Crush Diabetes Knowledge, Attitudes, and Beliefs Survey (CDKA)

1. Abstract

The Crush Diabetes Knowledge, Attitudes, and Beliefs Survey (CDKA) is a standardized, psychometrically validated psychometric measurement tool designed by Mokhtari et al. (2023) to assess diabetes-related health literacy, psychological perceptions, personal health beliefs, and behavioral intentions among middle school students aged 12 to 14 years. Emerging alongside escalating global pediatric obesity and adolescent type 2 diabetes mellitus incidence, the CDKA provides public health researchers, pediatric clinicians, and school-based health educators with an empirically robust diagnostic instrument to measure the direct cognitive and psychosocial determinants of preventive health behaviors. Comprising 25 items structured across two broad conceptual domains—objective knowledge (12 items) and affective attitudes, beliefs, and behavioral intentions (13 items)—the instrument assesses dimensions including diabetes etiology, symptom presentation, insulin physiology, risk factors, personal vulnerability, self-efficacy, and behavioral intent to change diet, physical activity, and sedentary patterns.

The scale was developed by a panel of registered dietitians and clinical experts specializing in adolescent endocrinology and evaluated across diverse educational cohorts. Psychometric evaluations confirm high content validity, yielding an overall scale-level Content Validity Index (S-CVI) of 0.95. Reliability evaluations demonstrate acceptable to strong internal consistency across domains, with a Cronbach’s alpha coefficient of 0.70 for the cognitive knowledge subscale and 0.70 for the combined attitudes, beliefs, and intent to change subscale. The instrument demonstrates test-retest reliability across repeated administrations, exhibiting intraclass person-average coefficients of $R_{kf} = 0.80$ for the knowledge section and $R_{kf} = 0.70$ for the attitudes, beliefs, and intent to change section. Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) validate a four-factor dimensionality for each section, demonstrating acceptable model fit indices ($chi^2 = 79.4, p < 0.001, ext{RMSEA} = 0.04$ for knowledge; $chi^2 = 165.7, p < 0.001, ext{RMSEA} = 0.07$ for attitudes, beliefs, and intent). Consequently, the CDKA serves as an empirically sound assessment tool for evaluating school-based curriculum outcomes, designing pediatric lifestyle interventions, and diagnosing gaps in adolescent chronic disease literacy.

2. Keywords

Crush Diabetes Knowledge Attitudes and Beliefs Survey, CDKA, adolescent diabetes literacy, pediatric type 2 diabetes, health attitudes, health beliefs, intent to change, psychometrics, school health education, preventive lifestyle behaviors, chronic disease awareness, health behavior change

3. Authors

The Crush Diabetes Knowledge, Attitudes, and Beliefs Survey was developed and validated by a multidisciplinary team of clinical nutritionists, health scientists, and behavioral researchers from the University of Utah:

• Pari Mokhtari, PhD, RDN (ORCID: 0000-0003-0451-1580) — Department of Nutrition and Integrative Physiology, College of Health, University of Utah, Salt Lake City, Utah, United States. Email: [email protected].
• Kristine C. Jordan, PhD, RDN — Department of Nutrition and Integrative Physiology, College of Health, University of Utah, Salt Lake City, Utah, United States.
• Andrew Wilson, PhD (ORCID: 0000-0003-3679-9232) — College of Nursing, University of Utah, Salt Lake City, Utah, United States.
• Brittany Sisco-Taylor, MS (ORCID: 0000-0001-7028-2953) — Department of Population Health Sciences, School of Medicine, University of Utah, Salt Lake City, Utah, United States.
• Julie M. Metos, PhD, RDN — Department of Nutrition and Integrative Physiology, College of Health, University of Utah, Salt Lake City, Utah, United States.

