Eating Disorder Online Activity Questionnaire (EDOAQ)

1. Abstract

The Eating Disorder Online Activity Questionnaire (EDOAQ), formulated by Redmond, Lamonby, Harrington, and Wood (2023), is an innovative clinical assessment instrument engineered to investigate the nuanced intersection between Problematic Use of the Internet (PUI) and the etiology, maintenance, and phenomenology of eating disorders (ED) in young populations. Developed as a participatory, multi-stakeholder assessment tool within National Health Service (NHS) Child and Adolescent Mental Health Services (CAMHS) and Community Eating Disorder Services (CEDS) across North West England, the EDOAQ addresses a critical diagnostic and therapeutic void: the absence of tailored tools capable of decoding how contemporary digital ecologies—such as algorithmic social media, digital peer communities, and visually driven micro-targeting—interact with eating disorder pathology. The initial development and field evaluation utilized a qualitative participatory framework involving clinical psychologists, multidisciplinary healthcare specialists, adolescent service users aged 11 to 17, and their parents or carers. Through iterative co-production and reflexive thematic analysis, the authors delineated three core experiential and operational dimensions: (1) the usability and clinical acceptability of the EDOAQ, (2) the direct and indirect impact of online activity on eating disorder cognitions and behaviors, and (3) broader patterns of engagement with the digital world. The instrument is intended for dual delivery via electronic and paper formats. While the initial published foundational study focused predominantly on qualitative content verification, face validity, and thematic acceptability, systematic quantitative psychometric evaluations—encompassing internal consistency reliability, exploratory and confirmatory factor structures, and longitudinal predictive validity—remain active priorities for subsequent mixed-methods clinical research. The EDOAQ provides clinicians with an indispensable structured scaffold to demystify online digital behaviors, foster non-judgmental therapeutic alliance, and identify potent digital triggers in adolescent eating pathology.

2. Keywords

Adolescence, Adolescent Behavior, Eating Disorders, Problematic Use of the Internet, Problematic Internet Use, Online Activity, Social Media Influence, Body Image Disturbance, Clinical Assessment, Child and Adolescent Mental Health Services, Digital Phenotyping, Qualitative Instrument Development.

3. Authors

The Eating Disorder Online Activity Questionnaire was conceptualized, designed, and evaluated by a collaborative team of clinical researchers and NHS healthcare professionals specializing in adolescent psychological medicine and community eating disorder services:

• Tomos Redmond, ClinPsyD — Mersey Care NHS Foundation Trust, United Kingdom. (ORCID: 0000-0002-9216-9748; Correspondence: [email protected]).
• Jasmine Lamonby, MSc — Department of Psychology, University of Manchester, Manchester, United Kingdom.
• Lesley Harrington, PhD — Pennine Care NHS Foundation Trust, Greater Manchester, United Kingdom.
• Alison Wood, DClinPsy — Pennine Care NHS Foundation Trust, Greater Manchester, United Kingdom (Email: [email protected]).

4. Purpose

The primary clinical and research objective of the Eating Disorder Online Activity Questionnaire (EDOAQ) is to provide an evidence-informed, developmentally sensitive framework that enables healthcare professionals to elicit, evaluate, and interpret the complex digital behaviors of adolescents undergoing treatment for eating disorders. Over the past two decades, the ubiquity of smartphone devices, ubiquitous digital connectivity, and image-centric social media architectures (e.g., Instagram, TikTok, Snapchat, YouTube) has radically restructured the psychosocial landscape of youth. Despite widespread recognition among frontline clinicians that online environments can dramatically accelerate body dissatisfaction, dietary restriction, purging, and compulsive exercising, standard psychometric assessment batteries in psychiatric clinics have historically neglected dedicated examinations of patients’ online lived realities.

