OBJECTIVES:This World Federation of Societies of Biological Psychiatry (WFSBP) consensus paper aims to summarise and evaluate the published study results on objectively measurable biological markers associated with anorexia nervosa (AN). METHODS:The relevant literature was reviewed by the WFSBP Task Forces on Eating Disorders and on Biological Markers, and a consensus regarding the significance of the published evidence was reached. RESULTS:Candidate biological markers that have been associated with AN include clinical (e.g. body weight), molecular (e.g. genetic, epigenetic, hormonal, immunological, metabolomic), cellular (e.g. leukocytes), neuroimaging (e.g. structure, function, connectivity), digital, cardiac and neurophysiological parameters. Some clinical and laboratory parameters are risk markers in clinical practice. Biological markers have pathophysiological relevance in understanding the biological and metabolic pathophysiology of AN and its physical health consequences. Few studies have examined pharmacogenetics or therapeutic drug monitoring as tools to monitor and guide the treatment of AN. CONCLUSIONS:Biological markers will hopefully soon enable clinicians to intervene earlier in a more targeted manner to mitigate treatment resistance. However, the current scientific basis for most biological markers are group comparisons only. Studies on sensitivity, specificity and the prognostic value of these markers are lacking.

Compulsive exercise is commonly observed in adolescent anorexia nervosa (AN) and relates to more severe eating disorder pathology and worse outcomes. Within higher levels of care (HLOC) for eating disorders, physical activity is often restricted in an attempt to reduce compulsive exercise; yet, the effectiveness of these restrictions across levels of care remains understudied. The current study examined how compulsive exercise presents and changes within HLOC as individuals with AN transition from settings with greater supervision to those with more autonomy over movement. Given the central role of weight restoration in HLOC, the association between change in percent of expected body weight (%EBW) and compulsive exercise frequency at discharge was also explored. Participants were 443 adolescents with AN admitted to inpatient or residential treatment and stepped down to partial hospitalization or intensive outpatient programs. At admission, stepdown, and discharge, participants reported compulsive exercise frequency using the Eating Disorder Examination-Questionnaire. Among participants who endorsed compulsive exercise at admission (58.0%, n = 257), piecewise growth curve analyses found a significant decrease in compulsive exercise frequency from admission to stepdown (β = -0.21, p < .001) and no significant change from stepdown to discharge (β = -0.02, p = .080). Change in %EBW was not associated with compulsive exercise frequency at discharge (β = 0.17, p = .941, IRR = 1.19). Eating disorder recovery in the context of HLOC settings that initially restrict access to exercise is associated with reported decreases in compulsive exercise for many adolescents with AN, irrespective of weight stabilization. Research involving measures of exercise motivation, controlled designs, and follow-up data is needed to inform the development of personalized compulsive exercise interventions.

Marginalized populations experience increased eating disorder (ED) risk and encounter significant barriers to treatment. Intersectionality provides a framework for understanding how systemic oppression contributes to inequities in EDs; however, intersectional approaches have yet to be applied to a clinical ED sample. The current study examined inequities in ED severity and treatment outcome across the intersections of race/ethnicity, sexual orientation, and socioeconomic status (SES). Adult women (N=3016; M = 27.2 years) with transdiagnostic EDs presenting to affiliated treatment sites across the United States completed the Eating Disorder Examination-Questionnaire (EDE-Q) at admission and discharge. Race/ethnicity and sexual orientation were self-reported; SES was measured using the area deprivation index of participants’ neighborhoods. Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) was used to estimate baseline EDE-Q global score; change in EDE-Q global score and binge eating, self-induced vomiting, laxative use, and driven exercise frequency from admission to discharge; and reason for discharge (routine or non-routine) across intersectional subgroups. In this sample of women with access to treatment, MAIHDA models predicted higher baseline levels of overall ED pathology among sexual minorities (predicted M = 4.10). Few differences in ED symptom improvement were observed across intersectional subgroups, with some small yet potentially meaningful inequities. Racially/ethnically minoritized subgroups appeared slightly less likely to complete treatment (predicted percent non-routine discharge = 41.50%). Future research should build on these findings by analyzing other dimensions of inequity (e.g., gender, weight status, disability status) to further characterize and address intersecting systems of oppression that disparately influence ED outcomes.

