Disordered Eating Attitude Scale (DEAS)

Abstract

The Disordered Eating Attitude Scale (DEAS) is an established psychometric questionnaire developed by Marle dos Santos Alvarenga, Fernanda Baeza Scagliusi, and Sonia Tucunduva Philippi (2010) to comprehensively evaluate the cognitive, emotional, and behavioral dimensions that characterize an individual’s orientation toward food and nourishment. While traditional psychometric inventories—such as the Eating Attitudes Test (EAT-26) and the Eating Disorder Inventory (EDI)—were engineered primarily to screen for manifest psychiatric symptoms of clinical eating disorders (such as anorexia nervosa and bulimia nervosa), the DEAS fills a critical diagnostic void by operationalizing maladaptive eating attitudes across both non-clinical populations and clinical cohorts. Eating attitudes encompass an individual’s deeply held beliefs, food-related thoughts, affect during and anticipating meals, moralistic food categorization, and aberrant behavioral repertoires (including subtle dietary restraint and post-prandial compensatory mechanisms).

Comprising 25 items divided into two operational sections, the scale assesses five empirically robust latent factors: (1) Relationship with Food, (2) Concerns about Food and Weight Gain, (3) Restrictive and Compensatory Practices, (4) Feelings Toward Eating, and (5) Idea of Normal Eating. The instrument employs a mixed-response metric in Part I (dichotomous items, conditional Likert scales, and multidimensional categorical food evaluations) and a uniform 5-point Likert scale (ranging from 1 = “Rarely/Never” to 5 = “Always”) across Part II. Total scores span a theoretical continuum of 37 to 190 points, where elevated values index more dysfunctional, distorted, and pathological eating attitudes. The instrument has undergone extensive cross-cultural adaptation and psychometric validation across multiple linguistic contexts, including Brazilian Portuguese, English, Spanish, Japanese, and Turkish. Across validation cohorts, the DEAS demonstrates adequate to high internal consistency (Cronbach’s alpha coefficients commonly between .75 and .83 for the overall scale), solid test-retest reliability, well-substantiated construct validity, and proven sensitivity in discriminating clinical eating disorder patients from non-clinical community controls.

Keywords

Disordered Eating Attitude Scale, DEAS, eating attitudes, psychometrics, dietary restraint, eating behavior, restrictive eating, compensatory practices, food guilt, factor analysis

Authors

The Disordered Eating Attitude Scale was conceived, developed, and validated by a distinguished team of nutritional and public health psychometricians affiliated with the University of São Paulo (Universidade de São Paulo – USP), School of Public Health (Faculdade de Saúde Pública), São Paulo, Brazil:

• Marle dos Santos Alvarenga, PhD, RD: Senior Researcher and Professor affiliated with the Department of Nutrition, School of Public Health, University of São Paulo, and supervisor at the Eating Disorders Program (AMBULIM), Institute of Psychiatry, Hospital das Clínicas, University of São Paulo. Her research focuses on eating behavior, body image, eating disorder psychopathology, and intuitive eating.
• Fernanda Baeza Scagliusi, PhD: Professor in the Department of Nutrition at the School of Public Health, University of São Paulo. Her scientific scholarship encompasses qualitative and quantitative methods in nutritional epidemiology, sociology of food, and eating behavior measurement.
• Sonia Tucunduva Philippi, PhD, RD: Full Professor in the Department of Nutrition, School of Public Health, University of São Paulo. An authority on dietary guidelines, food pyramids, nutritional planning, and population-level dietary assessment methodology.

Secondary validation studies of international linguistic adaptations featured key scientific collaborations with scholars such as Roberta F. Pereira, Camila P. C. Estima, Jillian Croll (English adaptation), and Raquel Francischi (Spanish adaptation).

Purpose

The core objective of the Disordered Eating Attitude Scale is to identify, quantify, and track distorted beliefs, emotional conflicts, and maladaptive behaviors related to food intake that do not necessarily meet the threshold for full-syndrome diagnostic criteria under the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR). In both public health epidemiology and specialized clinical nutrition, a major conceptual challenge has been the tendency of traditional assessment tools to evaluate only extreme, acute behavioral symptoms (e.g., objective binge episodes, self-induced emesis, extreme laxative misuse). Consequently, subclinical dietary pathology—such as intense food-related moralization, cognitive preoccupation, persistent guilt, chronic caloric calculation, and functional impairment—frequently escapes detection.

