SCOFF Questionnaire

1. Abstract

The SCOFF Questionnaire is a brief, five-item screening instrument developed by Morgan, Reid, and Lacey (1999) at St George’s Hospital Medical School, London. Engineered to address the widespread under-detection of eating disorders in primary care and community settings, the tool utilizes an easily memorable acronym representing its core clinical targets: Sick (induced vomiting), Control (loss of control over eating), One stone (significant acute weight loss), Fat (perceptual body image distortion), and Food (pervasive food preoccupation). Each item is evaluated using a simple dichotomous (Yes vs. No) response format. The instrument is intentionally designed as an accessible, rapid screening device that can be administered orally or via self-report pencil-and-paper or digital formats by both specialist and non-specialist healthcare professionals. Psychometrically, the SCOFF exhibits robust diagnostic utility. Systematic reviews and meta-analyses, such as the comprehensive synthesis by Botella et al. (2013) evaluating 15 validation studies, confirm that a cut-off score of ≥2 achieves high pooled sensitivity (approximately 86% to 90%) and specificity (83% to 89%) for detecting anorexia nervosa and bulimia nervosa. Although its brevity inherently limits dimensional nuance and its sensitivity toward binge eating disorder or male-pattern muscularity concerns is comparatively attenuated, the SCOFF remains one of the world’s most extensively validated, translated, and cost-effective clinical triage instruments in behavioral medicine and psychiatric epidemiology.

2. Keywords

SCOFF Questionnaire, eating disorders, anorexia nervosa, bulimia nervosa, psychometrics, screening tool, primary care, sensitivity and specificity, body image distortion, diagnostic triage

3. Authors

The SCOFF Questionnaire was formulated and validated by a clinical research team within the Department of Psychiatry at St George’s Hospital Medical School (now St George’s, University of London), United Kingdom:

• Dr. John F. Morgan, MD, MRCPsych — Consultant Psychiatrist and Senior Lecturer in Eating Disorders, St George’s Hospital Medical School, London, UK.
• Fiona Reid, MSc — Lecturer in Medical Statistics, Department of Public Health Sciences, St George’s Hospital Medical School, London, UK.
• Professor J. Hubert Lacey, MD, MPhil, FRCPsych — Emeritus Professor of Psychiatry and Head of the Department of Psychiatry, St George’s Hospital Medical School, London, UK.

4. Purpose

Eating disorders, predominantly anorexia nervosa, bulimia nervosa, and related variants classified under Other Specified Feeding or Eating Disorders (OSFED), represent severe psychiatric illnesses characterized by elevated medical morbidity, marked psychosocial impairment, and disproportionately high mortality rates. Despite their severity, a substantial proportion of individuals with eating disorders go unrecognized in frontline medical settings. Patients frequently conceal behavioral symptoms due to shame, guilt, ambivalence toward recovery, or the egosyntonic nature of restrictive pathology. Concurrently, primary care practitioners and general physicians frequently operate under severe consultation time constraints and may lack specialized training in psychiatric semiology, leading to missed windows for early clinical intervention.

The primary clinical and epidemiological purpose of the SCOFF Questionnaire is to serve as an ultrabrief, memorable, and low-burden screening mechanism designed to identify individuals at high risk for eating pathology. Rather than functioning as a definitive diagnostic instrument, the questionnaire operates as a standardized behavioral filter. Its specific objective is to raise a high index of clinical suspicion that warrants immediate, comprehensive assessment via formal diagnostic interviews, such as the Eating Disorder Examination (EDE) or the Structured Clinical Interview for DSM Disorders (SCID).

The theoretical and practical rationale underpinning the SCOFF rests on four operational requirements identified by Morgan and colleagues:

• Simplicity and Memorability: Frontline clinicians require an instrument that can be recalled effortlessly during an unstructured interview without necessitating manual consultation or scoring templates.
• Versatility of Administration: The questions must retain psychometric equivalence whether administered verbally as conversational clinical inquiries or embedded within written and digital self-report batteries.
• Threshold Sensitivity: As a secondary preventive triage tool, the index must emphasize high diagnostic sensitivity to minimize false-negative outcomes, ensuring vulnerable individuals are not prematurely dismissed.
• Non-Specialist Feasibility: The phrasing must avoid dense psychiatric jargon, enabling effective deployment by school counselors, general practitioners, sports physicians, triage nurses, and university health services.