Traditional diagnostic measures of eating pathology, such as the Eating Disorder Examination (EDE) and the self-report Eating Disorder Examination Questionnaire (EDE-Q), focus meticulously on cognitive constructs (e.g., shape concern, weight concern, restraint, eating concern) and behavioral topographies (e.g., objective binge episodes, self-induced vomiting, laxative abuse). However, these conventional instruments entirely omit the digital interfaces where these cognitions are continually catalyzed, reinforced, and intensified. Conversely, generic instruments designed to evaluate Internet Addiction or Problematic Use of the Internet (PUI)—such as the Internet Addiction Test (IAT) or the Compulsive Internet Use Scale (CIUS)—measure generalized dependency criteria (e.g., salience, tolerance, withdrawal, conflict) without contextualizing the specific content, algorithms, or community dynamics that drive psychiatric distress in eating disorders.

The EDOAQ was engineered to bridge this diagnostic chasm. Specifically, its operational purposes encompass:

• Facilitating Open Clinical Dialogues: Mitigating adolescent shame, secrecy, and parental conflict regarding internet use by introducing a neutral, collaborative clinical tool that legitimizes discussing digital life within clinical consultations.
• Deconstructing Toxic Digital Exposures: Identifying patient engagement with pernicious online content, including deliberate searches for “thinspiration,” “fitspiration,” “bonespiration,” restrictive dieting hacks, fasting challenges, and pro-eating disorder (“pro-ana” and “pro-mia”) peer forums.
• Mapping Algorithmic Echo Chambers: Assisting clinicians and young individuals in understanding how recommendation algorithms detect vulnerabilities in mood and body image, systematically flooding user feeds with pathology-reinforcing content.
• Formulating Tailored Behavioral Interventions: Serving as a direct baseline assessment for digital hygiene interventions, algorithmic retraining, behavioral experiments, structured digital detoxes, and selective unfollowing within cognitive-behavioral and family-based treatment paradigms.

5. Psychological Construct

The Eating Disorder Online Activity Questionnaire assesses a multi-layered, hybrid psychological construct residing at the convergence of Problematic Use of the Internet (PUI) and Eating Disorder Pathology. Rather than treating internet use as an undifferentiated, homogenous activity, the construct captures how specific digital affordances interact bidirectionally with cognitive vulnerabilities, emotional dysregulation, and body image disturbance in young people.

Construct Dimensions and Phenomenological Themes

As identified in the foundational developmental qualitative investigation by Redmond et al. (2023), this hybrid construct comprises three primary dimensions:

1. Usability, Acceptability, and Disclosure Dynamics

This operational dimension pertains to the psychological and relational feasibility of disclosing digital habits in clinical settings. Adolescents experiencing eating disorders often harbor profound fears of external control, confiscation of digital devices by caregivers, or clinical condemnation regarding their online searches. This dimension assesses the degree to which an assessment tool feels non-punitive, supportive, clear, and developmentally attuned. A high level of clinical acceptability reflects a measurement environment that diminishes evaluative anxiety, counteracts defensive minimization, and establishes an open clinical alliance.

2. Impact of Online Activity on Eating Disorder Pathology

This core functional dimension captures the dynamic, bidirectional relationship between digital media consumption and the exacerbation or maintenance of eating disorder symptoms. It encompasses several distinct cognitive-behavioral manifestations:

• Upward Appearance Comparison: Chronic, automatic comparison of one’s own physical appearance with digitally manipulated, filtered, or highly curated images of peers, fitness influencers, and celebrities, leading to acute body dissatisfaction.
• Exposure to Pro-Eating Disorder Content: Deliberate or algorithmic exposure to explicit weight-loss techniques, extreme fasting regimens, purging guidance, self-harm imagery, and competitive illness validation communities.
• Digital Reinforcement and Metric Fixation: The psychological reliance on digital validation (e.g., “likes,” comments, follower counts, view tallies) as quantifiable metrics of self-worth and physical attractiveness.
• Algorithmic Trapping: The experience of involuntary immersion in disordered content loops driven by machine-learning recommendation engines that misinterpret obsessive surveillance of nutrition and exercise as standard user engagement.