Higher-level-of-care eating disorder (ED) treatment programs are often designed specifically for individuals with restrictive EDs, and those with binge-eating spectrum disorders (BESD) have different needs. The present study sought to understand predictors of outcome in a residential treatment program designed specifically for BESD. Ninety-nine participants (81% female, 80% White, Mage = 34) admitted to a residential program designed specifically for BESD completed assessments at admission, discharge, 6-month follow-up, and 12-month follow-up. We examined ED-related and comorbid symptom predictors of binge eating severity at discharge and follow-up. In multivariate regression models, cognitive restraint, depression, anxiety, and trauma symptoms significantly predicted binge eating severity at discharge, but at 12-month follow-up, only depression and trauma symptoms remained significant predictors. Individuals with lower cognitive restraint and comorbid symptoms may benefit more from this type of treatment. Further tailoring BESD residential treatment to provide additional skills for managing cognitive restraint and comorbid symptoms may be warranted.

OBJECTIVE:Few studies have examined differences in eating disorder (ED) and general psychopathology among those who self-induce vomiting only compared to those who misuse laxatives only and those who engage in both types of behaviors. The current study aimed to fill this gap in the literature. METHOD:Participants were 1652 adults receiving treatment for an ED at a large multisite treatment facility in the United States between January 2021 and December 2024. Participants completed measures of ED psychopathology (Eating Disorders Examination-Questionnaire; EDE-Q), depression (Patient Health Questionnaire-9; PHQ-9), anxiety (Generalized Anxiety Disorder-7; GAD-7), and obsessive-compulsive symptoms (Obsessive-Compulsive Inventory-Revised; OCI-R) at admission. RESULTS:Significant differences among groups were found on all EDE-Q subscales and the Global score, with the laxative only group and the group engaging in both behaviors scoring significantly higher than the vomiting only group. A similar pattern of results was found for the PHQ-9, while significant differences on the GAD-7 and OCI-R failed to remain significant after multiple comparison corrections. DISCUSSION:It is possible that those who misuse laxatives, or engage in both laxative misuse and vomiting, may benefit from more intensive treatment than those who engage in self-induced vomiting only. These individuals may also need a different form of treatment if laxative misuse is conceptualized as an addiction.

A dearth of information exists on differences in quality of life across eating disorder diagnoses. The purpose of the current study was to examine quality of life among a large sample of adults with one of six eating disorder diagnoses: anorexia nervosarestricting subtype (AN-R), AN binge/purge subtype (AN-BP), bulimia nervosa (BN), binge eating disorder (BED), avoidant/restrictive food intake disorder (ARFID), and otherwise specified feeding or eating disorder (OSFED). Participants were 1268 adults receiving treatment for an eating disorder across multiple levels of care who completed a measure of quality of life, the WHOQOL-BREF, at admission. Adjusted pairwise comparisons with multiple comparison correction found significant differences among eating disorder groups, even after controlling for age, sex, level of care, body mass index (BMI), and the interaction between BMI and eating disorder diagnosis. Among the five significant findings, four of them involved differences in those with ARFID in comparison to the other eating disorder diagnoses. On the physical health subscale, individuals with BN scored significantly better than did individuals with ARFID. On the psychological subscale, patients with ARFID had significantly better quality of life than those with BN, BED and OSFED. On the social relationships subscale, the BED group scored significantly worse than those with AN-BP. This large study of adults with eating disorders suggests that there are meaningful differences among eating disorder diagnostic groups in quality of life.

OBJECTIVE:Knowledge of family functioning (FF) for those with eating disorders (EDs) is driven by research with females, resulting in an overly gendered perception of FF. The current study: (1) descriptively examined FF among male adolescents with EDs, (2) compared FF among males with anorexia nervosa-restricting subtype (AN-R), AN-binge/purge subtype (AN-BP), and avoidant/restrictive food intake disorder (ARFID), and (3) compared FF between males and females with these EDs. METHOD:Participants were 175 males and 175 females who completed the Family Assessment Device (FAD). RESULTS:Males scored above the clinical cutoffs on most FAD subscales. No differences in FF were found among males across ED diagnoses. Significant differences were found between males and females with AN-R on four FAD subscales (affective involvement [OR = 4.70], affective responsiveness [OR = 2.52], communication [OR = 2.78], and general functioning [OR = 2.22]), with males reporting worse FF (all ps < 0.03). Differences between males and females with AN-BP or ARFID were not large enough to meet statistical significance. CONCLUSIONS:This study increases understanding of FF in EDs from a more diverse standpoint. Male adolescents with EDs experience poor FF. Qualitative studies could clarify possible reasons behind poor FF for adolescent males with EDs and help to identify specific targets for treatment.

