Diabulimia Knowledge Level Scale

Abstract

The Diabulimia Knowledge Level Scale (DKLS; Çetinkaya Özdemir, Semerci, & Sönmez Sarı, 2023) is a psychometrically validated, self-report assessment instrument designed to measure healthcare professionals’—specifically nurses’—clinical understanding, recognition capacity, and foundational knowledge concerning diabulimia. Diabulimia, clinically conceptualized as an eating disorder in diabetes mellitus type 1 (ED-DMT1), involves the deliberate, non-adherent restriction or total omission of insulin for the explicit purpose of weight reduction through intentional glycosuria. Because frontline healthcare providers frequently misidentify the warning signs of this co-occurring metabolic and psychiatric condition as routine treatment non-compliance or idiopathic diabetic instability, the development of a standardized psychometric tool was paramount. Grounded in a comprehensive synthesis of contemporary diabetology, clinical psychiatric nursing literature, and psychiatric nosology, the DKLS was established through an extensive item-generation process and evaluated among a cohort of practicing clinical nurses in Turkey.

The final instrument comprises 19 items structured across four distinct conceptual dimensions: General Information, Behaviors, Psychological Effects, and Physiological Effects. All items are rated on a standard 5-point Likert-type scale ranging from 1 (“Strongly disagree”) to 5 (“Strongly agree”). Psychometric evaluation demonstrated robust content validity, characterized by item- and scale-level Content Validity Index (CVI) values exceeding conventional thresholds (0.97 for relevance, 0.92 for simplicity, and 0.95 for clarity). Exploratory factor analysis using principal component analysis with varimax rotation supported a four-factor structure explaining 68.53% of the total variance, subsequently tested via confirmatory factor analysis (CFA). Reliability assessments revealed satisfactory internal consistency, with Cronbach’s alpha coefficients ranging between .70 and .92 across individual subscales and the global scale. Temporal stability evaluated across a three-week test-retest protocol demonstrated stable measurement properties without significant mean drift (p = .154). The DKLS serves as an objective, empirical instrument for continuing nursing education, hospital-wide diagnostic readiness benchmarking, and epidemiological research in medical-surgical and psychiatric settings.

Keywords

Diabulimia, Diabulimia Knowledge Level Scale, Type 1 Diabetes Mellitus, Eating Disorders, Insulin Restriction, Nursing Assessment, Psychometrics, Scale Validation, Behavioral Symptoms, Physiological Complications

Authors

The Diabulimia Knowledge Level Scale was conceptualized, designed, and psychometrically validated by a collaborative research team of academic nurse researchers specializing in adult health, internal medicine nursing, and clinical psychiatric nursing care:

• Serap Çetinkaya Özdemir, PhD, RN (Corresponding Author)
  Affiliation: Department of Internal Medicine Nursing, Faculty of Health Sciences, Sakarya University, Sakarya, Turkey.
  ORCID: 0000-0001-7706-4748
  Email: [email protected]
  Postal Address: Sakarya University, Faculty of Health Sciences, Department of Internal Medicine Nursing, 54050, Sakarya, Turkey.
• Vahide Semerci, PhD, RN
  Affiliation: Department of Nursing, Faculty of Health Sciences, Bayburt University, Bayburt, Turkey.
  ORCID: 0000-0003-3481-9500
• Ebru Sönmez Sarı, PhD, RN
  Affiliation: Department of Nursing, Faculty of Health Sciences, Bayburt University, Bayburt, Turkey.
  ORCID: 0000-0001-7337-4853

Purpose

The primary purpose of the Diabulimia Knowledge Level Scale (DKLS) is to quantify, evaluate, and monitor the diagnostic and clinical knowledge that registered nurses and allied healthcare professionals possess regarding diabulimia. Despite being recognized clinically as one of the deadliest manifestations of eating pathology, diabulimia currently lacks a discrete diagnostic code in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR), where it is generally coded under Other Specified Feeding or Eating Disorders (OSFED) or conceptualized alongside anorexia nervosa and bulimia nervosa as intentional medical non-adherence. This classification ambiguity frequently leads to systemic underdiagnosis, delayed medical triage, and inappropriate intervention protocols in acute inpatient and outpatient endocrinology settings.