3. Engagement with the Digital World

This structural dimension evaluates the breadth, intensity, and functional topology of an adolescent’s digital lifestyle. It examines the quantitative parameters of screen time across diverse platforms (e.g., visual social networks, video-sharing portals, private messaging groups, gaming environments), the timing of use (e.g., nocturnal screen use disrupting circadian sleep architecture), the motivation underlying digital immersion (e.g., experiential avoidance, emotional numbing, social connection vs. isolation), and the presence of covert or deceptive online behaviors hidden from parental oversight.

6. Theoretical Framework

The architecture of the EDOAQ is grounded in several well-established psychological theories that elucidate how media consumption, sociocultural pressures, and cognitive vulnerabilities intersect during adolescent neurodevelopment:

1. The Tripartite Influence Model

Formulated by Thompson, Heinberg, Altabe, and Tantleff-Dunn (1999), the Tripartite Influence Model posits that three primary sociocultural vectors—peers, parents, and the media—exert profound influences on body image dissatisfaction and eating disturbances. This pressure operates through two primary mediating mechanisms: the internalisation of the thin ideal (or lean/muscular ideal) and appearance-based social comparison. In the contemporary context, interactive social media has converged these three channels: peers and media influencers now occupy the same digital stream, dramatically amplifying appearance comparison frequency and immediacy.

2. Festinger’s Social Comparison Theory

According to Leon Festinger’s (1954) Social Comparison Theory, human beings possess an innate drive to evaluate their own abilities, attributes, and social standing through comparisons with others. In visual digital spaces, adolescents engage in pervasive upward social comparisons, contrasting their unedited real-life appearance with hyper-idealized, digitally altered depictions of others. This chronic upward comparison generates substantial internal dissonance, shame, and perceived physical inadequacy, which in vulnerable youths fuels restrictive eating and compensatory behaviors as compensatory control strategies.

3. Objectification Theory and Self-Objectification

Developed by Fredrickson and Roberts (1997), Objectification Theory asserts that individuals—particularly young women and girls—are socialized to internalize an observer’s perspective of their physical bodies. In the digital environment, this manifests as intensive self-objectification: adolescents treat their bodies as visual objects to be photographed, altered, displayed, monitored, and evaluated. The relentless digital gaze and quantification of aesthetic feedback trigger habitual body monitoring, heightened body shame, and elevated susceptibility to clinical eating disorders.

4. Cognitive-Behavioral Model of Pathological Internet Use

Richard A. Davis’s (2001) cognitive-behavioral model delineates between generalized problematic internet use and specific problematic internet use. Specific PUI occurs when an individual utilizes digital affordances to enact, magnify, or maintain a pre-existing psychiatric vulnerability or behavioral fixation. The EDOAQ relies heavily on this principle: it conceptualizes adolescent eating disorder online activity not merely as an addiction to the internet as a medium, but as an online manifestation of cognitive overvaluation of shape, weight, and food control (Fairburn et al., 2003).

7. Validity

The psychometric evaluation of the EDOAQ represents an evolving research program originating from rigorous qualitative participatory health design methodologies (Redmond et al., 2023).

Content Validity and Co-Design Methodology

High content validity was embedded into the EDOAQ from inception through collaborative development involving clinical psychologists, multidisciplinary healthcare professionals from NHS Community Eating Disorder Services, adolescent service users with lived experience of clinical eating disorders, and their parents or carers. This multi-perspective synthesis ensured that the instrument’s operational domains directly reflect real-world adolescent digital behaviors rather than speculative adult hypotheses. Clinicians confirmed that the questionnaire captured domains essential for formulation, risk assessment, and treatment planning, affirming robust qualitative content validity.

Face Validity

During the qualitative evaluation phase, adolescents aged 11 to 17 recently admitted to specialized CEDS protocols completed the preliminary EDOAQ and participated in structured qualitative feedback sessions. Participants reported that the questionnaire felt authentic, directly addressed their daily media struggles, used culturally and developmentally appropriate language, and avoided moralistic or punitive overtones. Clinicians echoed these observations, noting that young people engaged readily with the format without showing signs of heightened assessment resistance.