BACKGROUND:Fearlessness about death, an essential component of suicide capability, remains understudied in the context of actual death-related experiences, whether suicidal in nature or not. One population in which both suicidal intent and desire and life-threatening states are relevant is eating disorder patients in potentially lethal medical crises. METHODS:We examined changes in fear of death across hospitalization in a sample of 84 severely malnourished patients, diagnosed with anorexia nervosa (93%) or avoidant/restrictive food intake disorder (7%), undergoing intensive treatment for severe weight loss in an inpatient medical stabilization unit. RESULTS:In contrast to the typical distribution of fear of death scores found in other samples, scores in our intensive care unit eating disorder sample displayed significant bimodality, with some patients indicating elevated fear of death at admission and some indicating no fear of death at all. Still, despite elevated desire and intent to die on average at admission, the modal reaction to intensive medical care was initially high levels of fear of death that subsequently decreased across treatment. CONCLUSIONS:Findings indicate significant increases in fear of death, a protective factor against suicide, for individuals facing death as a distinct and real possibility, that decreases as stabilization occurs. However, results also suggest a potential bifurcation of patients such that a subgroup of patients experience alarmingly low levels of fear of death, even at the height of their medical crisis. Implications for recurrence of disorder leading to rehospitalization or increased suicidal capability leading to future suicide attempts are discussed.

OBJECTIVE:This retrospective cohort study sought to determine whether higher levels of engagement with the Recovery Record (RR) app were associated with better outcomes on eating disorder (ED) symptoms (Eating Disorder Examination-Questionnaire) (EDE-Q), depression (Patient Health Questionnaire-9), and generalised anxiety (GAD-7) for individuals receiving ED treatment. METHOD:Participants were 4852 adolescents and young adults receiving treatment between March 2021 and August 2024. RR was made available to all patients during their treatment. RESULTS:Cluster analysis identified three distinct groups: engagement, low engagement, and no engagement. After controlling for baseline severity and treatment duration, engagement was correlated with lower EDE-Q scores at discharge compared to low engagement (β = -0.18, p = 0.010). Engagement was also correlated with significantly lower PHQ-9 scores at discharge relative to both low (β = -0.93, p = 0.004) and no engagement (β = -0.84, p = 0.015) groups. Engagement was associated with lower GAD-7 scores at discharge compared to low engagement (β = -0.90, p = 0.003). Greater average daily logging in the RR app was associated with incremental improvements in all outcomes: EDE-Q (β = -0.042, p = 0.030), PHQ-9 (β = -0.287, p = 0.001), and GAD-7 (β = -0.216, p = 0.010). CONCLUSIONS:Findings suggest that engagement with the RR app is associated with better outcomes in terms of ED symptoms, depression, and anxiety.

OBJECTIVE:The purpose of this study was to add to the limited literature on family functioning among adolescents with avoidant/restrictive food intake disorder (ARFID). METHOD:Participants were receiving treatment for ARFID (n = 89) or anorexia nervosa-restricting subtype (AN-R) (n = 66), matched on age, percentage of expected body weight, and treatment length of stay, at a multisite eating disorder treatment facility. All patients completed the Eating Disorders in Youth-Questionnaire (EDY-Q) (a measure of ARFID symptoms) at admission and discharge. Family functioning was assessed with the Family Assessment Device (FAD). RESULTS:No family functioning scores were significantly different between patients with ARFID and patients with AN-R, and patients reported poor family functioning on nearly all FAD subscales. No family functioning scores were associated with EDY-Q scores at admission or discharge. No changes in family functioning were associated with discharge EDY-Q scores. CONCLUSIONS:Findings suggest that family functioning among adolescents with ARFID is impaired and comparable to those with AN-R.