Individuals diagnosed with Type 1 Diabetes Mellitus (T1DM) are forced to balance tight exogenous glycemic control with an unavoidable focus on dietary carbohydrates, caloric calculations, blood glucose monitoring, and body weight. The initiation of physiological insulin therapy frequently induces mild-to-moderate weight gain as fluid shifts normalize and anabolic energy storage resumes. In vulnerable populations—particularly adolescents, young adults, and females transitioning through puberty—this weight gain can trigger severe body image dissatisfaction, depressive affect, and disordered eating attitudes. Patients soon realize that reducing, under-dosing, or omitting insulin doses causes immediate osmotic diuresis and glucosuria, expelling hundreds to thousands of unabsorbed calories through the urine at the expense of severe intracellular starvation and metabolic derangement.

Because registered nurses serve on the front lines of diabetes self-management education, inpatient crisis intervention, and bedside therapeutic communication, their ability to identify nuanced behavioral indicators of diabulimia is critical. Clinical manifestations such as recurrent, unexplained episodes of diabetic ketoacidosis (DKA), persistent elevated glycated hemoglobin (HbA1c) despite claimed compliance, secrecy surrounding insulin administration, severe dietary restriction combined with binge episodes, and rapid weight fluctuations are frequently missed when nursing staff lack specific training. The DKLS was specifically engineered to address this pedagogical and clinical gap by fulfilling three fundamental functions:

• Diagnostic Gap Identification: Serving as an objective baseline assessment tool during institutional hospital onboarding, continuing professional education seminars, and undergraduate/graduate nursing curricula to identify knowledge deficits regarding intentional insulin manipulation.
• Curricular and Educational Intervention Assessment: Providing a sensitive pre-test and post-test measurement framework to evaluate the empirical efficacy of specialized eating disorder training modules in diabetes management settings.
• Translational Research and Policy: Facilitating multi-center correlational research examining whether higher nursing knowledge levels directly translate into earlier clinical detection, lowered incidence of recurrent DKA admissions, and improved interdisciplinary psychiatric referrals for diabetic patients.

Psychological Construct

The construct operationalized by the Diabulimia Knowledge Level Scale is a multi-layered cognitive domain encompassing the clinical, behavioral, psychological, and pathophysiological manifestations of eating disorder pathology in the context of Type 1 Diabetes Mellitus. Rather than measuring a respondent’s personal eating pathology or affective traits, the DKLS evaluates an externalized, professional comprehension of a complex syndromic phenomenon. The construct is articulated across four interrelated dimensions:

1. General Information

This dimension evaluates the healthcare provider’s fundamental epidemiological, demographic, and nosological comprehension of diabulimia. It captures the respondent’s recognition of the diagnostic boundaries separating T1DM management from non-diabetic eating disorders, the intentional nature of insulin manipulation, and the peak demographic vulnerability windows. Key aspects within this construct domain include:

• Recognition that diabulimia is fundamentally characterized by the deliberate, purposeful skipping or reduction of exogenous insulin doses rather than accidental oversight or cognitive failure (Item 1).
• Comprehension that the syndrome is etiologically confined to individuals diagnosed with Type 1 Diabetes who are pharmacologically dependent on daily exogenous insulin regimens (Item 2).
• Awareness that the peak developmental incidence occurs predominantly during adolescence and young adulthood, when physical development, psychosocial autonomy, and peer appearance comparison intersect (Item 3).

2. Behaviors

The behavioral subscale captures an understanding of the overt, compensatory, and maladaptive coping actions exhibited by individuals actively engaging in insulin omission. Beyond the omission of insulin itself, diabulimia frequently co-occurs with classical purging and restrictive eating behaviors seen in bulimia nervosa and anorexia nervosa. Nurses must recognize that patients utilize a multi-modal behavioral repertoire to alter their body weight and composition:

• Intentional dietary restriction, including purposeful meal skipping, fasting, or extreme carbohydrate avoidance aimed at manipulating glycemic response and promoting rapid weight loss (Item 4).
• Engagement in conventional compensatory purging mechanisms, such as the surreptitious misuse of laxatives, osmotic stimulants, or enemas to accelerate intestinal transit (Item 5).
• The deliberate induction of vomiting (self-induced emesis) following binge episodes or normal caloric meals as a secondary weight control strategy (Item 6).

