Interrater reliability of the Eating Disorder Examination among postbariatric patients

Some hope for a dimensional assessment? A critical review of psychometric validated (semi-)structured interview to assess eating disorders

OBJECTIVE

Given that eating disorders (EDs) are considered one of the deadliest mental illnesses, the development of appropriate assessment instruments is a necessity. Despite the extensive literature on assessing EDs, there has been a lack of focus on semi-structured interviews. The purpose of this article is to provide a comprehensive review of psychometrically validated semi-structured interviews for EDs.

METHODS

Included studies (N = 24) were required to present a semi-structured interview for EDs that has been validated through a psychometric process. The APA PsycNet, MEDLINE, APA Psycinfo, Pubmed, and Health & Psychosocial Instruments databases were searched. The literature search included publications through May 2024, with no earliest year restriction.

RESULTS

A total of six instruments were identified and reviewed in terms of conceptual design, purpose and content, psychometric characteristics, and strengths and limitations. Three main findings were highlighted: (a) only half of the instruments are up to date; (b) the instruments are based on either a categorical or a mixed categorical-dimensional approach; and (c) the predominance of the categorical approach.

CONCLUSIONS

The results are discussed regarding the conceptual approaches of the instrument to provide clinical and research implications. Despite the many strengths of the instrument, additional psychometric research is needed.

Reliability of the original and brief versions of the Eating Disorder Examination in binge-eating disorder

OBJECTIVE

Psychometric studies evaluating the reliability of eating-disorder assessment among individuals with binge-eating disorder (BED) have been limited. The current study documents the interrater reliability and internal consistency of the Eating Disorder Examination (EDE) interview when administered to adults with Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-5)-defined BED.

METHODS

Participants (N = 56) were adults seeking treatment for BED in the context of clinical trials testing pharmacological and psychological treatments. Doctoral-level, trained, and supervised clinical researchers evaluated eating-disorder psychopathology using the EDE interview and audio-recorded the interview. A second doctoral-level, trained, and supervised clinical researcher, who did not conduct the initial assessment, coded eating-disorder psychopathology using the audio recording.

RESULTS

Agreement among raters on the number of binge-eating episodes was near perfect. There was excellent interrater reliability for nearly all scales of the EDE interview. Agreement among raters for behavioral indicators of loss of control and marked distress regarding binge eating ranged from moderate to perfect. Internal consistency was variable for all scales, ranging from unacceptable to good.

CONCLUSIONS

Our study suggests that the EDE can be administered reliably by multiple interviewers to assess adults with BED. However, internal consistency was mostly subpar. Tests of reliability and other psychometric properties (e.g., validity) in other patient groups such as children with BED are warranted.

An exploratory structural equation modelling (ESEM) of the Eating Disorders Examination Questionnaire (EDE-Q) in bariatric patients

Several studies in clinical and non-clinical populations indicate differences between rationally and empirically derived subscales for the Eating Disorder Examination Questionnaire (EDE-Q), including samples of patients seeking bariatric surgery. This study aimed to use exploratory structural equation modelling (ESEM) to estimate the factor structure of the EDE-Q and assess for the additive value of alternative measurement of eating disorder symptoms. Adolescents and adults completed the EDE-Q and a psychiatric evaluation prior to bariatric surgery. Data from 330 participants were analysed using the original four-factor and modified three-factor structure of the EDE-Q using both confirmatory factor analysis (CFA) and exploratory structural equation modelling (ESEM). Age, ethnicity, and body mass index were examined as covariates in the best fitting model, and model subscales used to generate a predictive model of clinician screened DSM-5 eating disorder diagnoses for criterion validity. A CFA of the four-factor EDE-Q provided poor model fit for a pre-surgical bariatric population, but the three-factor EDE-Q and an ESEM of the four-factor EDE-Q provided excellent model fit. The Eating Concern subscale of the four-factor ESEM model significantly predicted eating disorder diagnosis and was positively correlated with age. Our results suggest the ESEM derived factors of the EDE-Q offered some improvements to the original empirically derived factor structure, as subscale scores based on the original items and cross-loading items yielded an adequate prediction of clinician diagnoses.