OBJECTIVE:Leptin and thyroid hormones are interrelated regulators of energy homeostasis that are altered during starvation, yet their within-subject trajectories during early nutritional rehabilitation in severe restrictive eating disorders (EDs) remain poorly understood. This study examined how leptin and thyroid hormones change with initial weight gain in severely malnourished inpatients with an ED. METHOD:Data from 62 adult inpatients admitted to a specialized medical unit were analyzed. Blood samples measuring leptin, triiodothyronine (T3), thyroxine (T4), and reverse T3 (RT3) were obtained at admission, each 1.5 kg/m2 increase in BMI, and discharge. Receiver operating characteristic (ROC) curves identified optimal BMI thresholds for clinically abnormal hormone levels, and multilevel models examined hormonal trajectories over BMI change. RESULTS:At admission, leptin was undetectable in 74.2% of patients, T3 was low in 96.8% of patients, T4 was normal in all patients, and RT3 was low-to-normal in most patients. By discharge, most patients still remained underweight. Leptin remained undetectable in 24.2% of patients, and T3 normalized in only 22.6%. No BMI thresholds were clinically significant for identifying abnormal lab levels. Leptin increased linearly with BMI gain, while thyroid hormones demonstrated a biphasic pattern: T3 rose exponentially before plateauing, while T4 and RT3 declined sharply before rebounding, all around +2.0 kg/m2. CONCLUSIONS:T4 and RT3 did not demonstrate expected trajectories for severe malnutrition, while T3 and leptin trajectories were proportional to weight gain but incomplete. Additional follow-up through complete weight restoration is needed to fully understand the interrelation of these hormones, and how and when they normalize.

OBJECTIVE:For multiple disorders, DSM-5-TR criteria include "recurrent" engagement in behaviors-for example, in the criteria for binge-eating disorder (BED), "binge eating is not associated with the recurrent use of inappropriate compensatory behavior."-without operationalizing how frequently a behavior needs to occur to be considered "recurrent." This lack of specificity may be particularly problematic in diagnosing BED; BED shares features with bulimia nervosa (BN) but is differentiated based on lack of "recurrent" ICB engagement. Some treatments (e.g., behavioral weight loss) may be recommended for BED but are contraindicated for BN, making accurate diagnosis critical. We sought to determine what frequency of past-month inappropriate compensatory behaviors (ICBs) best distinguishes BED from BN. METHOD:850 treatment-seeking adolescents and adults (76% female, Mage = 29) completed eating disorder and comorbid symptom questionnaires at program admission from 2020 to 2024. Structural Equation Modeling (SEM) Trees derived which ICB frequencies best distinguished diagnoses. RESULTS:SEM Tree analyses yielded four subgroups: past-month ICB frequency 0 (ICB0; n = 349); frequency 1-25 (ICB1; n = 307); frequency 26-55 (ICB2; n = 143); and frequency > 56 (ICB3; n = 51). ICB0 differed from higher frequency groups on 11/12 clinical characteristics, ICB1 differed from higher frequency groups on 11/12 characteristics, and ICB2 differed from ICB3 on 1/12 characteristics. DISCUSSION:Findings suggest one past-month ICB may appropriately distinguish between BED and BN, though it remains unclear how fasting may impact this categorization. While more research is needed, DSM's diagnostic criteria may require revision to specify that the presence of any past-month vomiting, laxative use, or maladaptive exercise should preclude a BED diagnosis.

OBJECTIVE:While historically viewed as illnesses of youth, restrictive eating disorders (EDs), including anorexia nervosa, restricting subtype (AN-R), and avoidant/restrictive food intake disorder (ARFID), affect women of all ages, yet there is limited understanding of effective interventions and best practices for refeeding of older patients. This study examined whether age affects kcal needs during refeeding and whether caloric needs differ between diagnoses. METHOD:In this retrospective cohort study, average kcal needs for a one-point increase in body mass index (BMI) were assessed for 417 female patients receiving inpatient medical stabilisation for AN-R or ARFID, comparing females < 20 years of age, 20-29, 30-39, and 40+ years using linear regression and chi-square tests of independence. RESULTS:Average kcals needed for a one-point increase in BMI was 1870 kcal (SD = 1410). Average kcals needed per one-point increase in BMI were not statistically different amongst the age groups (p = 0.30). Patients with AN-R needed significantly fewer kcals than patients with ARFID for a similar amount of weight gain. DISCUSSION:In this high-acuity inpatient sample, caloric needs for weight restoration were similar across age groups. Patients with ARFID required more calories than those with AN-R. Age-based prescription reductions may therefore be unwarranted, supporting individualised, diagnosis-informed refeeding.