3. Psychological Effects

This subscale assesses the clinician’s recognition of the profound psychological morbidity, psychiatric comorbidities, and cognitive distortions accompanying diabulimia. The interplay between intensive diabetes management and body dissatisfaction creates an intense psychological burden often termed “diabetes distress,” which cascades into comorbid affective and obsessive disorders. Knowledge in this domain requires understanding that individuals with diabulimia manifest:

• Pervasive body dissatisfaction, persistent body dysmorphic concerns, and an intense preoccupation with physical appearance, body shape, and societal beauty ideals (Items 7 and 8).
• Clinically significant depressive symptomatology, including anhedonia, profound feelings of worthlessness, hopelessness, and diminished self-worth or self-esteem (Items 9 and 10).
• Obsessive-compulsive cognitive patterns, characterized by intrusive, rigid rumination over blood glucose metrics, food composition, caloric calculations, and body measurements (Item 11).
• Cognitive impairment and executive dysfunction, manifested as marked deficits in concentration, sustained attention, and processing speed, exacerbated by fluctuating glycemic states (Item 12).
• Emotional lability, extreme mood instability, and affective dysregulation stemming from rapid neurochemical and metabolic fluctuations between hyperglycemia and euglycemia (Item 13).

4. Physiological Effects

The physiological dimension evaluates the clinician’s knowledge regarding the devastating, multi-system somatic consequences directly triggered by chronic, subacute, or episodic hyperglycemia and cellular energy starvation. Deprived of insulin, glucose cannot enter peripheral muscle or adipose tissues, leading to systemic catabolism, persistent osmotic diuresis, and progressive end-organ damage. Nurses must be able to link specific physical presentations to diabulimia pathology:

• Endocrine and developmental disruptions, such as severe growth retardation, stunted skeletal maturation, and delayed pubertal onset in pediatric and adolescent patients (Item 14).
• Generalized neuromuscular exhaustion, chronic fatigue, muscle wasting, and severe asthenia resulting from the uncoupling of cellular metabolism and prolonged proteolysis (Items 15 and 17).
• Immunological and urological vulnerability, specifically recurrent, treatment-resistant urinary tract infections (UTIs) and fungal vulvovaginitis driven by persistent, high-grade glycosuria (Item 16).
• Metabolic bone disease, characterized by premature bone mineral density loss, impaired osteoblast function, and early-onset osteoporosis (Item 18).
• Reproductive dysfunction in biological females, including secondary amenorrhea, oligomenorrhea, and chronic anovulation directly resulting from hypothalamic-pituitary-ovarian axis suppression under severe metabolic stress (Item 19).

Theoretical Framework

The theoretical architecture of the Diabulimia Knowledge Level Scale rests upon three converging conceptual frameworks: the Dual-Pathway Model of Eating Pathology in Chronic Illness, the Cognitive Behavioral Model of Disordered Eating, and the Health Belief and Competency Framework in Nursing Education.

1. The Dual-Pathway Model in Type 1 Diabetes

Originally formulated by Eric Stice (1994, 2001) in general populations and extended to pediatric and young-adult diabetology by Markowitz et al. (2010), the Dual-Pathway Model posits that eating pathology develops along two distinct yet reinforcing trajectories: dietary restriction and negative affect. When applied to diabetes, this model takes on unique biological and psychological dynamics. The standard medical treatment for T1DM requires constant, rigid behavioral self-regulation, including weighing food, calculating carbohydrate ratios, tracking caloric intake, and administering subcutaneous insulin injections.

This medical regimen mirrors the dietary preoccupation and cognitive restraint observed in classical eating disorders. However, while non-diabetic individuals must resort to extreme caloric starvation or difficult purging techniques (e.g., emesis, laxatives), individuals with T1DM have direct pharmacological access to a unique purge mechanism: insulin restriction. Omitting insulin prevents cellular glucose absorption, forcing the kidneys to excrete unmetabolized carbohydrates directly into the urine. This induces rapid weight reduction despite normal or increased caloric intake. The DKLS draws heavily upon this conceptual framework, structuring items to assess whether healthcare professionals recognize this dual mechanism of metabolic purging alongside conventional restrictive behaviors.