4. Purpose

The primary clinical and epidemiological purpose of the Crush Diabetes Knowledge, Attitudes, and Beliefs Survey (CDKA) is to provide an empirically validated instrument capable of measuring adolescent comprehension of diabetes pathophysiology, nutritional risk factors, self-efficacy, and behavioral intention to adopt preventive health practices. Historically, the onset of type 2 diabetes mellitus (T2D) was viewed almost exclusively as an adult metabolic disease. However, shifts in adolescent dietary patterns, marked by the consumption of ultra-processed food and sugar-sweetened beverages, coupled with sedentary behaviors driven by recreational screen time, have triggered an increase in adolescent T2D diagnoses. In response, public health organizations have shifted toward early preventive interventions within school settings. Despite this, prior to the development of the CDKA, few psychometrically validated tools existed to capture both the cognitive knowledge bases and the affective motivational beliefs specific to early adolescents aged 12 to 14 years.

In clinical and educational practice, the CDKA fulfills multiple assessment functions. First, it acts as a diagnostic needs assessment tool for school nurses, public health practitioners, and physical education instructors to identify community-specific health literacy deficiencies before deploying curriculum modules. For instance, adolescents frequently harbor fundamental misconceptions regarding the distinction between autoimmune type 1 diabetes and metabolic type 2 diabetes, often conflating their etiologies or assuming that youth confers immunity to metabolic dysfunction. Second, the CDKA serves as an outcome evaluation instrument within longitudinal research, randomized controlled trials (RCTs), and community-based participatory research projects aiming to quantify the effectiveness of lifestyle modification initiatives (such as diabetes prevention programs, school garden initiatives, and enhanced physical education).

From a theoretical perspective, the CDKA acknowledges that cognitive knowledge alone is often insufficient to induce enduring behavioral modifications. While an adolescent may comprehend that physical inactivity and elevated sugar consumption exacerbate metabolic risk, their actual health behaviors are heavily mediated by subjective perceived vulnerability, outcome expectancies, social barriers, and personal self-efficacy. By pairing a cognitive knowledge assessment with psychometric indicators of personal attitudes, perceived severity, and concrete behavioral intent across a defined timeframe, the CDKA bridges the gap between static health literacy and active behavioral change paradigms. Consequently, the instrument offers researchers an integrated profile of an adolescent’s readiness to adopt sustained, health-promoting lifestyle habits.

5. Psychological Construct

The Crush Diabetes Knowledge, Attitudes, and Beliefs Survey (CDKA) evaluates a multidimensional construct comprised of cognitive, affective, and intentional domains. Rather than treating health literacy as a simple binary measure of factual recall, the CDKA operationalizes diabetes prevention readiness as a dynamic interplay of factual knowledge, subjective psychological beliefs, perceived behavioral barriers, and behavioral commitment. These dimensions are divided across two major survey sections encompassing multiple subscales:

1. Objective Diabetes Knowledge Domain

The cognitive knowledge section of the CDKA assesses an adolescent’s foundational comprehension of endocrine physiology, systemic etiology, clinical symptomatology, and modifiable lifestyle risks. This domain captures three specific sub-constructs:

• Insulin Physiology and Endocrine Function: Measures an individual’s comprehension of the biological mechanisms governing glycemic control, specifically identifying the pancreas as the insulin-producing organ and understanding insulin’s function in facilitating cellular glucose uptake. Example items evaluate whether the student understands that diabetes occurs when the body fails to make or properly use insulin.
• General Disease Definition and Symptom Awareness: Assesses clarity regarding the clinical characteristics of the disease, including recognizing common symptoms (e.g., polyuria, polydipsia, fatigue) and understanding that diabetes is characterized by elevated blood glucose rather than hypoglycemia. Crucially, it identifies misconceptions regarding age susceptibility, demonstrating that youth does not confer immunity to metabolic dysregulation.
• Prevention, Etiology, and Modifiable Risk Factors: Measures comprehension of the distinction between the non-preventable autoimmune etiology of type 1 diabetes and the modifiable behavioral, dietary, and excess adiposity risks associated with type 2 diabetes. This subscale also evaluates an understanding of the protective mechanisms of daily moderate-to-vigorous physical activity (MVPA) and balanced nutrition, as well as the non-modifiable contribution of genetic family history.