Background: While virtual therapy has proven effective in treating eating disorders (EDs), little work has examined virtual therapy at higher levels of care, which are treatment options providing more support than weekly outpatient therapy including intensive outpatient (IOP) treatment. Objective: This study aimed to add to the limited research on in-person versus virtual treatment at a higher level of care by comparing treatment outcomesbetween an in-person IOP and a virtual IOP (VIOP) for patients with EDs. We hypothesized that there would be no differences in improvements between VIOP and in-person IOP groups. Methods: This study has a nonrandomized multiple cohort design. Patients with EDs receiving treatment who completed both admission and discharge questionnaires in VIOP treatment (n=231) and in-person IOP treatment (n=39) between 2021 and mid-2022 within a large ED health care system in the United States were included. The Eating Disorder Examination-Questionnaire (EDE-Q) was used to measureED symptoms. The Patient Health Questionnaire-9 (PHQ-9) was used to measuredepression, and item 9 of the PHQ-9 was used to measure suicidal ideation. Welch t tests on admission, discharge, and raw change scores were conducted. Logistic regressions were conducted predicting treatment program (reference group VIOP vs in-person IOP) from the residualized change in each outcome and were adjusted for all significantly different factors between groups. Results: VIOP patients were significantly older (mean 28.03, SD 11.09) than in-person IOP patients (mean 19.51, SD 6.98) and displayed significantly different numbers of ED diagnoses and morecomorbid psychiatric diagnoses (VIOP: mean 1.23, SD 1.12; in-person IOP: mean 0.33, SD 0.84) but no differences in race (VIOP: 175/231, 75.6% White; in-person IOP: 30/39, 76.9% White), gender (VIOP: 196/231, 84.8% female; in-person IOP: 35/39, 89.7%female), or length of stay (VIOP: mean 58.84, SD 26.69; in-person IOP: mean 57.33, SD 19.67). When compared to in-person IOP patients, controlling for age, diagnosis, number of comorbid diagnoses, and admission scores, VIOP patients did not exhibit significantly different improvements in ED symptom scores (EDE-Q Global: b=0.01, SE 0.18, t=0.04, odds ratio [OR] 1.01, 95% CI 0.71-1.43; P=.97). However, VIOP patients exhibited significantly greater improvements in depression scores (PHQ-9: b=-0.14, SE 0.05, t230=-2.85, OR 0.87, 95% CI 0.79-0.96; P=.004) andthePHQ-9suicidal ideation item(PHQ-9 item9: b=-0.72, SE 0.34, t230=-2.13, OR 0.49, 95%CI 0.25-0.93; P=.03). Conclusions: ED outcomeswere similar for VIOP and in-person IOP patients. Contrary to our hypotheses, depression and suicidal ideation outcomesimproved more for VIOP patients than for in-person IOP patients. Furthermore, treatment access for non-White and older adults does not appear descriptively worsefor VIOP treatment compared to in-person IOP treatment, though these trends should be further explored. VIOP treatment may improve treatment access in an equitable fashion without reducing treatment quality.

The Eating Disorder Examination Questionnaire (EDE-Q) is one of the most widely utilized eating disorder (ED) assessments. However, its measurement structure remains obscured by mixed findings, which may be due to studies primarily featuring community samples with limited clinical ED symptom relevance and potential measurement noninvariance across ED types. The present study aimed to employ both factor analytic and item response theory analyses in a clinical sample of individuals (n = 2,032) seeking ED treatment at a higher level of care facility to discern the EDE-Q's structure and invariance across sociodemographic and clinical characteristics including age, race, gender, ED type, and treatment setting. Study aims also included generation of a reduced-item EDE-Q that reflected its truer measurement structure and scores with greater interpretability. Factor analytic and item response theory models uniformly indicated the EDE-Q is unidimensional with items reflecting overall ED symptom severity. Removing eight items based on item residual covariance patterns and ED theory based on expert consensus yielded a 13-item EDE-Q that had improved unidimensional fit and retained majority of the information conveyed by the original scale. The 13-item EDE-Q was also invariant across age, race, gender, level of care, and ED type. The 13-item EDE-Q is recommended as an alternative to the original and previously proposed models, as it has a more reliable total score, has better goodness of fit, and is invariant across sociodemographic and clinical characteristics. Nonetheless, more work is required to develop scales that capture specific cognitive, behavioral, and affective components of disordered eating. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

