Areas for Further Research

• Determine whether identification of an eating disorder using routine or targeted screening is associated with benefits on patient-oriented outcomes

• Determine whether patient characteristics and symptoms can be used to identify patterns of disordered eating that would warrant early intervention to prevent onset of an eating disorder

• Identify risk factors for development of an eating disorder that could be used in defining subgroups of individuals who warrant prospective screening or could benefit from preventive interventions

• Identify population-based approaches to preventive interventions (e.g., address impact of social media on eating disorder development; Chung et al. 2021)

• Validate existing rating scales for screening, assessment, and session-by-session treatment change in the major types of eating disorders (e.g., AN, BN, BED, ARFID, other specified feeding and eating disorders [OSFED]) and among a broad range of ages, genders, cultures, languages, symptom patterns (e.g., focusing on eating, body shape, muscularity, driven exercise), and settings (e.g., primary care, specialty care)

• Determine whether additional screening, assessment, or longitudinal rating scales need to be developed to ensure validity and reliability in the major types of eating disorders (e.g., AN, BN, BED, ARFID, OSFED) among a broad range of ages, genders, cultures, languages, symptom patterns (e.g., focusing on eating, body shape, muscularity, driven exercise), and settings (e.g., primary care, specialty care)

• Determine whether useful clinical assessment measures might be developed based on NIMH’s Research Domain Criteria (RDoC) frameworks (Monteleone et al. 2020; Schaefer and Steinglass 2021; Wildes and Marcus 2015) for measures such as appetitive signaling, anxiety related to social processes, reward learning, and reward prediction errors related to eating and thinness

• Determine ways to optimize short- and long-term patient outcomes, including recovery, using factors and approaches such as

  1. Early identification and intervention

  2. “Stepped-care” approaches, which start with less intensive treatment and shift to more intensive interventions, as needed, to achieve recovery

  3. Telehealth (individual, group, and family)

  4. Remote physiological monitoring

  5. Large-scale data analytics and predictive algorithms

  6. Self-help and GSH approaches, including groups, manual-based approaches, or computer-based programs (including Web-based, phone applications, chat bots, and other modalities)

  7. Family/caregiver interventions, including support groups and psychoeducation

  8. Involving certified peer support specialists as part of the multidisciplinary team

  9. Augmenting treatment with other psychosocial therapies (e.g., creative art therapies, cognitive remediation therapy) or complementary therapies

  10. Modifying treatment to improve physical health and address co-occurring health conditions, including substance-related and addictive disorders and other psychiatric disorders

  11. Modifying treatment to address significant symptoms such as suicidal ideas and behaviors, obsessions and compulsions, and perfectionism

  12. Modifying treatment to address attachment-related issues or traumatic experiences, including adverse childhood experiences

  13. Developing new treatments to target key processes in eating disorders (e.g., satiety, hunger, energy expenditure, cognitive rigidity, self-efficacy, body dissatisfaction, self-image disturbances)

  14. Developing treatments to address transdiagnostic processes involved with eating disorders and with common co-occurring conditions

• Identify clinical indicators, biomarkers, and other factors that can help in individualizing treatment selection, frequency, and duration to achieve optimal patient outcomes

• Identify clinical indicators, biomarkers, and other factors that can help in determining an optimal sequence of treatments if an initial therapeutic modality is not associated with response or recovery

• Identify approaches to individualizing treatment selection and delivery to optimize outcomes for individuals of different ages, developmental stages, sexes, genders, races, ethnicities, and cultural groups, among other individual facets

• Identify optimal approaches to treatment of individuals with a longstanding eating disorder as compared to individuals with a more recent onset of symptoms

• Obtain additional evidence on novel or existing psychotherapies (e.g., DBT, mindfulness, acceptance and commitment therapy, mentalization-based therapy) in treatment of eating disorders

• Obtain additional evidence on novel or existing pharmacotherapies in the treatment of eating disorders

• Conduct studies on the comparative effectiveness of psychotherapies and other interventions in treating eating disorders

• Identify optimal approaches to providing multidisciplinary team–based care of eating disorders

• Determine the circumstances in which “bundled” treatment programs are appropriate to use, including the elements of these programs that enhance patient outcomes

• Identify optimal dietary and nutritional interventions for each of the eating disorders, including the ways in which these interventions may need to be adjusted to specific patient needs, symptom severity, or clinical progress

• Identify clinical considerations in assessment and monitoring as well as optimal approaches to providing treatment to individuals with an eating disorder who wish to become pregnant, are pregnant, or are breastfeeding.

