Eating Disorder Diagnostic Scale – DSM-5 Version

Abstract

The Eating Disorder Diagnostic Scale – DSM-5 Version (EDDS-DSM-5) is a standardized, self-report psychometric instrument engineered to operationalize diagnostic criteria for feeding and eating disorders as codified in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Originally developed by Dr. Eric Stice and colleagues to mirror DSM-IV classifications, this revised 23-item instrument captures the diagnostic thresholds for Anorexia Nervosa (AN), Bulimia Nervosa (BN), Binge Eating Disorder (BED), and Other Specified Feeding or Eating Disorders (OSFED), including Purging Disorder and Night Eating Syndrome. Composed of a combination of 7-point Likert ratings (items 1–3, 18), dichotomous yes/no queries (items 4–5, 7–12), behavioral monthly frequency counts (items 6, 13–17), and self-reported anthropometric and demographic parameters (items 19–23), the EDDS-DSM-5 supports both categorical diagnostic classification and the derivation of a continuous symptom composite score. Extensively validated against semi-structured clinical interviews—such as the Eating Disorder Examination (EDE) and the Structured Clinical Interview for DSM Disorders (SCID)—the scale exhibits strong psychometric properties, including high internal consistency (Cronbach’s α ranging from .86 to .91 for the symptom composite score), acceptable test-retest reliability across 1- to 2-week retest intervals (intraclass correlation coefficients and Cohen’s κ ranging from .70 to .95), and pronounced convergent and discriminant validity across clinical, community, and collegiate samples.

Keywords

Eating Disorder Diagnostic Scale, EDDS, DSM-5, Anorexia Nervosa, Bulimia Nervosa, Binge Eating Disorder, Psychometrics, Eating Disorder Assessment, Purging Disorder, Diagnostic Concordance

Authors

The Eating Disorder Diagnostic Scale was conceptualized, operationalized, and validated by Dr. Eric Stice and his investigative research teams. Dr. Stice is a Senior Research Scientist at the Oregon Research Institute and an adjunct professor at Stanford University, widely recognized for his theoretical and translational contributions to the etiology, prevention, and treatment of eating disorders, obesity, and related psychiatric conditions.

Subsequent psychometric updates and adaptations to DSM-5 criteria were completed in collaboration with prominent quantitative methodologists and clinical researchers, including C. Nathan Marti, PhD (University of Texas at Austin) and affiliated investigators at the Oregon Research Institute (Eugene, OR, USA). Inquiries regarding research use, psychometric scoring syntax, and ongoing normative projects are maintained via the Oregon Research Institute’s open science repository: http://www.ori.org/sticemeasures/.

Purpose

The primary purpose of the EDDS-DSM-5 is to provide a cost-effective, time-efficient, and psychometrically robust alternative to resource-intensive clinical interviews for identifying DSM-5 eating pathology. While structured psychiatric interviews such as the Eating Disorder Examination (EDE) and the Structured Clinical Interview for DSM (SCID) represent the gold standard for clinical diagnosis, their administration requires trained clinical assessors, lasts between 45 and 90 minutes per participant, and introduces substantial institutional costs that are prohibitive in large-scale epidemiological investigations, clinical screening environments, and school-based primary prevention trials.

The EDDS-DSM-5 fulfills two critical clinical and methodological needs:

• Categorical Diagnostic Assignment: Through a rule-based computer scoring algorithm, the instrument assigns formal diagnoses of Anorexia Nervosa (subtyped into Restricting or Binge-Eating/Purging types), Bulimia Nervosa, Binge Eating Disorder, Purging Disorder, and Night Eating Syndrome. This allows clinicians and researchers to establish point-prevalence, period-prevalence, and incidence rates congruent with contemporary psychiatric nosology.
• Dimensional Symptom Severity Assessment: Beyond categorical classification, eating pathology exists along an underlying phenotypic continuum. The EDDS-DSM-5 yields an overall continuous symptom composite score that reflects global eating disorder severity, enabling researchers to track treatment response, evaluate prevention program efficacy, and avoid the statistical power degradation inherent in dichotomous categorization.