The DEAS addresses this limitation by functioning as a sensitive measurement tool capable of evaluating the broader spectrum of eating pathology. In clinical settings, the scale serves several key functions:

• Baseline Clinical Assessment: It establishes a baseline measure of a patient’s psychological relationship with food prior to medical, psychotherapeutic, or nutritional intervention.
• Monitoring Cognitive Trajectories: In patients recovering from anorexia nervosa, bulimia nervosa, or binge eating disorder, behavioral stabilization (such as weight restoration or cessation of purging) often precedes cognitive recovery. The DEAS helps clinicians identify persistent cognitive distortions that signal ongoing relapse risk.
• Differentiating Nutritional Beliefs: The instrument helps distinguish evidence-based nutritional knowledge from rigid, pathological dietary dogmatism driven by fear of weight gain.

In epidemiological and non-clinical research, the DEAS provides investigators with a validated scale to examine the prevalence of disordered eating patterns within vulnerable populations, such as university students, adolescent cohorts, athletes, fitness practitioners, and individuals seeking commercial weight-loss treatments. Furthermore, the instrument evaluates the cognitive mechanisms underlying modern eating trends, such as orthorexic tendencies and the functional dichotomization of foods into “safe” versus “forbidden” categories.

Psychological Construct

The construct assessed by the DEAS is disordered eating attitudes. In contemporary psychometrics and nutritional psychology, an eating attitude is operationalized as the comprehensive constellation of cognitive beliefs, affective evaluations, and behavioral tendencies directed toward food, eating, and bodily nourishment. When these attitudes become disordered, food loses its normative function as a source of sustenance, pleasure, and social connection. Instead, it becomes an emotional battleground characterized by anxiety, cognitive control, moral judgment, and compensatory rituals.

The DEAS measures this multi-faceted construct across five validated domains:

1. Relationship with Food

This central dimension captures the general disruption of an individual’s psychological and emotional relationship with food. It assesses the presence of internal conflict, loss of behavioral control, social isolation caused by eating habits, and the intrusion of food-related thoughts into everyday life. High scores reflect an adversarial relationship with eating, where food is perceived as a threat or an obsession that interferes with social and occupational functioning (e.g., feeling that one’s relationship with food disrupts daily life, or dreaming of a pill that could eliminate the physiological need to eat).

2. Concerns About Food and Weight Gain

This subscale assesses the cognitive burden associated with anticipating caloric intake and weight gain. It captures pervasive anxiety regarding the metabolic consequences of eating, chronic caloric calculation, and the mental effort spent deciding what, when, and how much to eat. Elevated scores reflect cognitive hyper-vigilance, where every meal is scrutinized for its potential to alter body shape or weight.

3. Restrictive and Compensatory Practices

This behavioral domain operationalizes both proactive and reactive strategies designed to manipulate energy balance. It measures intentional meal skipping, sustained fasting, radical caloric restriction, and compensatory behaviors (such as intense post-meal exercise or the use of laxatives, diuretics, and self-induced vomiting) triggered by perceived overeating. The construct reflects an all-or-nothing approach to self-regulation, alternating between rigid restriction and severe post-prandial distress.

4. Feelings Toward Eating

This dimension focuses on the affective responses evoked by the act of eating. In healthy eating patterns, food intake is accompanied by sensory pleasure, satisfaction, and biological satiety. In contrast, disordered eating is characterized by negative emotional conditioning, including intense guilt, feelings of emotional impurity or “dirtiness,” and the perception that eating is unnatural. This subscale measures the degree to which an individual experiences visceral negative affect during and after eating.

5. Idea of Normal Eating

This factor evaluates an individual’s cognitive schema regarding what constitutes appropriate, healthy, and permissible eating behavior. It explores the cognitive dichotomization of foods into rigid “good” versus “bad” categories, the belief that major food groups (such as carbohydrates, fats, or meats) must be entirely eliminated, and attitudes toward non-homeostatic eating (such as eating in response to emotional cues like sadness, boredom, or celebration). Elevated scores indicate rigid, black-and-white dietary thinking, whereas low scores reflect flexibility and an understanding of eating as an adaptable human experience.

Theoretical Framework

The conceptual architecture of the DEAS draws upon several foundational models in clinical psychology, cognitive-behavioral theory, and nutritional science:

1. The Cognitive-Behavioral Transdiagnostic Model

The DEAS is heavily informed by Christopher G. Fairburn‘s transdiagnostic cognitive-behavioral model of eating disorders. According to Fairburn, the core cognitive pathology underpinning disordered eating is the over-evaluation of eating, shape, and weight. While primary diagnostic tools traditionally focus on the shape and weight dimensions, the DEAS expands on the “control over eating” component. Under this framework, individuals equate self-worth with their ability to exert rigid control over dietary intake. Any perceived lapse in control triggers catastrophic thoughts of weight gain and failure, leading to profound guilt, emotional distress, and secondary compensatory behaviors.