• Determine which factors can be used in selecting an optimal treatment setting

• Determine optimal monitoring frequencies and approaches to detect treatment-related benefits and side effects

• Identify optimal approaches to preventing relapse once remission from an eating disorder has been achieved

• Develop “step-down” approaches to care to reduce relapse and avoid discontinuities in care

• Identify the treatment elements and approaches that are viewed as most and least helpful by individuals who have recovered from an eating disorder

• Identify methods that will allow information from mobile technologies, wearable technology, and large-scale data analytics to inform assessment, treatment, and future research

• Identify approaches to redesigning workflows and models of care delivery to improve the use of best practices and reduce inequities in the care of individuals with an eating disorder

• Determine the ways in which health system factors and treatment delivery characteristics influence patient outcomes

• Determine approaches to maximize patient engagement, increase motivation for change, and facilitate treatment retention for individuals with AN

• Identify optimal nutritional approaches to weight restoration, including targeted meal-based interventions aimed at normalizing food choice, intake, dietary variety, and macronutrient content (e.g., percent fat intake)

• Determine whether supplemental approaches to feeding (e.g., NGT feeding) are indicated in AN, and if so, the optimal approaches and circumstances in which supplemental feeding can improve outcomes as compared to meal-based approaches

• Determine whether specific interventions (e.g., exposure with response prevention, mindfulness) can be used to address specific symptoms or concerns (e.g., anxiety about eating, body image disturbance)

• Identify predictive factors that distinguish between individuals who respond more quickly to treatment and those who have longer illness courses, with the aim of developing new treatment approaches for individuals with severe and enduring AN

• Determine the circumstances under which specific medications (e.g., olanzapine, antidepressants) may be useful in an individual with AN

• Determine whether there is a role for exercise in the treatment of individuals with AN and, if so, the optimal type, amount, and timing of exercise recommendations

• Identify specific approaches for maintaining weight and behavioral gains and reducing relapse risk in AN once weight restoration is achieved

• Determine ways in which treatments for AN and monitoring for medical sequelae of AN may need to be adjusted for older adults

• Determine optimal approaches for treating co-occurring disorders in an individual with AN, including whether such treatments should occur simultaneously or sequentially

• Determine optimal approaches for minimizing symptoms associated with renourishment (e.g., GI dysmotility, edema, electrolyte abnormalities, cardiac effects)

• Determine optimal approaches to prevent, identify, and treat short- and long-term medical sequelae of AN, including individual characteristics (e.g., age, gender, race, ethnicity, co-occurring conditions, family history) that can affect development of these sequelae.

• Determine the physiological and other factors that contribute to low BMD in AN, approaches to addressing BMD, and gender-associated differences in physiology and treatment of low BMD

• Determine the ways in which individuals with AN experience inequity in assessment, treatment, and outcomes due to factors such as age, gender, sexual orientation, race, ethnicity, culture, weight, body size, social determinants, and insurance status, so that these health inequities can be ameliorated

• Determine whether fluoxetine or other SSRIs are effective in adolescents and emerging adults with BN

• Determine optimal approaches to treating individuals with BN who have multiple co-occurring conditions or whose diagnosis has shifted from AN-binge/purge subtype to BN

• Determine the ways in which individuals with BN experience inequity in assessment, treatment, and outcomes due to factors such as age, gender, sexual orientation, race, ethnicity, culture, weight, body size, social determinants, and insurance status so that these health inequities can be ameliorated

• Determine optimal psychotherapeutic and pharmacological approaches to treating adolescents and emerging adults with BED

• Determine whether optimal treatment approaches for individuals with BED differ in individuals who are obese as compared to those who are not obese

• Determine whether specific treatments for obesity (e.g., diet approaches, surgical approaches) are associated with a different profile of benefits and harms when used in individuals with BED

• Determine the ways in which individuals with BED experience inequity in assessment, treatment, and outcomes due to factors such as age, gender, sexual orientation, race, ethnicity, culture, weight, body size, social determinants, and insurance status so that these health inequities can be ameliorated

• Determine optimal approaches for screening, assessment, and evaluating session-by-session treatment change in individuals with ARFID among a broad range of ages, genders, cultures, languages, and symptom patterns

• Validate potential subtypes of ARFID including their patterns of signs and symptoms as well as their natural history

• Determine optimal approaches to treating individuals with ARFID, expanding on work with parent-based approaches (Shimshoni and Lebowitz 2020) and CBT (Thomas et al. 2020, 2021) as well as developing new approaches to treatment

• Identify variations in ARFID presentation across the lifespan and whether adjustments in treatment are needed for individuals in different age groups

• Determine optimal approaches to treating individuals with night eating syndrome

• Determine the optimal approach to setting target weights for individuals with atypical AN

• Determine ways in which treatment for atypical AN may need to differ from treatment of AN

• Determine modifications in treatment that may be needed for individuals who have had a shift in diagnosis (e.g., from the restricting subtype to the binge/purge subtype of AN, from the binge/purge subtype of AN to BN).