5. Psychological Construct

The SCOFF Questionnaire measures the general latent construct of core eating disorder psychopathology. Rather than evaluating the full dimensional spectrum of feeding behavior, the tool distills the nuclear cognitive, perceptual, and behavioral features that characterize the clinical phenotypes of anorexia nervosa and bulimia nervosa according to the diagnostic parameters of the ICD and DSM. The five items capture distinct, highly discriminatory facets of eating pathology:

1. Self-Induced Compensatory Vomiting (‘S’)

The first item (Do you make yourself Sick because you feel uncomfortably full?) targets postprandial purging via self-induced emesis. In clinical psychopathology, this behavior represents a cardinal compensatory mechanism utilized to counteract the physiological and cognitive anxiety associated with food consumption. It evaluates both the physical behavior of purging and the intolerable visceral sensation of gastric fullness, which individuals with eating disorders frequently conflate with catastrophic weight gain.

2. Cognitive and Behavioral Loss of Control (‘C’)

The second item (Do you worry you have lost Control over how much you eat?) assesses the subjective dysregulation of consumption. Loss of control (LOC) eating is recognized across both cognitive-behavioral and psychiatric diagnostic models as the essential psychopathological criterion distinguishing objective and subjective binge eating episodes from normal overeating. This dimension measures psychological distress regarding consumption boundaries and perceived vulnerability to behavioral disinhibition.

3. Acute, Clinically Significant Weight Depletion (‘O’)

The third item (Have you recently lost more than One stone in a 3 month period?; adapted in non-UK jurisdictions to 14 or 15 pounds, or 6.35 to 7 kilograms) functions as an objective somatic index of rapid negative energy balance. Sustained and precipitous weight reduction over a bounded temporal frame (three months) is an empirical hallmark of severe dietary restriction, medical cachexia, or unchecked metabolic purging, serving as a physiological red flag for acute anorectic transition.

4. Perceptual and Cognitive Body Image Distortion (‘F’)

The fourth item (Do you believe yourself to be Fat when others say you are too thin?) evaluates body dysmorphic valuation, severe perceptual distortion, and the overvaluation of shape and weight. Crucially, the phrasing incorporates a social comparison anchor (when others say you are too thin), operationalizing the classic clinical phenomenon of cognitive dissonance and delusional-like conviction wherein an emaciated patient maintains an intransigent subjective self-appraisal of corpulence despite objective external refutation.

5. Obsessive Food and Nutritional Preoccupation (‘F’)

The final item (Would you say that Food dominates your life?) appraises cognitive intrusion and obsessive-compulsive preoccupation with dietary intake, caloric counting, meal timing, and somatic avoidance. As demonstrated in historical starvation experiments (e.g., the Minnesota Starvation Experiment), biological energy deprivation combined with psychiatric overvaluation produces pervasive cognitive capture, wherein ruminations regarding nutritional intake displace functional occupational, relational, and recreational thoughts.

6. Theoretical Framework

The conceptual architecture of the SCOFF Questionnaire is rooted in the intersection of descriptive psychiatric phenomenology and the transdiagnostic cognitive-behavioral model of eating disorders formulated by Fairburn, Cooper, and Shafran (2003). This theoretical paradigm posits that while anorexia nervosa, bulimia nervosa, and atypical eating disorders manifest with distinct overt behavioral profiles (e.g., extreme dietary restriction versus cyclical binge-purge episodes), they share a single central cognitive nucleus: the overvaluation of eating, shape, and weight.

In classical psychodynamic formulations (such as those of Hilde Bruch), eating disorders were framed as struggles for autonomy and perceptual self-control manifested through visceral control over the somatic self. Bruch highlighted profound disturbances in body image perception and interoceptive awareness, such as the inability to recognize visceral satiety or somatic emaciation. The SCOFF operationalizes Bruch’s foundational concepts directly within its somatic items: item 1 captures the rejection of bodily satiety, while item 4 targets the failure of interoceptive and perceptual self-appraisal.

