Clinician Practices in Determining Expected Body Weights Across Eating Disorders: A Mixed Methods Study

ABSTRACT

Objective

Eating disorder (ED) treatment often involves establishing expected body weights (EBWs). While individualised approaches are commonly used, whether and how EBWs are set across EDs, clinician perspectives on different approaches, and clinician training in this practice remain poorly understood. This study addressed these gaps, examining how clinicians navigate determination of EBWs.

Method

A convergent mixed methods design was employed. Semi‐structured qualitative interviews were conducted with 24 multidisciplinary ED clinicians to explore their experiences determining EBWs. Interviews were analysed using reflexive thematic analysis with a critical realist lens. Quantitative data, collected from 161 multidisciplinary ED clinicians via an anonymous online survey, captured the frequency of different approaches across ED diagnoses and age groups, common challenges, and training experiences.

Results

Qualitative and quantitative data underscored clinicians' use of individualised approaches, frequent involvement of nutritional and medical professionals, and the challenge of limited access to patient's premorbid data. While interview participants described determining EBWs for all patients, quantitative descriptive data highlighted that EBWs were primarily determined for youth with anorexia nervosa and atypical anorexia nervosa.

Conclusions

Findings suggest that EBW determination practices vary, in terms of how and for whom EBWs are determined. Evidence‐based guidelines are needed to support consistent EBW determination.

Highlights

• Clinicians commonly use individualised approaches to determine expected body weights.

• Although nutritional and medical professionals are most frequently involved in expected body weight determination, clinicians across professions determine expected body weights.

• Lack of access to patients' historical growth chart data presents a key challenge to determining expected body weights.

1. Introduction and Aims

Weight suppression, the difference between one's highest lifetime weight and current weight, is common across eating disorders (EDs), including anorexia nervosa (AN; Stice et al. ²⁰²⁰; Berner et al. ²⁰¹³; Steinglass et al. ²⁰²⁰), atypical anorexia nervosa (AAN; Hebebrand et al. ²⁰²⁴), bulimia nervosa (BN; Stice et al. ²⁰²⁰; Shaw et al. ²⁰¹²), purging disorder (PD; Stice et al. ²⁰²⁰), and avoidant‐restrictive food intake disorder (ARFID; Brigham et al. ²⁰¹⁸; Thomas et al. ^{2020, 2021}). ED treatment often involves establishing expected body weights (EBWs) to guide weight restoration and alleviate a state of weight suppression, particularly for AN (Lebow et al. 2018) and AAN (Johnson‐Munguia et al. ²⁰²⁴). Clinicians typically determine EBWs with the expectation that physical, psychological, and medical symptoms will improve as patients reach these weights (Golden et al. 2012); however, there is currently no consensus regarding how EBWs should be determined.

Historically, the field has used population‐based approaches, such as setting paediatric EBWs at the median body mass index (BMI) for age and sex (Le Grange et al. 2012) or restoring adults to a BMI above the underweight threshold (e.g., BMI > 18.5; Khalsa et al. ²⁰¹⁷). However, these methods are unlikely to be effective for all patients and may underestimate necessary weight restoration, particularly for patients in larger bodies (Kennedy et al. 2017; Lebow et al. ²⁰¹⁵; Forney et al. ²⁰¹⁷; Harrop et al. ²⁰²¹; Loeb et al. ²⁰²³; Nagata et al. ²⁰¹⁸; Steinberg et al. ²⁰²³; Jary et al. ²⁰²⁴). In response to the limitations of population‐based approaches for determining EBWs, there has been a shift towards using individualised approaches that incorporate a variety of patient‐specific factors, including premorbid weights, height, pubertal stage, BMI percentiles, paediatric growth charts, and parental heights (Schaumberg 2025; Norris et al. ²⁰¹⁸; Steinberg et al. ²⁰²³; Golden et al. ²⁰¹²). Clinician survey data suggest that individualised methods are frequently used in practice. For instance, in a survey of 113 multidisciplinary child and adolescent ED clinicians treating patients with AN, 40.7% reported using data from individual growth charts to determine EBWs (Lebow et al. 2018). Another survey of 141 multidisciplinary clinicians treating patients with AAN found that among the 90 (64%) who used EBWs when treating these patients, 85.6% considered patients' growth charts, 73.3% considered the weight at which a patient's vital signs were previously normal, and 73.3% considered the weight at which a patient's laboratory values were previously normal (Johnson‐Munguia et al. ²⁰²⁴).

Despite understanding that individualised approaches are used regularly to determine EBWs, multiple unknowns remain regarding clinicians' practices setting EBWs. First, research on EBW determination has focused on the treatment of AN (Lebow et al. 2018) and AAN (Johnson‐Munguia et al. ²⁰²⁴), leaving a gap in empirical knowledge of whether and how EBWs are set for other EDs frequently accompanied by weight suppression (i.e., BN, PD, and ARFID). Additionally, while many clinicians report using growth charts to establish EBWs, it remains unclear how they apply this data, such as by referencing the patient's BMI or BMI percentile just before ED onset (i.e., the most recent pre‐onset value), considering the patient's mean BMI or BMI percentile prior to ED onset (i.e., the mean of all pre‐onset values), factoring in the patient's highest recorded BMI or BMI percentile, and/or incorporating information on parental weight status. Finally, clinician perspectives on different approaches for determining EBWs, as well as their training in different approaches, are empirically unexamined to date.

To address these gaps, this study employed a convergent mixed methods design to systematically examine how clinicians determine EBWs for patients with EDs. Mixed methods research enables the integration of qualitative and quantitative data to generate a more comprehensive understanding than either approach alone (Creswell and Inoue 2024). By integrating clinician perspectives from both interviews and survey data, this study offers an in‐depth look at current EBW determination practices to identify specific needs related to evidence‐based practice guidelines for determining and discussing EBWs.