Predictors of early weight loss in post-bariatric surgery patients receiving adjunctive behavioural treatments for loss-of-control eating

This study examined baseline patient characteristics as predictors of early weight loss, defined as any weight loss within the first month of treatment, among patients receiving adjunctive behavioural treatments for loss-of-control (LOC) eating about 6 months after bariatric surgery. Participants were 126 patients in a treatment trial for LOC-eating (roughly 6 months postoperatively) categorized by early weight change following 1 month of treatment. Early weight-loss, defined as any weight loss following 1 month of treatment, and weight-gain, defined as any weight gain, groups were compared on sociodemographic and clinical variables assessed using a battery of reliably administered diagnostic and clinical interviews and established self-report measures, and on surgery-related variables (time since surgery, percent total [%TWL], and percent excess weight loss). Most patients (n = 99; 78.6%) lost weight after the first month of adjunctive treatments. Black patients (n = 24; 61.5%) were significantly less likely to achieve early weight loss compared to patients identifying as White (n = 60; 83%) or 'other' (n = 15; 100%) which was not predicted by any other sociodemographic variable. Severity of eating-disorder psychopathology, psychiatric comorbidity, and a broad range of psychosocial measures were not significantly predictive of early weight changes. Duration since surgery and percent weight loss from time of surgery to study enrolment 6-months post-surgery differed by early weight-loss and weight-gain groups. Findings suggest that among post-bariatric surgery patients receiving adjunctive behavioural treatments for LOC-eating, baseline patient characteristics, aside from race and surgery-related variables, do not predict early weight loss.

Psychiatric comorbidity as a prospective predictor of long-term weight and psychosocial outcomes after bariatric surgery

OBJECTIVES

Psychiatric comorbidity is common among the bariatric population although the prognostic significance of psychiatric comorbidity on outcomes is uncertain. This prospective study examined differences in weight and psychosocial functioning outcomes based on lifetime and current (post-surgical) psychiatric comorbidity.

METHODS

Participants were 140 adults in a RCT for loss-of-control (LOC)-eating approximately six months post-bariatric surgery. Two structured interviews were administered: the Eating Disorder Examination-Bariatric Surgery Version (EDE-BSV) to assess LOC-eating and eating-disorder psychopathology, and the Mini International Neuropsychiatric Interview (MINI) to assess lifetime and current (post-surgical) psychiatric disorders. The EDE-BSV and Beck Depression Inventory (BDI-II) were repeated at post-treatment and 24-month follow-ups.

RESULTS

Lifetime (75.7%) and current/post-surgical (25%) psychiatric diagnoses were common. Groups with and without psychiatric comorbidity did not differ significantly in weight loss outcomes at any timepoint but psychiatric comorbidity was associated significantly with greater LOC-eating, eating-disorder psychopathology, and depression.

CONCLUSIONS

Among participants with LOC-eating post-bariatric surgery, lifetime and post-surgical psychiatric comorbidity was not associated with acute or longer-term weight outcomes but predicted poorer psychosocial functioning. Findings challenge prevailing views that psychiatric comorbidity is related to poorer longer-term weight outcomes following bariatric surgery but highlight its clinical significance as it is associated with broad psychosocial difficulties.

Factor structure and measurement invariance of the English- versus Spanish-language Eating Disorder Examination Questionnaire: Brief Form (S-EDE-Q-BF) in Hispanic/Latino/a/x persons seeking bariatric surgery

BACKGROUND

Assessment of eating disorder psychopathology during preoperative psychological evaluations could be facilitated with psychometrically valid measures. One of the most commonly used measures, the Eating Disorder Examination Questionnaire (EDE-Q), is lengthy and has been found to have psychometric limitations. Research has identified a shorter version that has received reliable support across diverse samples but requires further validation for use with patients being evaluated for bariatric surgery.

OBJECTIVES

To cross-validate the factor structure of the EDE-Q: Brief Form (EDE-Q-BF, standalone, nonnested version) with patients being evaluated for bariatric surgery across English- and Spanish-language versions and establish measurement invariance for gender and language.

SETTING

Northeastern hospital in the United States.

METHODS

Participants (n = 618) undergoing evaluations prior to bariatric surgery who identified as Hispanic/Latino/a/x and consented to participate in this research study (which did not influence eligibility for bariatric surgery) completed self-reports. Of the 618 participants, 92 were male and 526 were female, 318 preferred English and were administered English versions of the measures, and 300 preferred Spanish and were administered Spanish versions of the measures.