2. Attitudes, Beliefs, and Intent to Change Domain

The affective and behavioral domain of the CDKA assesses the internal cognitive appraisal mechanisms that translate knowledge into preventive behavior. Grounded in health psychology, this section captures several sub-constructs:

• Perceived Severity and Disease Burden: Evaluates the adolescent’s perception of the clinical and psychosocial severity of living with diabetes. It assesses the realization that diabetes management requires continuous adherence, affecting emotional stability and daily routines, while challenging false beliefs that diabetes is trivial or that type 2 diabetes carries fewer long-term complications than type 1 diabetes.
• Outcome Expectancies and Health Values: Measures personal beliefs regarding the efficacy of lifestyle modifications. This captures whether the respondent believes that personal dietary changes (e.g., reducing sugar-sweetened beverages) and regular physical exercise can directly prevent or delay the onset of metabolic disease and enhance daily energy levels.
• Self-Efficacy and Perceived Behavioral Control: Reflects the respondent’s confidence in executing health-promoting behaviors within everyday social and environmental contexts. This includes self-efficacy in selecting nutritious foods in the presence of peer pressure, negotiating environments where healthy options are limited, and scheduling regular exercise amidst academic and family responsibilities.
• Intent to Change Behavior: Assesses actionable, proximate behavioral intentions across target domains over specific timeframes (e.g., “in the next week”). Measured variables include clear commitments to reduce sugar-sweetened beverage and soda consumption, increase moderate-to-vigorous physical activity, reduce recreational screen time, consume whole foods and vegetables, and model healthy habits for peers and family members.

6. Theoretical Framework

The architecture of the Crush Diabetes Knowledge, Attitudes, and Beliefs Survey is grounded in social-cognitive and behavioral decision-making models. Chief among these theoretical frameworks are the Health Belief Model (HBM), formulated by Becker and Rosenstock, and Albert Bandura’s Social Cognitive Theory (SCT), supplemented by the behavioral sequencing principles of the Theory of Planned Behavior (TPB) developed by Icek Ajzen.

The Health Belief Model posits that individuals will not engage in preventive health actions unless they perceive themselves to be personally susceptible to a condition (perceived susceptibility), believe the condition carries significant clinical or social consequences (perceived severity), and conclude that adopting a specific preventive behavior yields positive benefits (perceived benefits) that outweigh physical, psychological, or logistical obstacles (perceived barriers). The CDKA integrates these constructs directly: items measuring family history risk and prediabetes awareness capture perceived susceptibility; items highlighting the persistent management challenges of diabetes reflect perceived severity; items asserting that exercise and balanced nutrition protect metabolic health measure perceived benefits; and items addressing food availability at home or school evaluate perceived barriers.

Bandura’s Social Cognitive Theory contributes the construct of self-efficacy—an individual’s confidence in their ability to execute behaviors necessary to produce specific performance attainments. In pediatric and early adolescent health psychology, self-efficacy is a strong predictor of lifestyle change. Adolescents frequently possess theoretical health knowledge but lack the self-regulatory efficacy to decline hyper-palatable snacks among peers or establish personal exercise routines without parental structuring. The CDKA explicitly integrates self-efficacy items that probe an adolescent’s perceived capability to maintain balanced nutritional intake and regular physical activity across varied social settings.

Finally, the Theory of Planned Behavior provides the theoretical link between subjective attitudes, perceived normative pressure, behavioral control, and actual clinical practice through the central mechanism of behavioral intention. According to the TPB, behavioral intention is the most proximate antecedent to the actual execution of a given health practice. By incorporating targeted “intent to change” metrics covering short-term time horizons (e.g., planning to drink water instead of soda, intending to increase physical activity, reducing screen time), the CDKA creates a validated psychometric pathway tracking how factual knowledge informs subjective beliefs, how beliefs foster health self-efficacy, and how self-efficacy resolves into concrete, measurable intentions to avoid chronic lifestyle-related diseases.