2. Methods

2.1. Participants and Procedure

Participants were eligible if they had at least 1 year of experience treating patients with EDs, were currently working with this population, and were English‐speaking. Recruitment occurred in two stages via social media and professional listservs. In the first stage, participants were recruited for semi‐structured interviews. Twenty‐four participants were purposively sampled from a pool of eligible individuals, to ensure representation across professional backgrounds (i.e., medicine, nutrition, mental health). After providing online consent and completing a brief demographic survey, participants completed an audio‐recorded, video call‐based semi‐structured interview, originally estimated to last 60 min but ultimately ranging from 30 to 75 min. During the interview, participants answered questions relevant to the present study (described below), as well as questions about sharing EBWs with patients, the findings of which are reported separately (Laboe et al. 2025). They received $100 for their participation. In the second stage of recruitment, 161 survey participants completed an eligibility screen, provided online consent, and proceeded directly to an anonymous online survey, which assessed practices regarding EBW determination (described below) and disclosure to patients, as well as demographic and clinical practice characteristics. They were not compensated. All procedures were approved by the University of Wisconsin‐Madison Institutional Review Board.

2.2. Measures

2.2.1. Participant Demographics and Characteristics

All participants self‐reported their gender, age, race, ethnicity, and highest level of education attained. Participants also provided details about their professional background, including their role, practice setting, client age range, and the types of treatment settings in which they work. Additionally, participants disclosed any personal experience with disordered eating.

2.2.2. Qualitative Data

Audio‐recorded semi‐structured interviews were used for qualitative data collection and were conducted via web‐based videoconferencing. The semi‐structured interview guide (available in Supporting Information S1, page 10) covered experiences and rationale for determining (or not determining) EBWs for patients with EDs, challenges faced when determining EBWs, and perspectives on individualised versus population‐based approaches for determining EBWs. Semi‐structured interviews were transcribed verbatim based off audio recordings.

2.2.3. Quantitative Data

A 12‐item survey (available in Supporting Information S1, page 15), partially derived from what was heard in qualitative interviews, was used to collect quantitative data on clinicians' practices around determining EBWs. Three questions assessed professional training in this area, two measured the frequency of determining EBWs for paediatric and adult patients with various EDs (i.e., AN, AAN, BN, PD, ARFID, binge eating disorder [BED]), and two examined the frequency of methods (when applicable) used for each patient population. One question covered challenges in this process, two explored reasons (when applicable) for not determining EBWs, and two assessed practice‐specific factors, including which clinician was responsible for this task. All questions were multiple‐choice. Response items varied by item and included Likert‐type scales (e.g., never to always) for assessing frequency of using different approaches and how often EBWs are determined across EDs, as well as categorical selections (e.g., Therapist/Dietitian/Medical Doctor or Nurse/Other, Yes/No) for assessing who determines EBWs and challenges encountered.

2.3. Data Analysis

2.3.1. Qualitative Analysis

Six researchers were involved in qualitative data collection and analysis. Two trained interviewers conducted the interviews, and all six researchers participated in transcription, coding, and theme development. Each transcript was orthographically transcribed from audio recordings and then independently reviewed by at least two members of the coding team to ensure accuracy. During coding, each transcript was double‐coded by two separate team members, who met to discuss discrepancies. Team‐based, reflexive thematic analysis was used to develop themes from interviews with participants (Braun and Clarke 2022). See Table 1 for a full description of the analytic process.

TABLE 1.

Description of reflexive thematic analysis.

Component	Description
Interview transcription	Prior to analysis, interviews were orthographically transcribed based on audio recordings. During transcription, ‘likes’ and ‘ums’ were replaced with ‘…’ for ease of reading, identifying information (e.g., name of treatment centre) was replaced with a non‐identifying description (e.g., [Treatment Centre Name]), and participants were assigned participant pseudonyms which are used when quoting individual interviews.
Researcher bracketing and reflexivity	Reflexive thematic analysis acknowledges that research is inherently subjective, requiring researchers to engage in reflexivity throughout both the analytic and writing phases (Braun and Clarke 2022). In this study, reflexivity was approached both individually and collaboratively. Prior to coding, team members reflected on their positionality and preexisting assumptions regarding determination of EBWs, considering how their social identities and personal experiences might shape their interpretations of the data (Savin Baden and Major 2013). For instance, the first author, a white, heterosexual woman in a smaller body with personal experience of AAN and treatment at higher levels of care, felt a strong connection to this research. During her own treatment, her EBW was underestimated. Because her BMI was in the ‘Normal’ category, she was not put on a weight restoration plan, even though she had lost a substantial amount of weight during her ED. This underestimation delayed her recovery, contributing to her scepticism about population‐based approaches for determining EBWs. During coding and theming meetings, members of the coding team often verbally reflected how their backgrounds were shaping engagement with the data. Members of the coding team have disciplinary backgrounds in psychology, neuroscience, and human development. Multiple individuals have lived experience with an ED and/or lived experience working with patients with EDs. The team also shared the following identities: White (6/6); smaller‐bodied (6/6); highest degree: Master's (2/6), Bachelor's (2/6), high school (2/6); and woman‐identifying (6/6). Together, the coding team acknowledged several assumptions about determination of EBWs, including: (a) Population‐based approaches underestimate EBWs for patients in larger bodies, preventing full recovery; (b) there are divergent methods for determining EBWs that often conflict with one another; and (c) lack of patient growth chart data limits the ability to determine EBWs using individualised approaches. These assumptions shaped interpretation of the data, but the team remained attentive to perspectives that challenged or expanded their assumptions.
Coding guide development	Members of the coding team (A.A.L., A.H., M.F.S., S.V., S.K., C.R.) independently reviewed all transcripts. Upon reviewing all transcripts, they evaluated the dataset's adequacy and comprehensiveness based on information power, considering the richness of the data in relation to the study's objectives and methodological approach (Malterud et al. 2016). The review indicated that the dataset provided sufficient depth and breadth for analysis. From there, each coding team member independently and inductively generated preliminary coding labels. Then, the team met to refine these labels and establish an initial codebook, which was applied in a test‐coding phase of 12 transcripts (each team member coded two transcripts). Following test‐coding, the team met again to finalise the codebook, which primarily contained semantic codes reflecting explicit meanings in the data.
Coding process	Transcripts were coded in Dedoose (version 9.2.22), with each team member coding between six and 12 transcripts. Each transcript was coded twice by separate coders, who then met to discuss any discrepancies in coding.
Theme development	For the initial identification of themes, the team utilised an online whiteboard to cluster related codes, drawing on their familiarity with the data. This clustering process was guided by the study's central research question: How do clinicians experience determination of EBWs in ED treatment? Once preliminary themes were identified, the first author revisited the dataset to ensure alignment between the themes and the data and revised the themes, which the coding team reviewed.

