International Journal of Eating Disorders最新文献

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: Describe the feasibility, acceptability and preliminary efficacy of family-based treatment for adolescents with avoidant/restrictive food intake disorders (FBT-ARFID) in a real-world clinical cohort. Explore differences between adolescents and younger children with ARFID using FBT-ARFID.

Method: A retrospective chart review from 2017 to 2023 identified 39 children (< age 13) and 40 adolescents (age 13-18) who were assessed and then treated with FBT-ARFID in an outpatient clinic at a large academic medical hospital. We pulled data from patient charts on demographics and clinical characteristics and completed chart reviews for clinical outcome data (acceptability, feasibility, hospitalization rates, concurrent treatments, and parental learning). We used independent sample t tests and χ ² testing, and reported effect size differences on comparisons.

Results: A total of 46/55 (84%) children began FBT-ARFID after an initial evaluation (feasibility), and of those, 39 (85%) received more than four sessions (acceptability). A total of 42/66 (64%) adolescents started FBT-ARFID; of those, 40 (95%) received more than four sessions of FBT-ARFID. Children and adolescents presented similarly prior to treatment and achieved comparable practical clinical outcomes when treated with FBT-ARFID. There were more neurodivergent adolescents receiving FBT-ARFID than children, defined as diagnoses of comorbid ASD, ADHD, or both (37% vs. 15%). Adolescents used a greater number of sessions (M = 21, SD = 11.9) compared to children receiving FBT-ARFID (M = 15.2, SD = 7.4).

Conclusions: Findings from this retrospective outpatient clinic chart review indicate that FBT-ARFID is feasible for treating, retaining, and obtaining clinical outcomes in adolescents.

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: 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: 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: Restrictive eating disorders (EDs), including anorexia nervosa (AN) and atypical AN (Atyp-AN), are often associated with cognitive rigidity that can impede treatment. The dorsolateral prefrontal cortex (dlPFC) plays a central role in cognitive control, but it remains unclear whether its activation during cognitive flexibility will differ across restrictive ED presentations.

Method: Eighty-seven females with restrictive EDs (aged 14-35) (AN: n = 31; atyp-AN with history of AN [hx-AN]: n = 33; atyp-AN without history of AN [Atyp-AN]: n = 23) completed a task-switching paradigm during functional magnetic resonance imaging. We examined dlPFC activation during sustained (task-switching vs. completing a single task) and transient control (switching between task rules vs. repeating the same rule), testing for group differences and symptom associations.

Results: The AN group showed a greater difference in left dlPFC activation during task switching vs. the single task compared to the hx-AN and Atyp-AN groups, driven by reduced activation during the single task condition. All groups showed similar increases in dlPFC activation during task switching and no differences in task performance. Across all participants, higher dlPFC activation during task switching vs. single task was associated with greater restraint symptoms.

Discussion: These novel findings identified shared versus unique neural features of sustained and transient control across restrictive ED groups. Heightened dlPFC activation during transient control associated with restraint may represent a transdiagnostic feature shared across restrictive EDs. Alternatively, reduced dlPFC activation during the low-demand, single-task condition in those with typical, but not atypical AN, may reflect a difference in sustained control, with implications for tailoring interventions to distinct restrictive ED presentations.

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: 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.

Objectives: Binge eating is the most common disordered eating behavior among pregnant women. This study examined the association of binge-eating frequency with the presence of a self-reported current preeclampsia diagnosis in a sample of U.S. military active-duty Service women.

Methods: Active-duty Service women (N = 134), 20-27 weeks gestation, completed an online survey assessing binge eating (number of episodes in the past 28 days), preeclampsia diagnosis, sociodemographics, pregnancy-related variables, and perceived stress. Firth logistic regression models evaluated the associations between binge-eating frequency and preeclampsia and between stress and preeclampsia, adjusting for relevant covariates and controlling for multiple comparisons.

Results: A total of 91 (67.9%) participants reported experiencing at least one objective binge-eating episode in the past 28 days, and 11 (8.2%) reported a current preeclampsia diagnosis. Evaluated in separate models, both binge-eating frequency and stress were significantly associated with preeclampsia, adjusting for gestational weight gain. Specifically, for each additional binge-eating episode a participant reported, their odds of having a preeclampsia diagnosis were 14% higher (p = 0.008), and for every one unit increase in stress, the odds of having a preeclampsia diagnosis were 38% higher (p = 0.007).

Discussion: Results indicated that a substantial majority of pregnant active-duty Service women in this sample reported binge eating. Binge eating and stress were both associated with increased odds of preeclampsia. While these relationships need further exploration, identifying high levels of stress or binge eating early in pregnancy could be an important addition to screening and prenatal care.

Objective: To examine baseline predictors of treatment completion and clinical outcomes in 16 to 25-year-olds referred for early intervention for a recent-onset eating disorder (ED).

Method: Participants (n = 228; 93% female, 63% White British) were drawn from the FREED-Up study, which evaluated First episode Rapid Early intervention for EDs (FREED) in England across 2017-2018. Measures were completed at baseline and 3-, 6- and 12-months. Generalized estimating equations were used to identify baseline predictors of treatment completion and symptom remission by 12-months. Linear mixed models were used to identify baseline predictors of change in global eating disorder examination-questionnaire (EDE-Q) scores, binge eating, purging and BMI (anorexia nervosa [AN] only) over 12-months. Possible predictors included baseline ED symptoms, depression, anxiety, stress, psychosocial impairment, expressed emotion from close others, functioning against personalized goals, age, ethnicity, duration of illness, waiting times and BMI.

Results: There were no significant predictors of treatment completion. Lower stress at baseline predicted increased likelihood of symptom remission by 12 months. Lower purging and psychosocial impairment at baseline predicted lower Global EDE-Q scores over the following 12 months; lower purging predicted lower rates of binge eating; and lower Global EDE-Q scores, binge eating and assessment waiting times predicted lower rates of purging. Higher purging at baseline was associated with higher BMI in AN.

Discussion: This study provides new data on predictors of treatment outcomes in an early intervention ED sample. In addition to ED symptoms and waiting times, psychosocial impairment and stress warrant consideration as factors that may influence treatment outcomes.