7. Validity

Psychometric evaluation of the CDKA demonstrates robust validity across multiple testing phases during its initial validation by Mokhtari et al. (2023) at the University of Utah:

Content and Face Validity

Initial item generation was conducted by an expert panel of five registered dietitian nutritionists (RDNs) and pediatric researchers with clinical experience in adolescent diabetes, childhood obesity, and health education curriculum design. Items were developed to map against the National Health Education Standards and clinical pediatric endocrinology guidelines. An external review panel of adolescent medicine experts, dietitians, and pediatric health specialists evaluated each item for developmental appropriateness, linguistic clarity, reading level suitability (assuring readability at a 6th-to-8th grade reading level), and clinical accuracy. Quantitative evaluation yielded an overall scale-level Content Validity Index (S-CVI) of 0.95, exceeding the conventional academic threshold of 0.80 for exceptional instrument content validity. Face validity was qualitatively established via cognitive pre-testing and pilot debriefings with middle school students, ensuring that item wording, syntax, and conceptual choices were interpreted accurately by adolescents aged 12 to 14 without semantic ambiguity.

Construct and Structural Validity

Construct validity was evaluated by examining whether the instrument could differentiate between distinct cognitive knowledge constructs and psychological attitudes, as well as testing internal structural integrity through exploratory and confirmatory factor analyses. The separation of factual true/false/unsure knowledge questions from continuous Likert-scale attitudinal and intentional metrics preserved construct boundaries, preventing cognitive recall biases from contaminating affective ratings. Confirmatory factor analysis verified that the latent dimensions underlying both the 12-item cognitive structure and the 13-item psychosocial structure fit the proposed theoretical models well, exhibiting root mean square errors of approximation (RMSEA) within empirically acceptable parameters ($ ext{RMSEA} = 0.04$ for knowledge; $ ext{RMSEA} = 0.07$ for attitudes, beliefs, and intent).

Convergent, Discriminant, and Known-Groups Validity

Convergent validity was evidenced by factor loadings across components; all retained items exhibited factor loadings exceeding 0.40, with primary loadings on designated subscales reaching up to 0.80 in the knowledge section and 0.90 in the attitudes, beliefs, and intent section. Discriminant validity was evidenced by the low to moderate inter-factor correlations between the objective knowledge factor scores and subjective attitudes/intent scores, confirming that while factual knowledge provides a foundation for attitudes, it operates as a statistically distinct psychological dimension. Furthermore, during intervention pilot testing, the CDKA demonstrated known-groups and educational responsiveness validity: cohorts participating in structured adolescent diabetes prevention programs exhibited statistically significant, measurable increases in both objective knowledge scores and positive behavioral intentions compared to baseline pre-intervention assessments.

8. Reliability

The reliability of the Crush Diabetes Knowledge, Attitudes, and Beliefs Survey was established using classical test theory methodologies, evaluating both internal consistency and temporal stability (test-retest reliability) among middle school student populations.

Internal Consistency

Internal consistency reflects the extent to which items within an instrument or subscale measure the same underlying construct. Given the conceptual distinction between objective factual recall and subjective psychological perspectives, internal consistency was evaluated separately for the primary sections of the instrument:

• Knowledge Section (12 items): Yielded a Cronbach’s alpha ($lpha$) coefficient of 0.70. In educational measurement involving dichotomous cognitive test items with diverse categorical content (covering anatomy, dietary sources, symptom recognition, and disease types), a Cronbach’s alpha of 0.70 satisfies the accepted standard for academic and school-based assessment tools.
• Attitudes, Beliefs, and Intent to Change Section (13 items): Yielded a Cronbach’s alpha ($lpha$) coefficient of 0.70. This indicates adequate internal consistency across items assessing perceived severity, self-efficacy, health values, and future behavioral commitments.