2.3.2. Quantitative Analysis

Descriptive analyses were conducted to characterise clinician practices related to determining EBWs, including clinician training, who determines EBWs, the EDs for which EBWs are determined, methods used to initially determine and reassess EBWs over the course of treatment, and challenges encountered in determining EBWs.

2.3.3. Integration of Qualitative and Quantitative Data

A convergent mixed methods design involves collecting and analysing qualitative and quantitative data on the same topic to identify areas of convergence and divergence, facilitating a more nuanced understanding of the research question (Creswell and Plano Clark ²⁰¹⁷). For the present study, the research question was: How do clinicians experience determination of EBWs for patients with EDs? Table 2 presents an interview prompts display (McCrudden et al. ²⁰²¹), illustrating the alignment between qualitative and quantitative questions. In addition to qualitative and quantitative analyses, this study included an integration analysis, where qualitative and quantitative findings were connected through a joint display. Joint displays are a tool for visually linking qualitative and quantitative data (Fetters and Molina‐Azorin ²⁰¹⁷).

TABLE 2.

Interview prompts display: Triangulation of qualitative and quantitative data.

Topic	Qualitative interview prompts	Quantitative survey items
How EBWs are determined	How do you calculate EBWs? Is this practice specific to you, or is it clinic‐wide? What do you think about other approaches to setting EBWs? (for example, if a participant discusses individualised approaches, ask them about population‐based approaches)	How do YOU determine EBWs for CHILD/ADOLESCENT (separate question for ADULT) patients with eating disorders? Select all that apply. Do you belong to a larger treatment programme that dictates your policy with regard to setting patients' EBWs?
Who determines EBWs	Do you ever set EBWs for your patients with eating disorders? 1. If no: is there someone you work with who calculates EBWs? If so, who?	Who in your practice is responsible for setting EBWs for patients? Select all that apply. How often do you determine EBWs when treating CHILD/ADOLESCENT (separate question for ADULT) patients with anorexia nervosa? Atypical anorexia nervosa? Bulimia nervosa? Purging disorder? Avoidant‐restrictive food intake disorder?
EDs for which EBWs are determined	With which patients do you set EBWs? Why?	How often do you determine EBWs when treating CHILD/ADOLESCENT (separate question for ADULT) patients with anorexia nervosa? Atypical anorexia nervosa? Bulimia nervosa? Purging disorder? Avoidant‐restrictive food intake disorder?
Challenges of determining EBWs	Think about a recent time you set an EBW. What difficulties did you face when setting this weight?	What challenges do you face when setting EBWs? Select all that apply.
Training in determining EBWs	If no to setting EBWs themselves: in an ideal world, would you want to be able to set EBWs for your patients? If yes: What training or resources would you need to be able to do so?	Have you received training on how to determine EBWs? To what extent do you agree that with your training, you feel confident in your ability to determine EBWs for patients with eating disorders? Would you be interested in receiving additional training on setting EBWs?

3. Results

3.1. Participant Descriptives

Interview participants were predominantly woman‐identifying (n = 22, 91.7%), and all were non‐Hispanic White and residing in the United States, with a mean age of 41.7 years. They came from a variety of professional backgrounds, including nutrition (i.e., registered dietitian, nutritionist; n = 8, 33.3%), medicine (i.e., medical physician, nursing professional; n = 9, 37.5%), psychology (i.e., PhD or PsyD clinician, n = 4; 16.7%), and other mental health fields (i.e., clinical social worker, mental health therapist; n = 3, 12.5%).

Of the 158 (98.1%) survey participants who completed demographic and clinical practice questions, 145 (91.8%) were woman‐identifying, 146 (90.7%) were White, and 148 (93.7%) were non‐Hispanic. The majority of participants resided in the United States (n = 109, 69.0%). The mean age of participants was 40.5 years. Survey participants represented various professional backgrounds, including nutrition (n = 82, 50.9%), psychology (n = 26; 16.2%), other mental health fields (n = 37, 23.0%), and medicine (n = 12, 7.5%). See Table 3 for full details of participant demographics and clinical practice characteristics.

TABLE 3.

Participant demographics and clinical practice information.