2. Cognitive Behavioral Theory and Affective Dysregulation

The cognitive behavioral model developed by Christopher Fairburn emphasizes that the core psychopathology of eating disorders is the over-evaluation of shape, weight, and their control. In patients with diabulimia, this cognitive schema is intensified by the physiological weight gain that naturally follows the establishment of euglycemia. Patients experience profound body dissatisfaction, leading to low self-esteem, clinical depression, and obsessive checking behaviors. The resulting affective distress fuels further insulin avoidance, creating a self-reinforcing vicious cycle. The DKLS incorporates this theoretical cycle into its Psychological Effects subscale, testing whether providers recognize that insulin skipping is not merely casual non-compliance, but a deeply rooted cognitive and emotional coping mechanism.

3. Professional Nursing Knowledge and Clinical Competency Theory

The measurement philosophy of the DKLS is informed by Patricia Benner’s Novice to Expert Model of clinical nursing competence and educational taxonomy principles. According to Benner, transitioning from an inexperienced clinician to a proficient or expert practitioner requires moving beyond rule-based knowledge to holistic situational recognition and perceptual acuity. Frontline nurses cannot assess or intervene in clinical phenomena they cannot cognitively categorize. By establishing a dedicated knowledge assessment instrument, the DKLS provides the empirical foundation necessary to evaluate the cognitive readiness of nurses, operationalizing knowledge across informational, behavioral, affective, and somatic domains.

Validity

The structural, content, and construct validity of the Diabulimia Knowledge Level Scale was evaluated using rigorous psychometric testing protocols during its initial validation study (Çetinkaya Özdemir et al., 2023).

Content Validity

The initial item pool was generated through an exhaustive synthesis of international clinical literature, psychiatric guidelines, and established diabetes non-adherence studies (e.g., Hoffmann, 2019; Markowitz & Friedman, 2010; Ruth-Sahd & Power, 2009). Following the draft formulation, content validity was formally scrutinized by an expert panel of multidisciplinary clinicians and academicians specializing in internal medicine nursing, endocrinology, and psychiatric healthcare. The panel evaluated each item across three qualitative criteria: relevance/relationship to construct, simplicity of phrasing, and linguistic clarity, using a standardized 4-point rating index.

The computed Content Validity Index (CVI) values demonstrated outstanding consensus across the expert panel:

• Relevance/Relationship: Scale-level CVI = 0.97
• Simplicity: Scale-level CVI = 0.92
• Clarity: Scale-level CVI = 0.95

These values comfortably exceed the conservative threshold of 0.80 established by psychometricians (Lynn, 1986; Polit & Beck, 2006), verifying that the operationalized item statements accurately reflect the target domain of diabulimia knowledge without extraneous or confounding content.

Construct and Structural Validity

Construct validity was evaluated through a sequential exploratory and confirmatory factor analytic framework. The exploratory factor analysis supported a four-dimensional theoretical architecture that accounted for 68.53% of the total cumulative variance. The construct validity was confirmed by the logical grouping of items: items assessing diagnostic definitions and demographic risk loaded cleanly onto General Information; behavioral markers, laxative use, and vomiting loaded onto Behaviors; self-esteem, depression, and mood lability loaded onto Psychological Effects; and physical sequelae such as osteoporosis, growth deficiency, and recurrent infections loaded onto Physiological Effects.

The CFA confirmed that each observed indicator exhibited statistically significant path coefficients onto its respective latent factor, supporting the theoretical dimensionality of the construct. The absence of cross-factor contamination in the final 19-item set confirms adequate discriminant and convergent validity among the four sub-dimensions.

Reliability

The reliability of the Diabulimia Knowledge Level Scale was examined using measures of internal consistency and temporal stability (test-retest reliability).

Internal Consistency

Internal consistency was calculated using Cronbach’s alpha coefficients across the total instrument and for each of the four identified subscales. In the primary validation sample of Turkish registered nurses, the subscale and overall scale alpha coefficients ranged from .70 to .92:

• Subscale Cronbach’s alphas: Ranged from an acceptable .70 to an excellent .92.
• Overall Total Scale Alpha: Demonstrated high internal consistency, indicating that the items share strong common variance while assessing distinct facets of the construct.

According to established psychometric standards (Nunnally & Bernstein, 1994), reliability coefficients between .70 and .80 are deemed acceptable for early-stage or screening instruments, while values exceeding .80 reflect strong reliability suitable for both research and clinical benchmarking.