2. Dietary Restraint Theory

The scale integrates key principles from Herman and Polivy’s Restraint Theory. Restraint theory posits that chronic, cognitively mediated efforts to suppress food intake create psychological and biological vulnerabilities. When dietary restraint replaces physiological hunger and satiety cues with rigid cognitive rules, individuals become susceptible to disinhibition (the “what-the-hell effect”). The DEAS captures these dynamics across items assessing meal skipping, calorie counting, and cognitive responses to perceived dietary transgressions.

3. The Dichotomous Food Moralization Hypothesis

The instrument draws upon social and cognitive theories that examine the moralization of health behaviors in modern societies. Influenced by public health messaging and social pressures, individuals often categorize foods into moral binaries: foods are labeled as “clean,” “pure,” and “good,” or conversely, “toxic,” “fattening,” and “bad.” The DEAS explicitly evaluates this cognitive distortion in Part I, where respondents assess whether completely avoiding essential food groups is necessary for health. This taps into the cognitive rigidity that underpins restrictive eating and orthorexic tendencies.

4. The Intuitive Eating and Competent Eating Paradigm

The theoretical framework of the DEAS is also shaped by models of healthy eating competence, such as Ellyn Satter’s Eating Competence Model and Tribole and Resch’s Intuitive Eating paradigm. These frameworks suggest that healthy eating is marked by internal regulation, sensory pleasure, cognitive flexibility, and freedom from guilt. By assessing constructs such as the capacity to feel pleasure from eating, positive food memories, and an acceptance of eating in response to emotional states, the DEAS conceptualizes healthy eating not merely as the absence of pathology, but as a flexible, positive relationship with food.

Validity

The psychometric properties of the DEAS have been evaluated across several validation studies involving diverse clinical and non-clinical samples.

Construct and Structural Validity

In the foundational validation study by Alvarenga, Scagliusi, and Philippi (2010), the instrument was administered to a normative cohort of female Brazilian university students ($N = 2,290$). Initial exploratory factor analyses identified a clean five-factor latent structure that accounted for a substantial portion of the total variance, confirming that the hypothesized conceptual domains (relationship with food, concerns about weight, restrictive practices, feelings toward eating, and normal eating concepts) formed distinct yet related dimensions.

Known-Groups and Discriminant Validity

To establish clinical discriminant validity, Alvarenga et al. (2010) administered the DEAS to a clinical cohort of female patients diagnosed with anorexia nervosa, bulimia nervosa, or eating disorders not otherwise specified (EDNOS) ($n = 61$) and compared their scores to the non-clinical university cohort ($n = 2,290$). The clinical cohort exhibited significantly higher mean scores across all five subscales and on the total scale score ($p < .001$), confirming the instrument’s capacity to discriminate between clinical eating pathology and normative eating attitudes. ROC curve analyses have supported the scale’s diagnostic utility in screening for pathological eating concerns.

Convergent and Concurrent Validity

Convergent validity has been evaluated by examining correlations between the DEAS and established psychometric benchmarks:

• Eating Attitudes Test (EAT-26): Total DEAS scores correlate strongly and positively with the EAT-26 ($r = .65$ to $.73, p < .001$), indicating that the DEAS captures core eating pathology while providing broader coverage of general food-related attitudes.
• Restraint Scale (RS): Moderate-to-high correlations have been observed with the Herman and Polivy Restraint Scale ($r = .55$ to $.68$), confirming the DEAS’s alignment with dietary restraint constructs.
• Bulimic Investigatory Test, Edinburgh (BITE): Significant positive associations ($r = .50$ to $.62$) have been demonstrated with measures of bulimic symptomatology, particularly for the Restrictive and Compensatory Practices subscale.
• Body Shape Questionnaire (BSQ): Strong positive correlations ($r = .58$ to $.66$) with body dissatisfaction inventories demonstrate the link between weight-related body image distress and disordered eating attitudes.