Variable	Survey (N = 161)	Interview (N = 24)
Gender—N (%)
Cisgender man	7 (4.4%)	2 (8.3%)
Cisgender woman	145 (91.8%)	22 (91.7%)
Genderqueer	2 (1.3%)	0 (0.0%)
Prefer not to answer	4 (2.5%)	0 (0.0%)
Unknown	3 (1.9%)	0 (0.0%)
Race—N a (%)
White	146 (90.7%)	24 (100.0%)
Asian Indian	1 (0.6%)	0 (0.0%)
Chinese	1 (0.6%)	0 (0.0%)
Filipino	2 (1.2%)	0 (0.0%)
Other Asian	1 (0.6%)	0 (0.0%)
Guamanian	1 (0.6%)	0 (0.0%)
Middle Eastern or north African	1 (0.6%)	0 (0.0%)
Prefer not to answer	7 (4.4%)	0 (0.0%)
Other	1 (0.6%)	0 (0.0%)
Unknown	3 (1.9%)	0 (0.0%)
Ethnicity—N (%)
Not of Hispanic, Latino, or Spanish origin	148 (93.7%)	24 (100.0%)
Mexican, Mexican American, Chicano	2 (1.3%)	0 (0.0%)
Puerto Rican	1 (0.6%)	0 (0.0%)
Another Hispanic, Latino, or Spanish origin	3 (1.9%)	0 (0.0%)
Prefer not to answer	4 (2.5%)	0 (0.0%)
Unknown	3 (1.9%)	0 (0.0%)
Educational attainment—N (%)
Bachelor's degree	29 (18.4%)	1 (4.2%)
Master's degree	94 (59.5%)	11 (45.8%)
Doctor of medicine (MD) or Doctor of osteopathic medicine (DO)	7 (3.8%)	7 (29.2%)
Doctorate degree (PhD, PsyD)	28 (17.8%)	5 (20.8%)
Unknown	3 (1.9%)	0 (0.0%)
Current country of residence—N (%)
United States	109 (69.0%)	24 (100.0%)
Australia	18 (11.4%)	0 (0.0%)
Canada	17 (10.8%)	0 (0.0%)
United Kingdom	9 (5.7%)	0 (0.0%)
Denmark	1 (0.6%)	0 (0.0%)
Iceland	1 (0.6%)	0 (0.0%)
Peru	1 (0.6%)	0 (0.0%)
Spain	1 (0.6%)	0 (0.0%)
Prefer not to answer	1 (0.6%)	0 (0.0%)
Unknown	3 (1.9%)	0 (0.0%)
History of disordered eating or an eating disorder—N (%)
Yes	79 (50.0%)	15 (62.5%)
No	74 (46.8%)	9 (37.5%)
Prefer not to answer	5 (3.2%)	0 (0.0%)
Unknown	3 (1.9%)	0 (0.0%)
Age (in years)—Mean (SD)	40.45 (10.7) b	41.71 (14.2)
Speciality—N (%) a
Clinical social worker	16 (9.9%)	2 (12.5%)
Psychologist	26 (16.1%)	4 (16.7%)
Mental health therapist	24 (14.9%)	1 (4.2%)
Medical doctor	6 (3.7%)	7 (29.2%)
Nurse (RN, RDN, LPN, or APRN)	2 (1.2%)	1 (4.2%)
Nutritionist	12 (7.5%)	0 (0.0%)
Occupational therapist	2 (1.2%)	0 (0.0%)
Physician assistant	1 (0.6%)	0 (0.0%)
Psychiatrist	2 (1.2%)	1 (4.2%)
Registered dietitian	82 (50.9%)	8 (33.3%)
Other	4 (2.5%)	0 (0.0%)
Psychotherapist	3 (1.9%)	0 (0.0%)
Consultant to treatment centres	1 (0.6%)	0 (0.0%)
Unknown	4 (2.5%)	0 (0.0%)
Years of experience treating patient s with EDs—Mean (SD)	11.4 (9.2)	11.3 (10.7)
% Of patients with EDs—Mean (SD)	78.8 (22.9)	78.9 (27.4)
Practice location a —N (%)
Urban	98 (62.0%)	16 (66.7%)
Suburban	84 (53.2%)	16 (66.7%)
Rural	30 (19.0%)	6 (25.0%)
Prefer not to respond	5 (3.2%)	0 (0.0%)
Unknown	3 (1.9%)	0 (0.0%)
Patient population a —N (%)
Children (11 or younger)	56 (35.4%)	13 (54.2%)
Teens (12–17)	117 (74.1%)	23 (95.8%)
Young adults (18–25)	149 (94.3%)	24 (100.0%)
Adults (26–64)	141 (89.2%)	20 (83.3%)
Older adults (65+)	61 (38.6%)	4 (16.7%)
Unknown	3 (1.9%)	0 (0.0%)
Treatment setting a —N (%)
Inpatient	12 (7.6%)	4 (16.7%)
Residential	11 (7.0%)	0 (0.0%)
Partial Hospitalisation	18 (11.4%)	5 (20.8%)
Intensive outpatient	16 (10.1%)	8 (33.3%)
Outpatient	81 (51.3%)	19 (79.2%)
Private practice	116 (73.4%)	8 (33.3%)
Academic medical centre	14 (8.9%)	7 (29.2%)
Non‐Academic medical centre	1 (0.6%)	0 (0.0%)
University Athletics	1 (0.6%)	0 (0.0%)
University health services	3 (1.9%)	4 (16.7%)
Other	2 (1.3%)	0 (0.0%)
Unknown	4 (2.5%)	0 (0.0%)
Therapeutic Model a —N (%) c
Acceptance and commitment therapy	23 (37.7%)	6 (85.7%)
Cognitive behavioural therapy	56 (91.8%)	7 (100.0%)
Dialectical behavioural therapy	31 (50.8%)	7 (100.0%)
Eclectic	18 (29.5%)	2 (28.6%)
Emotion‐focused therapy	7 (11.5%)	1 (14.3%)
Family‐based treatment	29 (47.5%)	4 (57.1%)
Internal family systems	17 (27.9%)	1 (14.3%)
Motivational interviewing	25 (41.0%)	3 (42.9%)
Psychodynamic	16 (26.2%)	2 (28.6%)
Other	10 (16.4%)	1 (14.3%)

^a Participants were allowed to self‐report multiple responses.

^b One participant put zero for their age, and three participants did not answer age questions, so the mean and SD for age for survey completers represents data from 157 participants.

^cOnly mental health professionals (i.e., clinical social worker, psychologist, mental health therapist) were asked about their therapeutic orientation; therefore, percentages are calculated based on a total of 61 respondents rather than the full sample of 161 for survey participants and a total of seven respondents rather than the full sample of 24 for interview participants.

3.2. Qualitative Results

To address how clinicians navigate determination of EBWs in ED treatment, five interrelated themes were identified: gathering data from multiple sources at the start of treatment to individualise estimations, prioritising ‘state’ over ‘weight’ in an ongoing manner, recognising limitations of population‐based approaches, setting EBWs for all patients with EDs, and deferring to medical professionals or nutritional professionals with training and expertise. As shown in Table 3, themes are organised in relation to the guiding research question and aligned with relevant quantitative findings. Individual participant demographics and descriptive information are presented in Supporting Information S1: Table 1 (page 1) to contextualise the excerpts below.

3.2.1. Gathering Data From Multiple Sources at the Start of Treatment to Individualise Estimations

Clinicians described a general two‐step process for determining EBWs, beginning with an initial estimation based on available data at the start of the treatment and adjusting the estimate as treatment progressed. In the first step, clinicians emphasised the importance of individualising EBWs by drawing on multiple sources of information—most commonly, patient growth charts in combination with Centres for Disease Control and Prevention growth curves (e.g., weight‐for‐age, height‐for‐age, BMI‐for‐age; Centre for Disease Control and Prevention, 2000). For example, Allison shared, ‘If a patient presents to me with weight loss, then I'm looking at their growth curve and trying to make an … educated guess of where they would need to get back to.’