Construct, Convergent, and Discriminant Validity Roadmap

Because Redmond et al. (2023) established the measure through an initial qualitative study, traditional quantitative indices of validity (e.g., Pearson correlations, multitrait-multimethod matrices) were not evaluated in the initial paper. To establish quantitative construct validity, the authors outlined a roadmap for future research employing mixed qualitative and quantitative methodologies. Necessary next steps include demonstrating:

• Convergent Validity: Statistically significant positive correlations between EDOAQ dimensional scores and validated eating disorder inventories, such as the Global Score and subscales of the Eating Disorder Examination Questionnaire (EDE-Q), as well as generalized internet addiction measures like the Young’s Internet Addiction Test (IAT) or the Smartphone Addiction Scale (SAS).
• Discriminant Validity: The ability of the EDOAQ to differentiate between specific eating-related digital pathology and non-clinical, normative recreational social media use, along with demonstrating distinction from generalized depressive withdrawal and social anxiety symptoms.
• Predictive and Ecological Validity: Longitudinal studies verifying whether baseline EDOAQ scores predict clinical outcomes, treatment engagement, relapse frequency, or urgent psychiatric hospitalizations among adolescents receiving outpatient care.

8. Reliability

At its current stage of psychometric development, the Eating Disorder Online Activity Questionnaire has not yet been subjected to large-scale parametric psychometric testing; therefore, empirical reliability coefficients (such as Cronbach’s alpha, McDonald’s omega, or intra-class correlation coefficients for test-retest reliability) are not reported in the foundational source literature (Redmond et al., 2023).

From a classical test theory (CTT) and modern psychometric perspective, calculating internal consistency across the instrument presents unique conceptual considerations. Because the EDOAQ functions concurrently as a clinical interview aid, risk detection inventory, and continuous rating scale, some item subsets may reflect causal indicators (e.g., exposure to specific harmful websites or nighttime screen time) rather than pure effect indicators of a single underlying latent trait. In such formative or mixed-indicator measurement frameworks, maximizing Cronbach’s alpha is not always appropriate or indicative of measurement quality.

To establish comprehensive reliability in future quantitative field trials, researchers must conduct:

• Internal Consistency Evaluation: Calculation of McDonald’s hierarchical omega ($\omega_h$) and categorical omega ($\omega_t$) across identified latent subscales within sufficiently powered clinical cohorts ($N > 250$).
• Test-Retest Stability: Determination of intra-class correlation coefficients (ICC) across a 1-to-2 week interval in clinically stable adolescent outpatients, accounting for rapid behavioral shifts common to digital device usage.
• Inter-Rater Reliability: Establishing high Cohen’s kappa ($\kappa$) or two-way mixed-effects ICCs when clinicians utilize the EDOAQ to rate clinical risk, treatment targets, or digital maintenance factors during intake formulations.

9. Factor Analysis

Statistical factor analyses, including Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), have not yet been published for the EDOAQ. The current developmental framework rests upon rigorous qualitative thematic analysis conducted following the reflexive principles of Braun and Clarke.

In the foundational investigation by Redmond et al. (2023), qualitative thematic analysis of extensive interview transcripts with adolescent service users (aged 11–17) and specialist clinicians identified three dominant structural themes that provide the structural foundation for upcoming latent factor modeling:

Subsequent psychometric phases will require administering the refined questionnaire to a representative, multi-center clinical sample across Child and Adolescent Mental Health Services. An EFA using polychoric correlation matrices and robust weighted least squares estimators (WLSMV)—appropriate for ordinal and categorical items—will be critical to determine whether a unidimensional, multidimensional oblique, or bi-factor structural model best characterizes the data. Subsequent CFA will evaluate goodness-of-fit using benchmark structural equation modeling indices (e.g., Comparative Fit Index $\ge 0.95$, Tucker-Lewis Index $\ge 0.95$, Root Mean Square Error of Approximation $\le 0.06$).