Cross-Cultural and Linguistic Validity

The DEAS has demonstrated cross-cultural validity in several translated editions:

• English Version: Validated by Alvarenga, Pereira, Scagliusi, Philippi, Estima, and Croll (2010) in an American female college cohort ($N = 439$), the English version replicated the core factor structure and displayed strong convergent validity with the EAT-26 and the Multidimensional Body-Self Relations Questionnaire (MBSRQ).
• Spanish Version: Evaluated by Alvarenga, Francischi, Fontes, Scagliusi, and Philippi (2010), the Spanish translation exhibited factor congruence and internal consistency comparable to the original Portuguese version.
• Cross-National Invariance: Subsequent cross-national investigations involving cohorts in Japan, Argentina, and Turkey have supported the metric and structural invariance of the DEAS across diverse cultural contexts.

Reliability

The reliability of the DEAS has been supported through evaluations of internal consistency and temporal stability.

Internal Consistency

In the original validation study (Alvarenga et al., 2010), the overall scale achieved a Cronbach’s alpha of $\alpha = .75$ to $.82$ across study phases, reflecting acceptable to good internal consistency for an instrument measuring multi-faceted psychological constructs. Subscale-level internal consistency estimates vary depending on item count and conceptual breadth:

• Subscale 1 (Relationship with food; 12 items): $\alpha = .75 – .81$
• Subscale 2 (Concerns about food and weight gain; 4 items): $\alpha = .71 – .78$
• Subscale 3 (Restrictive and compensatory practices; 4 items): $\alpha = .65 – .73$
• Subscale 4 (Feelings toward eating; 3 items): $\alpha = .60 – .68$
• Subscale 5 (Idea of normal eating; 2 composite sections): $\alpha = .58 – .66$

The comparatively lower alpha values observed for Subscales 4 and 5 are common in psychometrics for subscales comprising very few items (three or fewer), where coefficient alpha is mathematically constrained by scale length.

Test-Retest Stability

Temporal stability was established by administering the scale twice to non-clinical participants across a three- to four-week interval under neutral conditions. The intra-class correlation coefficient (ICC) and Pearson correlation coefficients for the total DEAS score typically range between $r = .80$ and $.86$ ($p < .001$), demonstrating high test-retest reliability over time. Subscale test-retest correlations consistently exceed $r = .70$, confirming that the instrument assesses enduring cognitive-attitudinal traits rather than transient emotional states.

Factor Analysis

The structural composition of the DEAS was established using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

During scale construction, an initial item pool generated through comprehensive clinical reviews and literature analysis was administered to an initial cohort of female university students ($n = 432$). Principal Axis Factoring (PAF) and Principal Component Analysis (PCA) were conducted alongside oblique (Promax/Oblimin) rotations to accommodate anticipated correlations between psychological dimensions of eating behavior. The analysis confirmed the retention of 25 items across five distinct factors based on Kaiser’s criterion (eigenvalues > 1.0), Cattell’s scree plot inflection points, and conceptual interpretability. Items demonstrating weak factor loadings (< .30) or uninterpretable cross-loadings were systematically eliminated.

Confirmatory Factor Analysis (CFA) and Goodness-of-Fit

In subsequent validation and cross-cultural studies, CFA was applied to evaluate the structural integrity of the 5-factor model. Across major evaluations, the 5-factor model has exhibited acceptable to good fit indices:

• Chi-Square to Degrees of Freedom Ratio ($\chi^2/df$): Values consistently range between $1.8$ and $2.8$, falling within the conventional threshold (< 3.0) for acceptable model fit.
• Comparative Fit Index (CFI): Estimates consistently range from $.90$ to $.94$, indicating acceptable to good comparative structural fit.
• Tucker-Lewis Index (TLI): Coefficients typically fall between $.89$ and $.93$.
• Root Mean Square Error of Approximation (RMSEA): Values consistently range from $.042$ to $.058$ (with 90% confidence intervals bounded below $.065$), well within the accepted cutoff for good approximation fit (< .06 to .08).
• Standardized Root Mean Square Residual (SRMR): Estimated between $.048$ and $.060$.

Item factor loadings within designated factors are generally robust, ranging between $.35$ and $.78$, with primary cognitive and behavioral items (such as feelings of guilt, constant preoccupation with food preparation, and meal skipping) exhibiting the highest standardized loadings.