Others mentioned using growth curves in tandem with other data, including ‘mid parental height’ (Mary) and ‘weight at last consistent menstrual period’ (Jessica). Susan highlighted the contextual nature of growth interpretation, noting, ‘The growth chart is a big part of it. Depending on the kid, I would put medical history into there too to try to help me interpret the growth chart and understand if there are data points or periods of time that probably actually aren't reflective of their…healthy growth patterns … Occasionally, I'll end up doing things like having to get a bone age that can help, again, with interpretation of the growth chart.’ Ultimately, she concluded, ‘The growth chart is where I get most of my information, but I think that has to be looked at within a context.’

Although using growth curves was most common, participants discussed how sometimes they were not able to access patient growth charts, particularly when working with adult patients. In these cases, participants described collecting ‘a detailed weight history’ (Samantha), including ‘highest lifetime weight’ (Bethany), that would inform their estimate for the EBW.

3.2.2. Prioritising ‘State’ Over ‘Weight’ in an Ongoing Manner

Although clinicians endorsed estimating EBWs at the start of treatment, they underscored that these estimates were not set in stone and were instead a moving target throughout treatment depending upon a patient's psychological, physical, and medical state. In other words, clinicians sought to align EBWs with actual recovery outcomes rather than arbitrary numbers. Lauren captured this commonly expressed sentiment, sharing ‘So we're looking at what you might call a healthy weight. And then there's a healthy state. If one is still very much in their ED at an X weight that we all agreed was the right weight, then it's not the right weight.’

Participants noted factors they considered in an ongoing manner to determine if someone was indeed in the ‘right state.’ e.g., Samantha shared, ‘For a menstruating female, we might look at their period regulation…We might look at blood chemistry … We might look at improved thinking and cognition and a decrease in eating disorder thoughts.’ Similarly, Ann expressed, ‘We take into account their weight charts, their family history, their physiologic functioning, and when you put all that stuff together, I mean, we may set a weight for a patient at a certain place and they get there, and still their period doesn't come … Then we sort of start thinking we have to readjust.’

3.2.3. Recognising Limitations of Population‐Based Approaches

When asked about population‐based approaches to determining EBWs, most participants expressed concern. Regarding use of BMI categories for adults (i.e., setting EBWs at the low to middle end of ‘Normal’), Mark shared, ‘We know that BMI was created as a population tool, not an individual measurement for health … So a client may present with a higher BMI but still be showcasing those classic presentations on the inside of GI distress, and cardiometabolic dysfunction, and hormone dysregulation. So I caution with using those statistical tools because I don't believe that would be helpful in determining someone's individualised care.’ Similarly, Brooke emphasised, ‘I think we're putting limits by using those population‐based BMI metrics, potentially capping people's recovery and not going high enough.’

Participants also expressed concern about using the median BMI for determining child or adolescent EBWs but recognised that sometimes it's the only data they have. For example, Melissa shared, ‘I certainly understand the desire to use the median. And if I'm stuck with nothing…that many times I get called in the middle of the night for kids, and I have no other information, and all I can do is calculate a median BMI for age and sort of know where they are in reference to that. That at least gives me an idea of some safety…I know if someone's 60% of a median, there are safety issues that we might be juggling … And so I don't think there's no use for a median, but I just don't want to be aiming for everyone to be a median.’ When not able to access growth chart data, participants often cited the median BMI or ‘Normal’ BMI as a useful starting point, that could then be reevaluated as treatment progressed.

3.2.4. Setting EBWs for All Patients With EDs

Clinicians framed determining EBWs as a means of combatting weight bias and ensuring adequate nutrition across the weight and diagnostic spectrum. Multiple participants endorsed determining EBWs for all patients with EDs. For example, Melissa shared, ‘I consider really all EDs to have restrictive components. So this is what I do with any ED.’ Similarly, Caroline emphasised, ‘Every single client will have an EBW … Say we have a client that doesn't necessarily have to weight restore, we'll still set an EBW for them.’

Reasons for setting EBW for all patients primarily revolved around providing better care for patients across the weight spectrum. For example, Mary shared, ‘I mostly do it as an anti‐fat bias, anti‐size bias advocate. Because I think in general people settle for really significant underfeeding of patients of most body sizes and most diagnoses, and I think the evidence tells us that, again, food first, that we can't expect somebody who is literally starving to recover to full robust wellness and that weight bias leads us to think differently about people who have different identified behaviours and different identified body sizes.’

3.2.5. Deferring to Medical or Nutritional Professionals With Training and Expertise

This theme revealed a widespread sense of role ambiguity when it comes to determining EBWs. Responses varied regarding which type of provider was responsible for determining EBWs, though most participants pointed to medical professionals or nutritional professionals. Many mental health professionals reported feeling unequipped to make these determinations, citing a lack of training or expertise. Consequently, they described deferring to medical or nutritional professionals to make these determinations. For example, Sandy shared, ‘I rely on the dietitian or if they have a [Primary Care Provider] who's super ED savvy and is tracking their weight trends, I let them do it. I probably could get a little bit better versed in … the numbers and how to calculate it. But I just … try and stay in my lane because I know there are other people who could do that job a lot better than me and probably more accurately than me.’ Similarly, Rachel shared, ‘I prefer not to be the one setting those numbers … I lack the medical background to be able to make those calculations. I don't have any specific training that has taught me how to do that.’

3.3. Quantitative Results

3.3.1. How EBWs Are Determined

Among the 104 participants who endorsed treating youth with EDs and determining EBWs, several methods for initially determining EBWs were frequently reported. The most frequently reported approach was using the weight consistent with the patient's mean Body Mass Index/BMI percentile prior to ED onset; 71 (68.3%) reported using this approach ‘always’ or ‘most of the time’. Fifty‐six (53.8%) reported using the weight at which hormone levels were previously normal (e.g., menstrual threshold) and 53 (51.0%) reported using the weight at which vital signs were previously normal ‘always’ or ‘most of the time’. See Supporting Information S1: Table 2 (page 3) for full details on approaches used to determine EBWs for youth patients with EDs.