10. Instrument / Measurement Tool

The operational specifications of the Eating Disorder Online Activity Questionnaire (EDOAQ) based on the development study by Redmond et al. (2023) are summarized below:

• Instrument Type: Original Clinical Inventory / Semi-Structured Clinical Assessment Questionnaire.
• Administration Format: Dual modality — available in both Electronic (digital tablet, web interface) and traditional Paper-and-pencil formats.
• Primary Target Population: Clinical and sub-clinical populations; Human; Male, Female, and Gender Diverse individuals.
• Target Age Demographic: Childhood to Adolescence, specifically validated with young people aged 11 to 17 years (incorporating late school-age and adolescent developmental brackets).
• Clinical Setting: Specialized Community Eating Disorder Services (CEDS), Tier 3/4 Child and Adolescent Mental Health Services (CAMHS), inpatient adolescent psychiatric units, and specialized pediatric consultation-liaison settings.
• Language: English (UK).
• Administration Time: Approximately 15 to 25 minutes when completed independently; 30 to 45 minutes when utilized as a clinician-guided semi-structured interview during comprehensive assessment sessions.
• Scoring and Clinical Interpretation: In its present clinical implementation, the EDOAQ provides both continuous tracking of problematic online behaviors and a qualitative thematic map of digital risks. Responses highlight specific maintenance factors (e.g., algorithmic exposure, digital triggers, nocturnal use) that are directly incorporated into individualized cognitive-behavioral case formulations.

11. Permissions & Fee and Test Year

• Test Year: 2023.
• Copyright and Governance: The conceptual framework, design, and publication rights reside with the authors (Tomos Redmond, Jasmine Lamonby, Lesley Harrington, and Alison Wood) and the publishing entity (Bentham Science Publishers / Adolescent Psychiatry).
• Commercial Status: Non-commercial instrument.
• Fee: Free of charge for non-commercial clinical, healthcare, and academic research applications.
• Permissions and Access: Researchers and clinicians wishing to obtain, review, pilot, or validate the complete EDOAQ item inventory are instructed to contact the corresponding author directly: Dr. Tomos Redmond (Mersey Care NHS Foundation Trust, UK; Email: [email protected]).

12. References

• Davis, R. A. (2001). A cognitive-behavioral model of pathological Internet use. Computers in Human Behavior, 17(2), 187–195. https://doi.org/10.1016/S0747-5632(00)00041-8
• Fairburn, C. G., Cooper, Z., & Shafran, R. (2003). Cognitive behaviour therapy for eating disorders: A “transdiagnostic” theory and treatment strategy. Behaviour Research and Therapy, 41(5), 509–528. https://doi.org/10.1016/S0005-7967(02)00088-8
• Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7(2), 117–140. https://doi.org/10.1177/001872675400700202
• Fredrickson, B. L., & Roberts, T. A. (1997). Objectification theory: Toward understanding women’s lived experiences and mental health risks. Psychology of Women Quarterly, 21(2), 173–206. https://doi.org/10.1111/j.1471-6402.1997.tb00108.x
• Redmond, T., Lamonby, J., Harrington, L., & Wood, A. (2023). Developing the Eating Disorder Online Activity Questionnaire (EDOAQ) for adolescents in eating disorder services; a qualitative study. Adolescent Psychiatry, 13(1), 32–48. https://doi.org/10.2174/2210676613666230823121744
• Rodgers, R. F., Slater, A., Gordon, C. S., McLean, S. A., Jarman, H. K., & Paxton, S. J. (2020). A biopsychosocial model of social media use and body image concerns, disordered eating, and muscle-building behaviors among adolescent girls and boys. Journal of Youth and Adolescence, 49(2), 399–409. https://doi.org/10.1007/s10964-019-01190-0
• Thompson, J. K., Heinberg, L. J., Altabe, M., & Tantleff-Dunn, S. (1999). Exacting beauty: Theory, assessment, and treatment of body image disturbance. American Psychological Association. https://doi.org/10.1037/10312-000

13. Items of the Scale