibited subcortical volume reductions in the bilateral pallidum and caudate, left putamen, and right thalamus, as well as cortical thinning in default mode network regions (bilateral inferior parietal lobule, right precuneus, posterior cingulate cortex) and the left cuneus, indicating potential disease-specific alterations. Comparisons between UHC and NHC revealed BMI-related alterations, reflected in surface area reductions of the right orbitofrontal cortex and left insula, and in volume reductions of the bilateral amygdala, right hippocampus, and left thalamus. Within AN, weight suppression was negatively associated with cortical thickness across 44 regions, suggesting a possible link with prior weight loss.

Conclusions: By including BMI-matched healthy control groups, this study provides preliminary evidence for distinguishing disease-specific from BMI-related neurostructural alterations in patients with AN. Future research may help clarify the role of weight suppression.

Objective: To synthesize the evidence on the associations between vegetarian and/or vegan diets (VVDs) and symptoms of orthorexia nervosa (ON) compared with omnivorous diets in the adult population.

Method: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and the Meta-analyses of Observational Studies in Epidemiology (MOOSE) guidelines, we conducted a systematic review and meta-analysis. We searched the MEDLINE/PubMed, Embase/Scopus, PsycINFO, and Web of Science databases up to June 17, 2025, with no language or date restrictions. Random effects models with the Sidik-Jonkman method were used to estimate pooled effect sizes.

Results: The meta-analysis included 26 cross-sectional studies with a total of 23,783 participants (72.0% female; mean age range: 19.6-51.0 years). Adults who followed VVDs had moderately higher ON symptoms compared to omnivores (standardized mean differences using Cohen's d index = 0.46; 95% confidence interval [CI]: 0.33, 0.60; inconsistency index [I²] = 81.0%). Additionally, categorical data revealed that VVD adherents were approximately twice as likely to report ON symptoms as omnivores (odds ratio = 1.99; 95% CI: 1.21-3.25; I² = 92.8%). Vegetarians and vegans were similarly associated with ON symptoms compared with omnivorous (p = 0.855).

Discussion: Adherence to VVD is associated with higher ON symptoms in young and middle-aged adults. However, these results should be interpreted with caution due to high heterogeneity and the low overall methodological quality of the exclusively cross-sectional studies included. Higher-quality longitudinal studies using validated assessment tools are needed to establish clearer causal relationships and inform clinical screening and intervention strategies.

Objective: Diet culture refers to the ubiquitous sociocultural system which conflates bodyweight with health, perpetuates myths about food and eating, and upholds a moral hierarchy of bodies derived from patriarchal, racist, and capitalist forms of domination. Diet culture promotes a hierarchy of bodies and weight-based moralism that can contribute to anti-fat bias and the development of eating disorders (EDs). Despite growing recognition of these harms, accessible interventions targeting diet culture beliefs remain limited. In this study, we evaluated a brief, single-session digital mental health intervention (DMHI) designed to challenge diet culture beliefs (e.g., fat = unhealthy).

Method: A sample of 455 cisgender women (M_age = 32.6) completed the DMHI, which included psychoeducation and cognitive restructuring techniques aimed at disrupting weight stigma and misinformation about weight and health. Of the 455 participants enrolled, between 284 and 296 completed post-intervention measures depending on the outcome. Attrition analyses indicated no significant baseline differences between completers and non-completers on ED pathology or weight stigma variables; the only exception was the interpersonal relationships subscale of fat acceptance, where non-completers reported slightly lower baseline scores. Pre- and post-intervention assessments measured ED pathology, fat acceptance, and anticipated weight stigma.

Results: We found increased fat-accepting health beliefs and decreased fear of weight stigma and ED symptoms, specifically fear of weight gain, following the intervention. However, interpersonal respect for fat individuals slightly decreased, and fat activism showed limited improvement, suggesting that certain components of fat acceptance may require more sustained or relational formats.

Discussion: These findings highlight the potential of DMHIs to shift belief systems related to diet culture and ED risk, particularly as scalable, low-barrier tools for prevention and early intervention. Future work should examine the durability of effects and test enhancements targeting activism and interpersonal change. Hypotheses, variables, and analyses were pre-registered at https://osf.io/27ym3.