Among the 125 participants who endorsed treating adults with EDs and determining EBWs, the most common approach for initially determining an EBW was using the weight at which hormone levels were previously normal (e.g., menstrual threshold). Seventy‐six participants (60.8%) endorsed considering this weight ‘always’ or ‘most of the time’. Using the weight at which vital signs were previously normal (n = 73, 58.4%) and using the weight consistent with the patient's mean adult BMI prior to ED onset (n = 66, 52.8%) were also common. See Supporting Information S1: Table 3 (page 5) for full details on approaches used to determine EBWs for adult patients with EDs.

Among the 132 participants (82.0%) who endorsed determining EBWs, many reported considering various factors that prompted reassessment of their initial EBW determinations. Fifty‐three (40.2%) participants reported adjusting EBWs based on hormonal values (including menstrual status) ‘always’ or ‘most of the time’, followed by 49 (37.1%) who adjusted based on blood pressure and 48 (35.3%) based on heart rate. See Supporting Information S1: Table 4 (page 7) for full details on factors considered when reassessing EBWs.

3.3.2. Who Determines EBWs

Clinicians reported the type of professional who typically determines the EBWs for their patients with EDs. Due to a branching logic error, 126 clinicians (78.3%) out of the final sample of 161 were included. Of those participants, most indicated that EBWs were determined by nutritional professionals (n = 88, 69.8%) or medical professionals (n = 77, 61.1%).

Figure 1 displays which professionals typically determine EBWs, with responses grouped by participants' own professions. This figure highlights that although nutritional and medical professionals were most frequently identified, clinicians across all professions determine EBWs. For example, among doctoral‐level psychologist respondents, over half reported that mental health professionals set EBWs. See Supporting Information S1: Figure 1 (page 9) for a breakdown of how often participants from each professional group reported that they themselves set EBWs ‘most of the time’ or ‘always’ across ED diagnoses and age groups.

FIGURE 1.

The figure shows the percentage of clinicians who report that each type of professional (i.e., doctoral‐level psychologist, medical professional, nutrition professional, other mental health professional) determines expected body weights for patients with eating disorders.

3.3.3. EDs for Which EBWs are Determined

Participants most frequently reported determining EBWs for patients with AN and AAN. Among those who treat youth with AN and determine EBWs (n = 103), 56 (54.4%) reported ‘always’ determining an EBW and 24 (23.3%) reported doing so ‘most of the time’. Among those who treat youth with AAN and determine EBWs (n = 95), 45 (47.4%) reported ‘always’ determining an EBW and 24 (25.3%) reported doing so ‘most of the time’. Participants also endorsed determining EBWs for youth with ARFID (64/92, 69.6%), BN (51/80, 63.7%), and PD (26/43, 60.5%) ‘always’ or ‘most of the time.’ Participants reported calculating EBWs for youth with BED less frequently. Twenty‐nine (39.7%) reported doing so ‘always’ or ‘most of the time.’

Among participants who treat adults and determine EBWs, EBWs were most frequently determined for AN (81/118, 68.6%) and AAN (63/110, 57.4%) ‘always’ or ‘most of the time’. Fewer than half of clinicians reported determining EBWs ‘always’ or ‘most of the time’ for adults with PD, ARFID, or BN. Determination of EBWs for patients with BED was less common (35/108; 32.4%). See Supporting Information S1: Table 4 (page 8) for full distributions across diagnoses and age groups.

3.3.4. Challenges of Determining EBWs

The 132 participants who endorsed determining EBWs reported various challenges they face when determining EBWs, including lack of available patient growth history data (n = 104, 64.6%), EBW not being acceptable to the patient (n = 87, 54.0%), patient history of being underweight (n = 83, 51.6%), the EBW being a moving target (n = 81, 50.3%), and lack of an empirically‐validated, individualised approach (n = 68, 47.2%) to determining EBWs.

3.3.5. Training in Determining EBWs

Across the full sample, slightly more than half of participants (n = 87, 54.0%) reported receiving training in how to determine EBWs through continuing education or workshops, and 69 participants (42.9%) reported receiving training as part of their professional education. Training experiences varied across professional groups, with nutritional professionals (n = 42, 51.2%) and medical professionals (n = 6, 50.0%) most likely to report receiving training during their professional education.

Based on their training, 65.4% of psychologists (n = 17), 58.3% of medical professionals (n = 7), 82.9% of nutritional professionals (n = 68), and 41.7% of other mental health professionals (n = 17) agreed or strongly agreed that they felt confident in their ability to determine EBWs. Across the whole sample, most (n = 118; 73.3%) agreed or strongly agreed that they would be interested in further training.

3.4. Integrated Results: A Joint Display

Integration of qualitative and quantitative findings revealed areas of strong convergence as well as points of partial divergence. Convergence occurred regarding the use of multiple information sources to determine individualised EBWs, the challenges associated with limited historical growth data, and, among professionals, nutritional and medical professionals having the most expertise and practice setting EBWs. However, while qualitative data underscored that mental health professionals outsource the task of determining EBWs to nutritional or medical professionals, quantitative data revealed that clinicians across professions, including mental health professionals, determine EBWs at least some of the time. Divergence also occurred related to the EDs for which EBWs are determined. While interview participants described determining EBWs for all patients as a strategy to promote equitable care, quantitative data highlighted that EBWs were primarily determined for youth with AN and AAN, with less consistent application for other diagnoses and age groups. See Table 4 for a joint display of integrated results.

TABLE 4.

A joint display of integrated results.