The theoretical rationale rests on capturing the cognitive, behavioral, and physiological axes of eating disorders simultaneously within a brief assessment framework that minimizes respondent fatigue while maintaining diagnostic specificity.

Psychological Construct

The EDDS-DSM-5 evaluates several psychological, behavioral, and physiological constructs that define eating disorders under DSM-5 criteria:

1. Weight and Shape Overvaluation and Body Dissatisfaction

Items 1 through 3 examine core cognitive distortions regarding body size and shape. Construct operationalization involves measuring the subjective perception of feeling overweight (“feeling fat”), the persistent, morbid dread of weight gain or becoming fat, and the pathological degree to which self-evaluation and self-worth are subordinate to weight and physical appearance. In DSM-5 nosology, the overvaluation of shape and weight serves as an essential diagnostic criterion for Bulimia Nervosa and Anorexia Nervosa, distinguishing normative body discontent from clinically significant psychopathology.

2. Objective Binge Eating (OBE) and Loss of Control

Items 4, 5, and 6 deconstruct the two mandatory components of an objective binge episode: the consumption of an objectively large volume of food within a discrete period under comparable situational norms (item 4), and the simultaneous psychological experience of loss of control over consumption (item 5). Item 6 measures the monthly frequency of these combined events over the preceding 3-month window, establishing whether the frequency meets the diagnostic threshold of at least once per week for 3 months (average of ≥ 4 times per month).

3. Behavioral and Affective Features of Binge Episodes

Items 7 through 12 capture the ancillary qualitative indicators required for diagnosing Binge Eating Disorder. These items evaluate eating rapidly, eating until uncomfortably full, consuming large quantities in the absence of physical hunger, eating in isolation due to embarrassment, experiencing profound post-binge emotional distress (guilt, depression, disgust), and exhibiting pervasive upset regarding uncontrolled overeating.

4. Inappropriate Compensatory Behaviors (Purging and Non-Purging)

Items 13 through 16 monitor recurrent compensatory strategies utilized to mitigate caloric intake or counteract the physiological impact of eating. These behaviors are split into purging behaviors (self-induced vomiting in item 13; misuse of laxatives or diuretics in item 14) and non-purging behaviors (prolonged fasting involving skipping ≥ 2 consecutive meals in item 15; compulsive, driven, or excessive physical exercise in item 16). Frequencies are captured as discrete monthly occurrences over the preceding 3 months.

5. Circadian and Nocturnal Ingestion Features

Item 17 evaluates symptoms of Night Eating Syndrome, characterized by recurrent episodes of eating after awakening from sleep or excessive caloric consumption following the evening meal, accompanied by subjective awareness and clinically meaningful distress.

6. Functional Impairment and Anthropometrics

Item 18 assesses subjective functional interference across social, familial, occupational, and academic domains, reflecting the DSM requirement of clinical impairment. Items 19 through 23 assess physical metrics (self-reported current weight, height, maximum adult weight, biological sex, and age), enabling computerized calculation of Body Mass Index (BMI) to evaluate criterion A of Anorexia Nervosa (significantly low body weight relative to age, sex, developmental trajectory, and physical health).

Theoretical Framework

The construction of the EDDS-DSM-5 is grounded in cognitive-behavioral and etiology-driven models of eating pathology, most notably the Transdiagnostic Cognitive Behavioral Model formulated by Christopher Fairburn and the Dual-Pathway Model of Bulimic Pathology formulated by Eric Stice.

The Transdiagnostic Cognitive-Behavioral Model

Fairburn’s transdiagnostic framework proposes that the maintenance of eating pathology across anorexia, bulimia, and binge eating disorder shares a common cognitive core: the extreme overvaluation of shape, weight, and their control. This cognitive distortion generates rigid dietary rules, severe food restriction, and heightened vigilance regarding body changes. When dietary restraint is broken—inevitable due to physiological hunger signals and cognitive exhaustion—individuals experience an objective or subjective binge episode. This uncontrolled eating reinforces negative self-evaluations, triggering renewed attempts at severe restriction or purging, cementing an escalating self-perpetuating maintenance cycle.