• Determine the ways in which individuals with other eating and feeding disorders experience inequity in assessment, treatment, and outcomes due to factors such as age, gender, sexual orientation, race, ethnicity, culture, weight, body size, social determinants, and insurance status so that these health inequities can be ameliorated

• Determine the optimal approaches to assess patients’ capacity to accept or decline treatment in eating disorders, particularly restrictive eating disorders

• Identify circumstances under which compulsory or coercive treatment of an eating disorder may be ethically justifiable

• Determine the outcomes of compulsory or coercive treatment of an eating disorder (e.g., hospitalization, NGT feeding) as compared to declining treatment or receiving voluntary treatment

• Identify optimal approaches to providing palliative care to individuals with severe and enduring AN

• Determine optimal approaches (e.g., verbal communications, electronic information sharing via patient portals or open notes) for involving family in treatment while also protecting the privacy and confidentiality of adolescents and emerging adults

• Identify ways in which social media influences eating disorder symptoms and treatment engagement

• Determine whether specific policy recommendations, regulatory requirements, or adjustments to social media algorithms can reduce the deleterious effects of social media on individuals who have an eating disorder

• Develop methods to ensure that screening occurs in all patients undergoing a psychiatric evaluation and that diagnostic assessments are conducted in all patients reporting symptoms consistent with disordered eating, because there is no current evidence supporting ethnic or racial differences in the prevalence and presentation of eating disorders

• Identify ways eating disorder risks, assessment, treatment, and outcomes are affected by biases and discrimination (by society and by health care professionals) related to factors such as age, gender, sexual orientation, race, ethnicity, culture, weight, body size, and social determinants

• Identify effective approaches to reducing and eliminating health disparities due to bias and discrimination in the assessment and treatment of individuals with an eating disorder

• Determine whether specific policy recommendations, regulatory requirements, or health care service delivery interventions can reduce disparities in patients’ access to care based on factors such as age, gender, sexual orientation, race, ethnicity, culture, weight, body size, and social determinants as well as insurance status and geographical location

In addition to these specific topics that would benefit from additional research, our ability to draw clinically meaningful conclusions from research would be augmented by improvements in the design of studies. These include

• Improve the generalizability of study populations

• Enhance study recruitment approaches and use a priori specification of analyses to obtain data on treatment effects in subgroups that have been underrepresented in prior research (e.g., inpatients; older individuals; individuals with multiple psychiatric or physical health conditions; individuals with severe and/or persistent illness; diverse samples of individuals in terms of gender, sexual orientation, race, ethnicity, culture, weight, body size, and social determinants)

• Develop approaches to data collection and transparent reporting of sociodemographic factors to facilitate pooling of data from multiple studies and permit assessment of treatment effects in subgroups that have been underrepresented in previous research

• Standardize collection of key data elements and outcome variables as well as information on patient characteristics that are important to risk adjustment of outcomes (e.g., BMI at admission, illness duration, age of illness onset, co-occurring conditions)

• Provide detailed information on processes used for random assignment and masking or blinding to treatment condition

• Report data separately for each diagnostic group in studies that use transdiagnostic samples

• Augment self-report observations with direct measurements of outcome, insofar as possible

• Ensure that sample sizes in clinical studies are adequate to achieve statistical power

• Ensure that studies report data in a consistent fashion with pre-specification of outcomes of interest

• When observations are missing, use appropriate data analytic approaches and perform sensitivity analyses, when indicated, to determine effects of missing data

• Identify instruments for measuring eating disorder symptoms that are efficient and accurate in measuring key outcomes for AN, BN, BED, and other eating disorders and foster standardized and consistent use of such instruments across studies

• Identify standardized approaches for collecting information about factors that ultimately may be useful in individualizing treatment selection (e.g., biomarkers, family history, symptom history, treatment history, and personality traits)

• Ensure that studies identify the magnitude of change in scale scores that would constitute a clinically meaningful difference

• Increase collection of data on patient-centered outcomes (e.g., quality of life, social functioning, physical health, recovery)

• Develop consensus definitions of response, remission, and recovery that can be applied consistently across studies

• Ensure that studies of new treatments, technologies, delivery system modifications, or clinical decision support systems include specific attention to health equitability in implementation methods

• Develop mechanisms such as registries for systematic collection of information on program outcomes as a complement to collecting clinical trial data

• Improve systematic collection of information on harms, including in studies of psychotherapies

• Ensure that studies assess longer-term treatment (e.g., at least 1 year) and long-term follow-up assessments (e.g., 3–5 years) to identify possible long-term harms and patterns of relapse after treatment completion