Test-Retest Temporal Stability

To establish that the DKLS measures stable knowledge structures rather than transient, state-dependent responses, temporal stability was evaluated using a test-retest design across a three-week interval in a randomly selected sub-sample of the nursing cohort. The resulting correlation coefficients and paired-sample analyses yielded the following findings:

• Bivariate Correlations: Pearson correlation coefficients (r) between the initial baseline administration (Time 1) and the retest administration (Time 2) ranged from 0.44 to 0.76 across the total score and individual subscale dimensions, reflecting moderate to high temporal consistency.
• Paired-Samples Mean Comparisons: A paired-samples t-test was conducted to verify whether systematic mean score inflation or deflation occurred across the three-week window. The analysis revealed no statistically significant differences between Time 1 and Time 2 scores (p = .154).

The absence of significant mean drift confirms that the instrument possesses solid temporal stability and is resistant to short-term recall bias or random measurement fluctuation when no educational intervention has taken place.

Factor Analysis

The dimensional structure of the Diabulimia Knowledge Level Scale was established through a two-stage analytic process involving Exploratory Factor Analysis (EFA) and subsequent Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

The initial pool of candidate items was subjected to Principal Component Analysis (PCA) with orthogonal (varimax) rotation. The suitability of the data for factor extraction was verified using the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett’s Test of Sphericity, both of which satisfied mathematical criteria for factorability.

During the exploratory factor analytic progression, rigorous item reduction rules were applied to optimize scale parsimony and structural clarity:

• Low Factor Loadings: Nine items (Items 6, 7, 13, 15, 16, 25, 35, 36, and 38 from the preliminary pool) were removed due to failing to achieve the minimum factor loading threshold of 0.40 on any latent factor.
• Cross-Loading Items: Two items (Items 8 and 24) were systematically removed because they demonstrated significant cross-loadings across multiple factors with insufficient loading differentials (< 0.10 difference).
• Under-Identified Factors: Eight items (Items 4, 5, 9, 14, 26, 27, 28, and 37) loaded onto unstable components containing two or fewer items. To maintain factor stability and prevent under-identification, these items were eliminated.

The final extraction yielded a robust four-factor solution comprising 19 items that explained a cumulative 68.53% of the total variance. The factors were designated as follows:

• Factor 1 (Psychological Effects): Encompasses items measuring mood instability, concentration problems, depressive symptoms, obsessive behaviors, and altered body image.
• Factor 2 (Physiological Effects): Encompasses somatic manifestations, including growth failure, severe fatigue, recurrent UTIs, osteoporosis, and menstrual irregularity.
• Factor 3 (Behaviors): Captures active behavioral strategies, including meal skipping, laxative abuse, and self-induced vomiting.
• Factor 4 (General Information): Encompasses foundational knowledge regarding insulin omission, target diagnostic group (T1DM), and adolescent epidemiology.

Confirmatory Factor Analysis (CFA)

To confirm the empirical tenability of the four-factor model derived from EFA, Confirmatory Factor Analysis was conducted. Model parameters were estimated using maximum likelihood techniques. The resulting goodness-of-fit indices were as follows:

• Chi-Square / Degrees of Freedom Ratio: χ² = 448.853, df = 146 (p < .001); χ²/df = 3.074. Ratios between 2.0 and 5.0 are generally considered acceptable in initial psychometric scale validations of complex psychological phenomena.
• Comparative Fit Index (CFI): 0.87
• Incremental Fit Index (IFI): 0.87
• Tucker-Lewis Index (TLI): 0.85
• Normed Fit Index (NFI): 0.82
• Relative Fit Index (RFI): 0.79
• Goodness-of-Fit Index (GFI): 0.78
• Root Mean Square Residual (RMR): 0.070
• Root Mean Square Error of Approximation (RMSEA): 0.104

While the CFI and TLI indices fell slightly below the conservative 0.90 cut-off and the RMSEA remained in the moderate-to-elevated range (0.104), these values reflect real-world clinical survey dynamics among heterogeneous healthcare professionals. The indices, combined with an explained variance of nearly 70% and strong factor loadings across all 19 indicators, substantiate the structural validity of the four-factor model for clinical assessment and educational research.