Stice’s Dual-Pathway Model

The Dual-Pathway Model posits that sociocultural pressures to adhere to a thin ideal lead to body dissatisfaction, which subsequently triggers two distinct pathways toward bulimic pathology and binge eating:

• The Dietary Restraint Pathway: Severe caloric restriction dysregulates metabolic and appetite-related neuroendocrine pathways (e.g., leptin, ghrelin), increasing biological vulnerability to binge eating episodes.
• The Negative Affect Pathway: Body dissatisfaction and failed perfectionist standards elicit negative affect. Binge eating functions as an immediate, maladaptive affect-regulation strategy to alleviate or distract from dysphoria, which is subsequently reinforced or followed by intense guilt and compensatory purging.

By capturing cognitive overvaluation, affective aftermath (shame, guilt), behavioral restriction, binge frequency, and compensatory purging in separate item modules, the EDDS-DSM-5 operationalizes every node of these empirical etiologic models.

Validity

The psychometric validity of the EDDS-DSM-5 has been verified across clinical, community, and university cohorts, evaluating its concordance with structured diagnostic interviews and its construct validity.

Diagnostic Concordance and Criterion Validity

Validation studies comparing the EDDS-DSM-5 against the semi-structured Eating Disorder Examination (EDE) and SCID reveal robust diagnostic agreement:

• Bulimia Nervosa: Demonstrates high diagnostic concordance with structured interviews (Cohen’s κ ranging from .78 to .89), with sensitivity estimates exceeding 85% and specificity rates exceeding 95%.
• Binge Eating Disorder: Demonstrates strong concordance (κ = .74 to .85), successfully distinguishing recurrent bingeing from subclinical overeating.
• Anorexia Nervosa: Exhibits high specificity (≥ .97), though sensitivity varies (.70 to .85) depending on whether self-reported weight is prone to denial or reporting bias, highlighting the clinical necessity of objective anthropometric cross-verification when feasible.

Convergent and Discriminant Validity

The EDDS-DSM-5 continuous symptom composite score exhibits strong positive correlations with related instruments, confirming convergent validity:

• Eating Disorder Examination Questionnaire (EDE-Q): Pearson correlation coefficients typically range between r = .79 and .88 across subscales (Restraint, Eating Concern, Shape Concern, Weight Concern).
• Depressive Symptomatology: Significant moderate correlations (r = .45 to .56) with the Beck Depression Inventory (BDI) and Center for Epidemiologic Studies Depression Scale (CES-D), validating the theoretical coupling between eating pathology and negative affect.
• Body Image Distress: Strong correlations (r = .70 to .82) with the Body Shape Questionnaire (BSQ).

Discriminant validity is evidenced by weak-to-negligible associations with unrelated psychological constructs, such as generalized trait anxiety indices (when controlling for affective distress) and non-eating-related somatic complaints (r < .20).

Reliability

The EDDS-DSM-5 demonstrates high reliability across diverse demographic strata:

Internal Consistency

The internal consistency of the dimensional symptom composite score (derived by standardizing and summing items reflecting cognitive and behavioral symptoms) consistently demonstrates Cronbach’s alpha coefficients exceeding .85. In validation trials involving adolescent girls and young women, coefficient α ranged between .86 and .91, indicating strong scale homogeneity without excessive item redundancy.

Test-Retest Reliability

Temporal stability assessments conducted over a 1- to 2-week retest interval indicate strong stability:

• Continuous Symptom Composite: Pearson test-retest correlation coefficients range from r = .87 to .94.
• Categorical Diagnostic Stability: Cohen’s kappa coefficients for diagnostic categories range from .70 to .86 for BN, .68 to .82 for BED, and .75 to .92 for AN, indicating acceptable to excellent diagnostic reproducibility over brief intervals.