RESULTS

The 3-factor structure ("Restraint," "Weight/Shape Concerns," and "Body Dissatisfaction") of the EDE-Q-BF fit the data well (χ2 [11] = 18.47; P = .071; root mean square error of approximation [RMSEA] = .033; comparative fit index [CFI] > .99; standardized root mean squared residual [SRMR] = .02). Scaler invariance was met for both gender and language. Correlations with external criteria further supported its validity.

CONCLUSION

The EDE-Q-BF can easily be administered as part of a preoperative psychological assessment battery to screen for eating disorder psychopathology and is valid for Hispanic/Latino/a/x men and women who speak either English or Spanish.

24-Month follow-up of randomized controlled trial of guided-self-help for loss-of-control eating after bariatric surgery

BACKGROUND

Loss-of-control (LOC)-eating postoperatively predicts suboptimal longer-term outcomes following bariatric surgery. This study examined longer-term effects through 24-month follow-ups after completing treatments in a randomized controlled trial testing guided-self-help treatments (cognitive-behavioral therapy [gshCBT] and behavioral weight-loss [gshBWL]) and control (CON) delivered postoperatively for LOC-eating.

METHODS

140 patients with LOC-eating 6 months after bariatric surgery were randomized (5:5:2 ratio) to 3-months of gshCBT (n = 56), gshBWL (n = 60), or CON (n = 24) delivered by trained allied-health clinicians. Independent assessments were performed throughout/after treatments and at 6-, 12-, 18-, and 24-month follow-ups; 83% of patients were assessed at 24-month follow-up.

RESULTS

Intention-to-treat analyses comparing the three groups (gshCBT vs. gshBWL vs. CON) in LOC-eating abstinence at posttreatment (30%, 27%, 38%), 12-month follow-up (34%, 32%, 42%), and 24-month follow-up (45%, 32%, 38%) revealed no significant differences. Mixed-models revealed significantly reduced LOC-eating frequency through posttreatment, no significant changes in LOC-eating frequency during follow-up, and no differences between the three groups. Weight reduced significantly, albeit modestly, through posttreatment but increased significantly and substantially during follow-ups, with no differences between groups.

CONCLUSIONS

Overall, the 12-week scalable guided-self-help treatments did not differ from each other or control, were associated with significantly reduced frequency of LOC-eating and modest weight loss at posttreatment but were followed by significant weight gain during the 24-month follow-up. Weight gain was substantial and nearly universal whereas the frequency of LOC-eating did not change over time (i.e., LOC-eating reductions and abstinence rates were well maintained through 24-moth follow-ups). Patients with postoperative LOC-eating require more intensive adjunctive treatments.

PUBLIC SIGNIFICANCE

Loss-of-control (LOC) eating postoperatively predicts poorer bariatric surgery outcomes and the longer-term effects of postoperative adjunctive postoperative interventions for LOC eating are unknown. In this 24-month follow-up of a controlled study of scalable guided-self-help treatments and a control condition, improvements in LOC-eating frequency, eating-disorder psychopathology, and depression during treatment were maintained well, with no differences between the three groups. Proportion of patients achieving abstinence from LOC-eating at the 24-month follow-up ranged from 38% to 45% across the three groups. In contrast, weight increased significantly during the 24-month follow-ups, with no differences between the three groups. Findings suggest LOC-eating following bariatric surgery might represent a "marker" for a subgroup of patients that are at risk for substantial weight gains over time. LOC eating following bariatric surgery is challenging to treat with low-intensity scalable treatments and may require more intensive specialist treatments.

Interrater reliability and internal consistency of the eating disorder examination in the longitudinal assessment of bariatric surgery study

BACKGROUND

Psychometric studies of eating disorder measures within bariatric surgery populations are limited.

OBJECTIVES

To examine the interrater reliability and internal consistency of the Eating Disorder Examination (EDE) among patients before and after bariatric surgery.

SETTING

Three clinical centers of the Longitudinal Assessment of Bariatric Surgery Research Consortium.

METHODS

The EDE-Bariatric Surgery Version was administered and audio-recorded by trained interviewers before and at annual assessments after bariatric surgery. Approximately 20% of interviews were randomly selected for rating by a second interviewer. Reliability of the original and brief EDE subscales was examined.