Instrument / Measurement Tool

• Instrument Name: Diabulimia Knowledge Level Scale (DKLS)
• Original Authors: Serap Çetinkaya Özdemir, Vahide Semerci, and Ebru Sönmez Sarı (2023)
• Test Type: Original objective knowledge inventory / educational assessment questionnaire
• Target Population: Adult healthcare providers (specifically registered nurses, nursing students, diabetes nurse educators, endocrinologists, and medical-surgical clinicians; age 18 and older)
• Administration Mode: Self-administered (paper-and-pencil or computerized survey format)
• Item Count: 19 items
• Dimensional Structure: Four subscales
  • General Information: 3 items (Items 1, 2, 3)
  • Behaviors: 3 items (Items 4, 5, 6)
  • Psychological Effects: 7 items (Items 7, 8, 9, 10, 11, 12, 13)
  • Physiological Effects: 6 items (Items 14, 15, 16, 17, 18, 19)
• Response Scale: The scale is structured as a 5-point Likert-type scale, with response options ranging from 1 = Strongly disagree, 2 = Disagree, 3 = Undecided, 4 = Agree, to 5 = Strongly agree.
• Scoring Guidelines:
  • All 19 items are scored in a direct positive direction (1 = Strongly disagree, 2 = Disagree, 3 = Undecided, 4 = Agree, 5 = Strongly agree).
  • There are no reverse-coded items in the validated 19-item version.
  • Subscale Scores: Calculated by summing the item scores within each respective domain (General Information: range 3–15; Behaviors: range 3–15; Psychological Effects: range 7–35; Physiological Effects: range 6–30).
  • Total Score: Calculated by summing all 19 items, yielding a composite score ranging from 19 to 95.
  • Interpretation: Higher aggregate and subscale scores denote higher levels of clinical knowledge, recognition capacity, and awareness regarding diabulimia. Lower scores indicate specific educational deficits requiring targeted instructional interventions.

Permissions & Fee and Test Year

The Diabulimia Knowledge Level Scale was developed and validated in 2023 and published in the peer-reviewed journal Archives of Psychiatric Nursing. In accordance with academic dissemination standards, the authors have made the scale accessible for non-commercial, scholarly research, hospital clinical auditing, and educational training purposes without requiring licensing fees:

• Test Year: 2023
• Commercial Use: Prohibited without explicit, formal written authorization from the authors and the copyright-holding publisher.
• Academic and Educational Use: Free of charge for research, university curricula, and clinical nursing training initiatives.
• Permissions & Contact: Researchers seeking to reproduce, translate, or adapt the scale for cross-cultural studies should contact the corresponding author: Dr. Serap Çetinkaya Özdemir ([email protected]), Sakarya University, Faculty of Health Sciences, Department of Internal Medicine Nursing, Sakarya, Turkey.

References

• Çetinkaya Özdemir, S., Semerci, V., & Sönmez Sarı, E. (2023). Development of the Diabulimia Knowledge Level Scale. Archives of Psychiatric Nursing, 46, 139–145. https://doi.org/10.1016/j.apnu.2023.09.003
• 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
• Hoffmann, M. (2019). Diabulimia: The clinical presentation, risks, and nursing care of a comorbid eating disorder and type 1 diabetes mellitus. Journal of Pediatric Nursing, 48, 120–123. https://doi.org/10.1016/j.pedn.2019.06.012
• Lynn, M. R. (1986). Determination and quantification of content validity. Nursing Research, 35(6), 382–386. https://doi.org/10.1097/00006199-198611000-00017
• Markowitz, L. A., & Friedman, J. D. (2010). Eating disorders in adolescents and young adults with type 1 diabetes: A review and framework for understanding the presentation and treatment. Diabetes Spectrum, 23(3), 138–144. https://doi.org/10.2337/diaspect.23.3.138
• Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). McGraw-Hill.
• Polit, D. F., & Beck, C. T. (2006). The content validity index: Are you sure you know what’s being reported? Critique and recommendations. Research in Nursing & Health, 29(5), 489–497. https://doi.org/10.1002/nur.20147
• Ruth-Sahd, L. A., & Power, S. (2009). Diabulimia: A dual diagnosis of diabetes and eating disorders. Dimensions of Critical Care Nursing, 28(5), 209–213. https://doi.org/10.1097/DCC.0b013e3181ac4c26
• 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

Items of the Scale