BACKGROUND:Weight variability (WV), or daily-to-weekly fluctuations in weight, associates with increased eating pathology in adults with bulimia nervosa and greater weight gain across the developmental span in healthy controls, but few studies have explored these relations in adolescents with anorexia nervosa (AN). Given the importance of early weight gain during treatment for AN prognosis, WV could impact outcomes via effects on weight trends and related psychopathology. The current study examined whether WV and the slope of weight change predict eating disorder symptoms at the end of treatment (EOT) among adolescents with AN. METHOD:Adolescents with AN (N = 284) receiving residential treatment completed the Eating Disorder Examination Questionnaire (EDE-Q) at admission and EOT. WV was calculated using the root-mean-squared-error of daily weights over the first 14 days of treatment. Linear regressions examined the effect of WV, slope of weight change over 14 days, and their interaction on percent expected body weight (%EBW) and EDE-Q scores at EOT. RESULTS:WV positively predicted EDE-Q Global Score (p = 0.033, sr 2  = 0.01), Shape Concern (p = 0.026, sr 2  = 0.01), and Weight Concern (p = 0.008, sr 2  = 0.02) at EOT. Neither WV nor the slope of weight change predicted %EBW at EOT (ps > 0.05). The slope of weight change did not predict EDE-Q, nor did it moderate the relation between WV and any outcomes (ps > 0.05). CONCLUSIONS:Patients who experience greater weight fluctuations early in treatment may be more susceptible to elevated psychopathology at discharge. In light of few significant effects and small effect sizes, more research is needed to determine the putative role of early weight trends in adolescent AN outcomes.

OBJECTIVE:Single-item questionnaire assessments may be summarily dismissed by some, but it is unclear if this is warranted. We suggest here that the answer to that question depends on their reliability and construct validity. Reliability of a single-item index cannot be assessed via indices like coefficient alpha, but it can be addressed otherwise; construct validity can be assessed as it would be with any other quantitative index. METHODS:In 489 patients with mood and anxiety disorders, we examined the Beck Depression Inventory Item 9's (BDI-Item 9) test-retest coefficient, and its validity, in part via invariance analyses. More specifically regarding invariance analyses, we determined whether a covariance structure including a multi-item assessment of suicidal ideation was invariant with a similar structure using a single-item assessment of suicidal ideation instead. RESULTS:Reasonably supportive evidence emerged for the reliability and construct validity of this specific one-item index. CONCLUSIONS:The BDI-Item 9 can reliably assess suicidal ideation when data lack multi-item assessments.

OBJECTIVE:Refeeding oedema, believed to result from the effects of insulin on renal sodium retention and subsequent oedema formation, typically occurs during the first 2 weeks after reintroduction of nutrition in individuals with severe malnutrition and can intensify body image distress in patients with eating disorders (EDs). Phosphate supplements have been found to increase insulin sensitivity, and it is hypothesised that they may also contribute to refeeding oedema in patients with EDs. METHOD:In this retrospective cohort study of 633 patients with severe malnutrition due to anorexia nervosa (AN) or avoidant restrictive food intake disorder (ARFID), the impact of phosphate supplementation on the rate of weight gain was investigated. RESULTS:39% of patients required treatment with phosphate supplements during their hospital course. Patients in the phosphate supplement group had significantly more weight gain than the cohort who did not receive a phosphate supplement (p < 0.05). The rate of weight gain also increased, in the phosphate supplement group, as admission body mass index decreased. CONCLUSIONS:This is the first study to suggest that crucial phosphate supplementation contributes to refeeding oedema in patients with EDs. Future studies should seek to better understand the relationship between phosphate supplementation and weight trends in order to optimise clinical management.

Inpatient care is a significant contributor to the health-care costs associated with the treatment of individuals with eating disorders (EDs). This study seeks to better understand the treatment outcomes of individuals with extreme EDs requiring inpatient medical stabilization at specialized medical versus nonspecialized medical units. Patients transferred via air ambulance to ACUTE Center for Eating Disorders and Severe Malnutrition (specialized care) from an outside hospital (nonspecialized care) between January 2022 and March 2024 were included in this retrospective, non-randomized cross-over trial. Weekly weight trends, length of stay, unit on which patients received medical care, and development of medical complications after admission were compared between the nonspecialized the specialized inpatient medical stabilization units. Final analysis included 92 patients. Those with restricting EDs experienced an average of -0.9 kg/week of weight gain while receiving nonspecialized medical care, compared to 1.6 kg/week of weight gained at the specialized medical stabilization unit. Individuals with purging EDs experienced an average of -0.2 kg/week weight change at the nonspecialized medical units compared to 1.7 kg/week gained at the specialized unit, after excluding patients who developed complications related to improper fluid management. Patients were also more likely to be treated in an intensive care unit and develop increased iatrogenic complications while receiving nonspecialized medical treatment compared to those patients treated on a specialized ED unit. Specialized, inpatient medical care for those with extreme EDs is associated with positive weight trends and reduced rates of iatrogenic complications, compared to nonspecialized inpatient medical care.