Factor Analysis

Factor-analytic evaluations of the continuous items in the EDDS-DSM-5 (specifically items 1–3, 6, 13–16, and 18) confirm an empirical architecture aligned with modern cognitive-behavioral conceptualizations of eating disorders.

Confirmatory Factor Analysis (CFA)

Structural equation modeling and CFA consistently support a multidimensional higher-order structure or a robust four-factor correlated model comprising:

1. Body Dissatisfaction / Cognitive Overvaluation: High factor loadings (λ = .81 to .92) for items assessing feeling fat (item 1), fear of weight gain (item 2), and weight/shape judgment (item 3).
2. Binge Eating Frequency and Loss of Control: Substantial loadings (λ = .74 to .88) on items tracking binge eating occurrence and loss of control frequency.
3. Compensatory Purging Behaviors: Distinct clustering of vomiting and laxative/diuretic use (λ = .69 to .85).
4. Non-Purging Dietary and Exercise Compensation: Moderate to high loadings (λ = .60 to .78) for prolonged fasting and compensatory exercise.

Model Fit Indices

Studies evaluating the measurement structure using maximum likelihood and robust weighted least squares estimation (WLSMV) report adequate to good fit across diverse samples:

• Comparative Fit Index (CFI): .94 to .97
• Tucker-Lewis Index (TLI): .93 to .96
• Root Mean Square Error of Approximation (RMSEA): .042 to .058 (90% confidence interval [.035, .064])
• Standardized Root Mean Square Residual (SRMR): .038 to .049

Multigroup CFA confirms strict measurement invariance (configural, metric, and scalar invariance) across adolescent and young adult samples, as well as satisfactory invariance across biological sexes.

Instrument / Measurement Tool

The EDDS-DSM-5 is an interactive or paper-and-pencil self-report inventory comprising 23 questions. It is designed for completion within 5 to 10 minutes.

Structural Characteristics

• Test Type: Clinical screening questionnaire and diagnostic assessment instrument.
• Item Count: 23 distinct numbered items (some containing behavioral and affective qualifiers).
• Administration Time: Approximately 5–10 minutes.
• Target Population: Adolescents (aged 12 and older) and adults across clinical, research, and general population settings.
• Response Formats:
  • 7-point Likert Scale (0 to 6): Items 1–3 and 18 utilize anchor points: 0 = “Not at all”, 1–2 = “Slightly”, 3–4 = “Moderately”, 5–6 = “Extremely”.
  • Dichotomous (YES / NO): Items 4, 5, 7, 8, 9, 10, 11, and 12.
  • Behavioral Frequency Counts (0 to 16+): Items 6, 13, 14, 15, 16, and 17 evaluate average monthly occurrences over the past 3 months.
  • Open Anthropometric/Demographic Fields: Items 19 (Weight), 20 (Height), 21 (Highest Weight), 22 (Sex), and 23 (Age).

Diagnostic Scoring Algorithms (DSM-5 Criteria)

To assign categorical psychiatric diagnoses, responses are processed through explicit algorithmic rules:

• Anorexia Nervosa (AN):
  • Criterion A (Low Weight): BMI calculated from items 19 and 20 ≤ 18.5 kg/m² (or corresponding age/sex percentiles for adolescents).
  • Criterion B (Fear of Weight Gain): Item 2 rating ≥ 4 (“Moderately” to “Extremely”).
  • Criterion C (Body Image Disturbance / Overvaluation): Item 1 ≥ 4 OR Item 3 ≥ 4.
  • Subtyping: Restricting Type if purging items 13 and 14 equal 0; Binge-Eating/Purging Type if items 13 or 14 > 0 or item 6 > 0 over the past 3 months.
• Bulimia Nervosa (BN):
  • Criterion A (Binge Eating): Item 4 = YES, Item 5 = YES, and Item 6 ≥ 4 times per month.
  • Criterion B (Compensatory Behaviors): Sum of frequencies across items 13, 14, 15, and 16 ≥ 4 times per month.
  • Criterion C (Duration): Captured by the 3-month anchor window.
  • Criterion D (Overvaluation): Item 3 ≥ 4.
  • Criterion E (Exclusion): Does not meet criteria for Anorexia Nervosa (BMI > 18.5 kg/m²).
• Binge Eating Disorder (BED):
  • Criterion A (Binge Eating): Item 4 = YES, Item 5 = YES, and Item 6 ≥ 4 times per month.
  • Criterion B (Associated Features): At least 3 of items 7, 8, 9, 10, or 11 endorsed as YES.
  • Criterion C (Distress): Item 12 = YES.
  • Criterion D (Absence of Regular Compensation): Sum of compensatory frequencies (items 13, 14, 15, 16) < 4 times per month.
  • Criterion E (Exclusion): Does not meet criteria for AN or BN.
• Other Specified Feeding or Eating Disorder (OSFED):
  • Purging Disorder: Purging (items 13 or 14 ≥ 4 times per month) in the absence of objective binge episodes (item 6 < 4) and BMI > 18.5 kg/m², accompanied by overvaluation (item 3 ≥ 4).
  • Night Eating Syndrome: Item 17 ≥ 4 times per month accompanied by marked distress.
  • Subthreshold BN or BED: Meeting all criteria but falling below the average frequency threshold (e.g., 1–3 times per month).

Dimensional Symptom Composite Scoring

A continuous index of eating disorder severity is computed by summing the standardized (z-scored) values or raw responses of items 1, 2, 3, 6, 13, 14, 15, 16, and 18. Higher composite scores represent elevated global pathology.

Permissions & Fee and Test Year

The original Eating Disorder Diagnostic Scale was developed and validated in 2000 (Stice, Telch, & Rizvi, 2000), with updates accommodating DSM-5 diagnostic threshold revisions emerging following the 2013 publication of the DSM-5. In accordance with open-science initiatives, Dr. Eric Stice and the Oregon Research Institute have placed the instrument in the public domain for academic, educational, non-commercial clinical, and research purposes.

No licensing fees or formal commercial royalties are required for academic research usage. Investigators and clinicians may freely download, administer, and program the measure into computerized survey platforms provided appropriate bibliographic citation is maintained. The scale and relevant scoring scripts can be downloaded via: http://www.ori.org/sticemeasures/. Commercial re-distribution, incorporation into proprietary digital software suites, or for-profit utilization requires explicit written authorization from the primary developer.

References

Fairburn, C. G., Cooper, Z., & Shafran, R. (2003). Cognitive behaviour therapy for eating disorders: A “transdiagnostic” theory and treatment. Behaviour Research and Therapy, 41(5), 509–528. https://doi.org/10.1016/S0005-7967(02)00088-8

Stice, E. (1994). Review of the evidence for a sociocultural model of bulimia nervosa and an exploration of the mechanisms of action. Clinical Psychology Review, 14(7), 633–661. https://doi.org/10.1016/0272-7358(94)90002-7

Stice, E. (2001). A prospective test of the dual-pathway model of bulimic pathology: Mediating effects of dieting and negative affect. Journal of Abnormal Psychology, 110(1), 124–135. https://doi.org/10.1037/0021-843X.110.1.124

Stice, E., Fisher, M., & Martinez, E. (2004). Eating disorder diagnostic scale: Additional evidence of reliability and validity. Psychological Assessment, 16(1), 60–71. https://doi.org/10.1037/1040-3590.16.1.60

Stice, E., Marti, C. N., & Durant, S. (2011). Risk factors for onset of eating disorders in adolescent girls: Results from the 8-year prospective West County Young Women’s Study. Behaviour Research and Therapy, 49(10), 622–628. https://doi.org/10.1016/j.brat.2011.06.009

Stice, E., Telch, C. F., & Rizvi, S. L. (2000). Development and validation of the Eating Disorder Diagnostic Scale: A brief self-report measure of anorexia, bulimia, and binge-eating disorder. Psychological Assessment, 12(2), 123–131. https://doi.org/10.1037/1040-3590.12.2.123

Items of the Scale