RESULTS

Interrater reliability of the EDE subscales ranged from .86-.97 for the original subscales and .83-.95 for brief subscales before surgery, and .90-.98 for the original subscales and .92-.97 for brief subscales after bariatric surgery. Interrater agreement (based on kappa) was almost perfect for overeating and binge-eating behaviors and substantial for loss-of-control eating before surgery. Similar interrater agreements (based on kappa) were observed after surgery for subjective overeating and binge-eating episodes. Internal consistency of the subscale and global scores was variable, ranging from .41-.97.

CONCLUSION

Findings provide support of the interrater reliability of the EDE, albeit with variable internal consistency, before and after bariatric surgery. Despite support for trained raters to reliably assess EDE constructs, variability in internal consistency suggests that further psychometric testing and rigorous scale development of disordered eating may be needed for the bariatric surgery population.

Nonnormative Eating Behaviors and Eating Disorders and Their Associations With Weight Loss and Quality of Life During 6 Years Following Obesity Surgery

IMPORTANCE

Individuals with severe obesity presenting for obesity surgery (OS) frequently show nonnormative eating behaviors (NEBs) and eating disorders (EDs), but the long-term course and prospective associations with weight loss and health-related quality of life (HRQOL) remain unclear.

OBJECTIVE

To examine the prevalence and prospective relevance of presurgical and postsurgical NEBs and EDs according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, diagnosed through clinical interview, for weight loss and HRQOL up to 6 years following OS.

DESIGN, SETTING, AND PARTICIPANTS

In the prospective, multicenter Psychosocial Registry for Obesity Surgery cohort study, patients seeking OS were recruited at 6 OS centers in Germany and assessed at baseline before surgery and at 6 months and 1 to 6 years after surgery. From a consecutive sample of 1040 volunteers with planned OS from March 1, 2012, to December 31, 2020, a total of 748 (71.92%) were included in this study. Across follow-up, 93 of the 748 patients (12.43%) dropped out. Data were analyzed from April to November 2021.

INTERVENTIONS

Laparoscopic Roux-en-Y gastric bypass or sleeve gastrectomy.

MAIN OUTCOMES AND MEASURES

Both NEBs and EDs were identified using the Eating Disorder Examination interview. Main outcomes were the percentage of total body weight loss (%TBWL) and HRQOL (Impact of Weight on Quality of Life-Lite; range, 0-100, with 0 indicating worst and 100 indicating best).

RESULTS

In 748 patients undergoing OS (mean [SD] age, 46.26 [11.44] years; mean [SD] body mass index [calculated as weight in kilograms divided by height in meters squared], 48.38 [8.09]; 513 [68.58%] female), the mean (SD) %TBWL was 26.70% (9.61%), and the mean (SD) HRQOL improvement was 35.41 (20.63) percentage points across follow-up. Both NEBs and EDs were common before surgery, with postsurgical improvements of varying degrees. Whereas NEBs and EDs did not reveal significant prospective associations with %TBWL, loss-of-control eating at follow-up was concurrently associated with lower %TBWL (estimate, -0.09; 95% CI, -0.14 to -0.04). Loss-of-control eating (estimate, -0.10; 95% CI, -0.17 to -0.03 percentage points) and binge-eating disorder of low frequency and/or limited duration (estimate, -6.51; 95% CI, -12.69 to -0.34 percentage points) at follow-up showed significant prospective associations with lower HRQOL.

CONCLUSIONS AND RELEVANCE

This cohort study found prospective relevance of loss-of-control eating and binge-eating disorder of low frequency and/or limited duration for reduced long-term HRQOL following OS. These findings underline the importance of monitoring both NEBs, especially loss-of-control eating, and EDs in the long term postsurgically to identify patients in need of targeted prevention or psychotherapy.

Disordered eating following bariatric surgery: a review of measurement and conceptual considerations

Eating disorders are associated with significant medical morbidity and mortality and serious psychological impairment. Individuals seeking bariatric surgery represent a high-risk group for evidencing disordered eating and eating disorders, with some patients experiencing the persistence or onset of disordered eating postsurgery. This review synthesizes the available literature on problematic or disordered eating in the bariatric field, followed by a review of measurement and conceptual considerations related to the use of eating disorder assessment tools within the bariatric population.