Within the cognitive-behavioral framework, Fairburn and colleagues theorize that individuals evaluate their self-worth almost exclusively in terms of their body shape, weight, and capacity to exercise rigid dietary restraint. This core psychopathology produces secondary maintaining mechanisms:

• Cognitive Narrowing: The cognitive schema becomes dominated by food-related stimuli, hypervigilance, and obsessive calculation (captured by SCOFF Item 5).
• Strict Dietary Restraint: Unrealistic, perfectionistic dietary rules lead directly to dramatic weight loss (captured by SCOFF Item 3).
• Dichotomous Dietary Beliefs: Inevitable perceived violations of rigid dietary standards trigger catastrophic interpretations of loss of control (captured by SCOFF Item 2).
• Compensatory Purging: Induced emesis (captured by SCOFF Item 1) is employed to manage severe panic following real or perceived excessive consumption, which paradoxically perpetuates further binge episodes by disrupting physiologic satiety cues.

By mapping its five inquiries directly onto the primary nodes of this cognitive-behavioral maintenance cycle, the SCOFF successfully isolates the operational core of the disorder using the fewest possible behavioral and cognitive indicators.

7. Validity

The psychometric validity of the SCOFF Questionnaire has been investigated across diverse clinical populations, epidemiological cohorts, and cultural translations:

Criterion and Diagnostic Validity

In the seminal development study conducted by Morgan, Reid, and Lacey (1999), the SCOFF was validated against a sample of 116 women diagnosed with anorexia nervosa or bulimia nervosa according to DSM-IV criteria, alongside 96 healthy control subjects. Utilizing a cut-off score of ≥2 positive responses, the instrument demonstrated 100% sensitivity (95% CI [97%, 100%]) for anorexia and bulimia, accompanied by an overall specificity of 87.5% (95% CI [79%, 93%]) against controls. Subsequent field testing within unselected primary care populations (Morgan et al., 1999) affirmed that the ≥2 threshold retained high diagnostic yield while generating manageable false-positive rates suitable for secondary referral.

Meta-Analytic Syntheses

Subsequent independent studies in general practice yielded more realistic, non-idealized psychometric indices. The meta-analysis conducted by Botella et al. (2013) aggregated data across 15 validation trials comprising over 3,000 individuals. Across all studies, the pooled sensitivity at a cut-off score of ≥2 was estimated at 0.86 (95% CI [0.77, 0.91]), with a pooled specificity of 0.83 (95% CI [0.77, 0.88]). When evaluated under a more conservative cut-off threshold of ≥3, pooled sensitivity shifted to 0.71 while specificity climbed to 0.92, indicating that researchers seeking to eliminate false positives in non-clinical epidemiological contexts frequently prefer the higher threshold.

Convergent and Concurrent Validity

Convergent validity has been established by comparing SCOFF scores against established, multi-item dimensional inventories. Validation trials consistently yield strong positive correlations between the SCOFF and total scores on the Eating Attitudes Test (EAT-26) (Pearson r typically ranging between .60 and .74), the Eating Disorder Inventory (EDI-2 / EDI-3), and the Eating Disorder Examination Questionnaire (EDE-Q). Cross-cultural evaluations, including the Spanish validation by Garcia-Campayo et al. (2005), the French translation by Garcia et al. (2010), and the Finnish nationwide cohort study by Lähteenmäki et al. (2009), have repeatedly corroborated high diagnostic receiver operating characteristic area under the curve values (ROC AUC ranging from 0.82 to 0.94), affirming structural and construct equivalence internationally.

8. Reliability

Because the SCOFF comprises only five dichotomous items designed to capture distinct behavioural and cognitive targets, classical metrics of internal consistency (such as Cronbach’s alpha) must be interpreted with caution. In brief, heterogenous screening tools, elevated alpha coefficients can reflect excessive item redundancy rather than diagnostic utility.