Topic	Qualitative results	Quantitative results	Comparison results
How EBWs are determined	Gathering data from multiple sources at the start of treatment to individualise estimations Prioritising ‘state’ over ‘weight’ in an ongoing manner Recognising limitations of population‐based approaches	Individualised approaches (e.g., weight consistent with mean BMI pre‐ED, weight where hormonal values were most recently normal) more commonly endorsed than population‐based approaches (e.g., median BMI, low end of ‘Normal’ BMI). Most participants reported monitoring at least one factor (e.g., hormonal values, vital signs) to reassess EBWs over time.	Individualised methods were prioritised over population‐based approaches, which were viewed as overly simplistic and potentially harmful due to risk of underestimating EBWs. Population‐based tools were used sparingly, typically when patient data was unavailable. Clinicians commonly reassessed EBWs based on functional recovery indicators rather than keeping the EBWs set in stone.
Who determines EBWs	Deferring to medical professionals or nutritional professionals with training and expertise	Most participants reported that EBWs were determined by nutritional professionals (69.8%) and/or medical professionals (61.1%). Nearly a quarter (23.81%) reported that mental health professionals determine EBWs. Nutritional professionals and doctoral‐level psychologists most frequently reported determining EBWs within their respective roles.	In practice, EBWs are most often determined by nutritional and medical professionals, aligning with participants' reported deference in qualitative interviews. However, qualitative data more strongly emphasised mental health professionals' reluctance to take on this role, and quantitative data highlighted that clinicians across professions determine EBWs.
EDs for which EBWs are determined	Setting EBWs for all patients with EDs	Most participants reported determining EBWs for youth with AN, AAN, ARFID, BN, and PD, and for adults with AN and AAN. Determination of EBWs for adults with ARFID, BN, and PD was less common. Determination of EBWs was the least frequent for patients with BED across both youth and adult populations.	While qualitative data emphasised determining EBWs for all patients to support equitable care, quantitative findings showed that this was most consistently applied to AN and AAN. EBWs were less frequently determined for other diagnoses, particularly for adults, highlighting a gap between aspirational practice and actual implementation across diagnoses.
Challenges of determining EBWs	Recognising limitations of population‐based approaches	The most frequently reported challenges to determining EBWs included lack of available growth data (64.6%), EBW not being acceptable to the patient (54.0%), prior history of being underweight (51.6%), and EBWs being a moving target (50.3%).	Unavailable patient data was identified across both qualitative and quantitative findings as a key barrier to using individualised approaches to determine EBWs. In the absence of individualised data, clinicians reported relying on population‐based methods as a pragmatic starting point, with the intention of reassessing EBWs as more information became available or as treatment progressed.
Training in determining EBWs	Deferring to medical professionals or nutritional professionals with training and expertise	Nutritional professionals and medical professionals were most likely to report receiving training during their professional education. Nutritional professionals and doctoral‐level psychologists were most likely to report receiving training through continuing education or workshops. Nutritional professionals were the least likely to report no training, and mental health professionals were most likely to report no training.	Both qualitative and quantitative findings highlighted gaps in formal training, particularly among mental health professionals, which contributed to role deference in clinical practice. Nutritional and medical professionals were most likely to receive formal training and were commonly viewed as responsible for determining EBWs.

Abbreviations: AAN, atypical anorexia nervosa; AN, anorexia nervosa; ARFID, avoidant‐restrictive food intake disorder; BED, binge eating disorder; BMI, body mass index; BN, bulimia nervosa. EBW, expected body weight; ED, eating disorder; PD, purging disorder.

4. Discussion

This mixed methods study examined the determination of EBWs for patients with EDs, including how and for whom clinicians set EBWs, challenges clinicians encounter in determining EBWs, and training clinicians receive in this practice. Qualitative and quantitative findings converged around three core ideas: (1) clinicians commonly use individualised approaches to determine EBWs, (2) clinicians across professions determine EBWs, with nutritional and medical professionals most frequently involved, and (3) lack of access to patients' historical growth chart data presents a key challenge.

4.1. Individualised Approaches to EBWs: Alignment and Ongoing Challenges

Consistent with prior research on EBW determination in AN (Lebow et al. 2018) and AAN (Johnson‐Munguia et al. ²⁰²⁴), participants reported commonly using pre‐ED growth chart data or previous benchmarks for biological functioning (e.g., menstruation, normal vital signs) to guide EBW determination. Population‐based approaches, such as median BMI and above the threshold for ‘Normal’ BMI (> 18.5) were rarely used and were described in interviews as underestimating appropriate EBWs, potentially delaying or undermining recovery, a concern supported by past research (Kennedy et al. 2017; Lebow et al. ²⁰¹⁵; Forney et al. ²⁰¹⁷; Harrop et al. ²⁰²¹; Loeb et al. ²⁰²³; Nagata et al. ²⁰¹⁸; Steinberg et al. ²⁰²³; Jary et al. ²⁰²⁴). Our findings build on prior work by clarifying that the individualised approach that clinicians most frequently use involves growth chart data to estimate a patient's average BMI or BMI percentile before illness.

Despite shared preference for individualised approaches, clinicians reported that lack of access to premorbid growth chart data was a challenge, sometimes leading to use of population‐based approaches. Previous work demonstrates that only 35% of adolescent ED patients had three or more growth chart data points prior to ED onset at initial ED assessment (Harrison 2013). As such, clinical guidance should provide structured options for determining EBWs when historical weight data is lacking. Guidance on assessment and integration of self‐reported weight history may help bridge this gap in real‐world settings. In addition to data access, lack of an empirically‐based, individualised approach to setting EBWs was also identified as a challenge for determining EBWs. While accessible tools for analysing growth chart data and providing EBW estimates are being developed (Schaumberg 2025; Singh et al. ²⁰²⁵), integration of additional metrics often considered in clinical reasoning, such as hormonal and pubertal status, may add precision to current algorithmic estimations.

Clinicians also emphasised that EBWs are not static targets. Many described reassessing EBWs throughout treatment based on patient progress, particularly when biological recovery markers (e.g., hormonal function, vital signs) failed to normalise. These findings echo prior work calling for greater integration between clinical nuance and research definitions of recovery (Bardone‐Cone et al. ²⁰¹⁸; LaMarre et al. ²⁰²³) and suggest that guidelines for determining EBWs must include mechanisms for reassessment.

4.2. While Nutritional and Medical Professionals May Be Most Well‐Equipped, EBW Determination Is a Common Practice Across ED Professionals

While qualitative findings suggest that medical and nutritional professionals are most likely to determine EBWs, most participants (82%) in our quantitative study across ED professional backgrounds reported personally engaging in this practice. Although determining EBW estimations was a common clinical responsibility across participants, only about half of these professionals reported receiving formal training on EBW determination, with 73% percent wanting more formal training. Lack of training may reflect the lack of standardised recommendations for EBW‐setting in general, preventing high‐quality training due to uncertainty about what should be taught. A growing body of research (Steinberg et al. 2023; Schaumberg ²⁰²⁵; Jary et al. ²⁰²⁴) highlights the direct relationship between individualised EBWs and recovery outcomes, suggesting that training in determining individualised EBWs is needed.