Instrument / Measurement Tool

The Disordered Eating Attitude Scale (DEAS) is a standardized self-report psychometric inventory designed for individual or group administration. The operational characteristics of the measurement tool are structured as follows:

• Administration Format: Standardized self-administered questionnaire (paper-and-pencil or secure digital survey platforms).
• Target Population: Adolescents and adults (typically validated for ages 14 to 65+ across both clinical and non-clinical cohorts).
• Administration Time: Approximately 8 to 12 minutes to complete.
• Total Item Count: 25 items, organized into two distinct structural parts: Part I (Items 1 to 12) and Part II (Items 13 to 25).
• Response and Scoring Rules:
  • Part I (Items 1 to 12): Employs diverse item structures. Item 1 contains multiple food items subdivided into three categories (a: sugar, French fries, oil; b: breads, rice, beans, pasta, red meat, whole milk, cheese; c: vegetables, fruits, white meat), scored conditionally based on perceived dietary necessity (1, 3, or 5 points). Items 2 through 10 are dichotomous (Yes/No), scored as either 1 or 5 points (with reverse scoring applied to items reflecting healthy attitudes, such as Items 2 and 9). Item 11 is a multiple-choice item scored conditionally (1 or 5 points). Item 12 provides four behavioral options scored 1, 3, or 5 points depending on the severity of the compensatory reaction.
  • Part II (Items 13 to 25): Employs a uniform 5-point Likert scale: Rarely/Never = 1 point; Sometimes = 2 points; Often = 3 points; Usually = 4 points; Always = 5 points.
• Total Score Range: The minimum possible score is 37 points, and the maximum possible score is 190 points. Higher scores indicate more severe disordered eating attitudes.
• Subscale Structural Allocation:
  • Subscale 1: Relationship with Food: Items 8, 10, 13, 17, 18, 19, 20, 21, 22, 23, 24, and 25.
  • Subscale 2: Concerns About Food and Weight Gain: Items 5, 14, 15, and 16.
  • Subscale 3: Restrictive and Compensatory Practices: Items 4, 6, 7, and 12.
  • Subscale 4: Feelings Toward Eating: Items 2, 3, and 9.
  • Subscale 5: Idea of Normal Eating: Items 1 (sub-items a, b, c) and 11.
• Interpretative Guidelines: While the DEAS is primarily a continuous measurement scale, higher scores reflect increasing cognitive distortion and behavioral disruption. In research, scores are typically analyzed continuously or divided into population quartiles/tertiles. In clinical contexts, scores approaching or exceeding published clinical sample means (typically > 80–90 points) warrant comprehensive multidisciplinary diagnostic evaluation.

Permissions & Fee and Test Year

The Disordered Eating Attitude Scale was developed and published in 2010. The scale was established as an open-access scientific instrument intended for academic, epidemiological, and non-commercial clinical use. No licensing fees or royalties are required to administer the DEAS in scientific research or public health investigations.

Researchers and clinicians wishing to employ the scale must adhere to standard academic protocol by maintaining the original item wording, scoring systems, and citing the primary validation papers (Alvarenga et al., 2010, Perceptual and Motor Skills; Alvarenga et al., 2010, Appetite). Inquiries regarding formal cross-cultural translation, commercial health-technology integration, or adapted clinical uses should be directed to the primary author, Dr. Marle dos Santos Alvarenga, through the School of Public Health at the University of São Paulo, Brazil.

References

• Alvarenga, M. S., Scagliusi, F. B., & Philippi, S. T. (2010). Development and validity of the Disordered Eating Attitude Scale. Perceptual and Motor Skills, 110(2), 379–395. https://doi.org/10.2466/pms.110.2.379-395
• Alvarenga, M. S., Pereira, R. F., Scagliusi, F. B., Philippi, S. T., Estima, C. P. C., & Croll, J. (2010). Psychometric evaluation of the Disordered Eating Attitude Scale (DEAS) – English version. Appetite, 55(3), 374–376. https://doi.org/10.1016/j.appet.2010.07.009
• Alvarenga, M. S., Francischi, R., Fontes, F., Scagliusi, F. B., & Philippi, S. T. (2010). Adaptación y validación al Español del Disordered Eating Attitude Scale (DEAS). Perspectivas en Nutrición Humana, 12(1), 11–23.
• 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
• Garner, D. M., Olmsted, M. P., Bohr, Y., & Garfinkel, P. E. (1982). The Eating Attitudes Test: Psychometric features and clinical correlates. Psychological Medicine, 12(4), 871–878. https://doi.org/10.1017/s0033291700049163
• Herman, C. P., & Polivy, J. (1975). Anxiety, restraint, and eating. Journal of Abnormal Psychology, 84(6), 666–672. https://doi.org/10.1037/0021-843X.84.6.666
• Tylka, T. L. (2006). Development and psychometric evaluation of a measure of intuitive eating. Journal of Counseling Psychology, 53(2), 226–240. https://doi.org/10.1037/0022-0167.53.2.226

Items of the Scale