Internal Consistency

Across published literature, estimates of internal consistency utilizing the Kuder-Richardson Formula 20 (KR-20) or Cronbach’s alpha typically fall within the moderate range: between .48 and .68 in non-clinical community cohorts, and between .60 and .76 in targeted clinical populations (e.g., Pannocchia et al., 2011; Solmi et al., 2014). Item-total correlations are consistently positive, typically ranging between .30 and .55, confirming that while each item evaluates an independent symptom domain, they share variance attributable to an overarching latent disorder dimension.

Test-Retest Stability and Inter-Format Equivalence

Temporal stability evaluated over short test-retest intervals (1 to 2 weeks) shows strong concordance. In a dedicated methodological investigation, Perry et al. (2002) evaluated the reliability and equivalence of the SCOFF when administered as an oral interview versus a written self-report questionnaire among women attending primary care practices. Cohen’s kappa coefficient between oral and written administration formats reached κ = 0.73, with overall test-retest stability of κ = 0.82. These findings confirm that the diagnostic yield remains resilient regardless of whether the questions are posed face-to-face during clinical consultation or completed autonomously by the patient.

9. Factor Analysis

Extensive factor-analytic evaluations of the SCOFF have yielded essential insights into its dimensional behavior across clinical and community samples.

Exploratory Factor Analysis (EFA)

Exploratory factor analyses conducted on tetrachoric correlation matrices (the appropriate statistical standard for dichotomous response options) generally demonstrate a dominant single-factor solution accounting for between 42% and 58% of the total item variance. The items evaluating body image distortion (Fat), cognitive preoccupation (Food), and loss of control (Control) consistently demonstrate the highest factor loadings, frequently exceeding .70. In contrast, the item evaluating rapid weight loss (One stone) occasionally manifests lower loadings (.35 to .50) in non-clinical populations, reflecting the low base-rate of acute somatic weight loss among mild or subthreshold community cases.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic investigations evaluating the unidimensional structural model have reported adequate to good model fit indices across various demographic strata. For example, structural modeling conducted on general population data by Solmi et al. (2014) affirmed that a single-factor configuration exhibited good fit when accounting for binary indicators:

• Comparative Fit Index (CFI): Values consistently range between .95 and .99.
• Tucker-Lewis Index (TLI): Values typically exceed .95.
• Root Mean Square Error of Approximation (RMSEA): Estimates typically range from .02 to .05, demonstrating acceptable boundary approximation.

Item Response Theory (IRT)

When evaluated through two-parameter logistic (2PL) Item Response Theory paradigms, the SCOFF items demonstrate distinct threshold and discrimination profiles:

• High Discrimination (α): The induced vomiting item (Sick) and the body image distortion item (Fat) exhibit high discrimination parameters (α > 1.5), functioning as severe markers located at the upper extremes of latent trait severity (θ > +1.5 standard deviations above the population mean).
• Broad Latent Coverage (β): The items addressing food domination (Food) and loss of control (Control) display lower difficulty thresholds (β ≈ 0.0 to +0.8), functioning effectively as entry-level indicators that identify individuals at earlier or subclinical phases of eating distress.

10. Instrument / Measurement Tool

The structured technical characteristics of the SCOFF Questionnaire are summarized below:

• Instrument Type: Brief Psychiatric Screening Questionnaire / Clinical Triage Instrument.
• Format: Available as a clinician-administered oral interview, paper-and-pencil self-report inventory, or digital questionnaire.
• Target Population: Adolescents and adults (ages 12 years and older) across primary healthcare, school-based, athletic, and general medical contexts.
• Number of Items: 5 primary screening items (supplemented optionally in clinical interviews by two adjunct questions assessing meal pattern satisfaction and secret eating).
• Response Scale: Dichotomous (Yes / No).
• Item Scoring Protocol:
  • Each affirmative response (Yes) is assigned a value of 1 point.
  • Each negative response (No) is assigned a value of 0 points.
  • Total score range: 0 to 5 points.
• Reverse Scoring Rules: None. All items are scored in a direct positive direction.
• Diagnostic Cut-off Guidelines:
  • Score 0 – 1: Low probability of an active clinical eating disorder. Routine surveillance advised if clinical concern persists.
  • Score ≥ 2 (Standard Threshold): Positive screen. High sensitivity for anorexia nervosa and bulimia nervosa. Indicates a high likelihood of eating pathology and warrants immediate referral for full clinical diagnostic evaluation.
  • Score ≥ 3 (Conservative Threshold): Enhanced specificity. Recommended in research settings and general epidemiological surveys to minimize false positives.
• Clinical Adjunct Questions (Bulimia Focus):
  • Are you satisfied with your eating patterns? (Answering ‘No’ raises suspicion).
  • Do you ever eat in secret? (Answering ‘Yes’ raises suspicion).