Notably, non‐doctoral mental health professionals reported the least amount of formal training in determining EBWs, despite being the most common ED provider (Ellis et al. 2009). While ED treatment ideally includes a multidisciplinary treatment team, this is less common in the treatment of patients with low medical acuity (Pehlivan et al. 2022) and among patients who are uninsured or underinsured and have less access to resources (Moreno et al. 2023; Vendlinski et al. ²⁰²⁵). Further, the current study primarily recruited individuals with ED treatment specialities. Patients often lack access to ED speciality care and may be seen in more generalist settings (e.g., general paediatrics or adolescent medicine; non‐specialised mental health professionals) when this care is not available. As such, while nutritional and medical professionals may have the most expertise in determining EBWs, tools and training to aid EBW estimations should prioritise accessibility and reproducibility for a range of professionals, including decision‐support tools or automated individualised EBW estimators that integrate growth chart data with clinical inputs (Schaumberg 2025; Massara et al. ²⁰²³).

4.3. Calculation of EBWs for All Patients Is Aspirational, But Less Common in Practice

Regarding diagnostic patterns, while interview participants acknowledged the potential utility of determining EBWs for patients across the diagnostic spectrum, survey data demonstrated that this practice was most often employed when treating patients with AN and AAN. Descriptively, clinicians endorsed determining EBWs for patients with AAN (72.6% youth, 57.3% adult ‘most of the time’ or ‘always’) less frequently than they did AN (77.7% youth, 68.6% adult ‘most of the time’ or ‘always’). It is possible that this discrepancy is a consequence of weight stigma. Indeed, in a qualitative study examining experiences of weight stigma among patients with AAN, multiple participants noted that their EBWs were the same as their admission weights or that clinicians suggested weight loss as a goal (Harrop et al. 2021). In addition to possibly indicating weight stigma, fewer EBWs being set for AAN may suggest uncertainty around determining EBWs for these patients (Nagata et al. 2018). Future work should investigate best approaches for setting individualised EBWs for AAN (Jary et al. 2024).

It is also notable that EBWs were not consistently determined for patients with BN, ARFID, and PD, according to the quantitative data. Although these EDs are frequently accompanied by weight suppression, which has been associated with a host of adverse clinical outcomes (Lowe et al. 2007; Butryn et al. ²⁰¹¹), weight restoration is not as much of a focus for recovery from these EDs compared to AN and AAN, potentially contributing to the lower rates of EBW determination observed. Additionally, while clinicians qualitatively reported setting EBWs across all ED diagnoses, this finding may reflect aspirational practices rather than consistent clinical practice. Diagnosis‐specific barriers may impede the application of EBWs in practice, including lack of consensus around the role of weight in recovery and challenges in determining appropriate targets. For example, in ARFID, early‐onset growth disruption can complicate determination of energy needs and growth trajectories (Tanner and Richmond 2024). In the case of BN, although weight suppression has been linked to greater symptom severity and poorer outcomes, it is not a universal feature of the disorder (Fisher et al. 2025). As a result, EBW determination may be clinically meaningful only for a subset of patients with BN, necessitating individualised assessments of weight history and trajectory. Future research should examine whether, when, and how EBW determination is helpful for patients with BN, ARFID, and PD, and identify diagnosis‐specific barriers that may limit the use of EBWs.

4.4. Strengths and Limitations

This study has several strengths. Its convergent mixed methods design allows for triangulation of clinician perspectives, and the use of nuanced response options (e.g., ‘never’ to ‘always’) captures greater variability in EBW practices than dichotomous items would. To our knowledge, this study is the first to examine EBW determination across multiple ED diagnoses in both youth and adult populations. Limitations include that the sample was predominantly based in the United States, limiting generalisability to international settings. Additionally, this study relied on clinician self‐report, so it's unclear how clinician biases may affect the determination of EBWs in practice. Future research should incorporate case‐based vignette methodologies to assess EBW determination more directly.

4.5. Conclusion

This study offers a comprehensive examination of EBW determination in ED treatment. Findings highlight use of individualised, iterative approaches to EBW determination and reveal key challenges, including limited access to historical weight data, variability in diagnostic application, and limited formal training among providers. To advance equitable and evidence‐informed care, future efforts should focus on standardising individualised EBW‐setting practices across diagnostic presentations, expanding training for providers, and leveraging algorithmic tools that integrate clinical nuance.

Author Contributions

Agatha A. Laboe: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, writing – original draft, writing – review and editing. Molly F. Steinhoff: data curation, formal analysis, investigation, writing – review and editing. Siena S. Vendlinski: data curation, formal analysis, investigation, writing – review and editing. Anna Kreynin: writing – original draft, writing – review and editing. Sarah Johnson‐Munguia: conceptualization, methodology, writing – original draft, writing – review and editing. Ava Heyrman: data curation, formal analysis, investigation, writing – review and editing. . Sophia Kreckler: data curation, formal analysis, investigation, writing – review and editing. Caitlyn Ruud: data curation, formal analysis, investigation, writing – review and editing. Katherine Schaumberg: conceptualization, project administration, investigation, methodology, software, supervision, writing – original draft, writing – review and editing.

Funding

This project was supported by a Psi Chi Graduate Research Grant, awarded to A.A.L. A.A.L. was supported by (TL1TR002375) within the Institute for Clinical and Translational Research at the University of Wisconsin‐Madison, supported by (U54TR002373). K.S. was supported by the National Institute of Mental Health (K01MH123914). M.F.S. was supported by the National Heart, Lung, and Blood Institute (T32HL130357) and National Institute of Mental Health (F31MH137979).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information S1

Laboe, Agatha A. , Steinhoff Molly F., Vendlinski Siena S., et al. 2026. “Clinician Practices in Determining Expected Body Weights Across Eating Disorders: A Mixed Methods Study,” European Eating Disorders Review: 34. no. 3), 584–598. 10.1002/erv.70049.

Data Availability Statement

Data is available upon request. Preprint: https://osf.io/preprints/osf/ca72b_v1?view_only=.