11. Permissions & Fee and Test Year

The SCOFF Questionnaire was developed and published in 1999 by Dr. John F. Morgan, Fiona Reid, and Professor J. Hubert Lacey at St George’s Hospital Medical School, London. The original seminal validation study appeared in the British Medical Journal (BMJ). In keeping with the explicit objectives of the authors and public health imperatives, the SCOFF is positioned within the public domain for non-commercial clinical use, medical education, and academic research. No licensing fees or royalty payments are required to administer the scale in routine clinical practice, primary care triage, or scholarly research. Commercial software integration or copyright incorporation into proprietary digital diagnostic suites requires adherence to standard journal attribution and fair-use guidelines as established by BMJ Publishing Group Ltd.

12. References

The following peer-reviewed literature provides the empirical foundation for the SCOFF Questionnaire’s design, psychometric integrity, and cross-cultural validation:

• Botella, J., Sepúlveda, A. R., Huang, H., & Gambara, H. (2013). A meta-analysis of the diagnostic accuracy of the SCOFF. The Spanish Journal of Psychology, 16, Article E92. https://doi.org/10.1017/sjp.2013.92
• 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
• Garcia, F., Grigioni, S., Chelali, S., Meyrignac, G., Thibaut, F., & Dechelotte, P. (2010). Validation of the French version of SCOFF questionnaire for screening of eating disorders among adults. The World Journal of Biological Psychiatry, 11(7), 888–893. https://doi.org/10.3109/15622975.2010.483251
• Garcia-Campayo, J., Sanz-Carrillo, C., Ibañez, J., Lou, S., Solano, V., & Alda, M. (2005). Validation of the Spanish version of the SCOFF questionnaire for the screening of eating disorders in primary care. Journal of Psychosomatic Research, 59(2), 51–55. https://doi.org/10.1016/j.jpsychores.2004.06.005
• Lähteenmäki, S., Aalto-Setälä, T., Suokas, J., Saarni, S. I., Perälä, J., Saarni, S. E., Aro, H., Lönnqvist, J., & Suvisaari, J. (2009). Validation of the Finnish version of the SCOFF questionnaire among young adults aged 20 to 35 years. BMC Psychiatry, 9, Article 5. https://doi.org/10.1186/1471-244X-9-5
• Morgan, J. F., Reid, F., & Lacey, J. H. (1999). The SCOFF questionnaire: Assessment of a new screening tool for eating disorders. BMJ, 319(7223), 1467–1468. https://doi.org/10.1136/bmj.319.7223.1467
• Pannocchia, L., Fiorino, M., Giannini, M., & Vanderlinden, J. (2011). A psychometric exploration of an Italian translation of the SCOFF Questionnaire. European Eating Disorders Review, 19(4), 371–373. https://doi.org/10.1002/erv.1105
• Perry, L., Morgan, J., Reid, F., Brunton, J., O’Brien, A., Luck, A., & Lacey, H. (2002). Screening for symptoms of eating disorders: Reliability of the SCOFF screening tool with written compared to oral delivery. International Journal of Eating Disorders, 32(4), 466–472. https://doi.org/10.1002/eat.10093
• Solmi, F., Hatch, S. L., Hotopf, M., Treasure, J., & Micali, N. (2014). Validation of the SCOFF questionnaire for eating disorders in a multiethnic general population sample. International Journal of Eating Disorders, 48(3), 312–316. https://doi.org/10.1002/eat.22373

13. Items of the Scale