Test-Retest Reliability and Temporal Stability

To establish whether the CDKA produces stable measurement over time in the absence of targeted educational interventions, temporal stability was evaluated by administering the survey to a sample of middle school students across a designated retest interval. Intraclass person-average reliability coefficients ($R_{kf}$) were computed:

• Knowledge Section: Exhibited an intraclass correlation coefficient of $R_{kf} = 0.80$, denoting high temporal stability and confirming that adolescent baseline factual recall remains consistent across repeated administrations without intervening instruction.
• Attitudes, Beliefs, and Intent to Change Section: Exhibited an intraclass correlation coefficient of $R_{kf} = 0.70$, demonstrating acceptable temporal stability for affective and motivational metrics, which naturally exhibit greater susceptibility to contextual and environmental fluctuations than factual knowledge.

9. Factor Analysis

The underlying dimensionality and latent variable structure of the CDKA were evaluated through a sequence of Exploratory Factor Analyses (EFA) followed by Confirmatory Factor Analyses (CFA), analyzing the cognitive knowledge subscale and the psychosocial subscale independently.

Exploratory Factor Analysis (EFA)

For both survey sections, sample correlation matrices were evaluated for factorability using Bartlett’s Test of Sphericity ($p < 0.001$) and the Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy. Scree plot analyses, combined with the Kaiser-Guttman criterion (retaining components with eigenvalues$> 1.0$), identified four distinct latent components within each section:

• Knowledge Components: The four extracted components mapped onto (1) Insulin Physiology, (2) General Disease Information and Symptoms, (3) Prevention and Modifiable Lifestyle Factors, and (4) Untrue Definitions / Common Misconceptions. Item factor loadings within the knowledge section demonstrated strong convergent validity, with all retained items exhibiting loadings between 0.50 and 0.80.
• Attitudes, Beliefs, and Intent Components: The four extracted components for this section mapped onto (1) Intent to Change, (2) Attitudes and Beliefs: Lifestyle and Prevention, (3) Attitudes and Beliefs: Psychosocial Barriers, and (4) Attitudes and Beliefs: Emotional and Disease Burden. Factor loadings across this 13-item section were robust, ranging between 0.40 and 0.90, with all retained items surpassing the standard structural retention cutoff of 0.40.

Confirmatory Factor Analysis (CFA)

To verify the four-factor structural models derived from the EFA, Confirmatory Factor Analyses were executed. Model fit was evaluated using multiple structural equation modeling indices, including the Chi-square goodness-of-fit test ($\chi^2$), degrees of freedom ($df$), and the Root Mean Square Error of Approximation (RMSEA):

• Knowledge Model: The four-factor CFA model demonstrated an acceptable structural fit to the observed data: $chi^2 = 79.4, p < 0.001$, with an$ ext{RMSEA} = 0.04$. The RMSEA value of 0.04 falls well below the conservative 0.05 threshold, indicating a close fit between the theoretical four-factor model and the observed adolescent response patterns.
• Attitudes, Beliefs, and Intent to Change Model: The four-factor CFA model for this section also exhibited an acceptable structural fit: $chi^2 = 165.7, p < 0.001$, with an$ ext{RMSEA} = 0.07$. An RMSEA value between 0.05 and 0.08 represents acceptable fit in psychosocial structural modeling involving early adolescent self-report measures.

10. Instrument / Measurement Tool

• Test Type: Original psychometric self-report survey combining an objective knowledge questionnaire with an affective Likert rating scale.
• Total Item Count: 25 items across two main conceptual sections: Knowledge (items 1–12) and Attitudes, Beliefs, and Intent to Change (items 13–25).
• Target Population: Human; School-age children and early adolescents (specifically middle school students aged 12–14 years; Grades 6–8; Male and Female).
• Available Language: English.
• Administration Format: Paper-and-pencil or interactive digital/computerized survey; self-administered in classroom, clinic, or community research settings.
• Completion Time: Approximately 10 to 15 minutes.
• Response Formats:
  • Knowledge Section: Multiple choice (with one correct answer) or True / False / Don’t know. The inclusion of an explicit “I do not know” option prevents respondents from guessing, preserving the validity of the cognitive measurement.
  • Attitudes, Beliefs, and Intent to Change Section: 5-point Likert scale (1 = Strongly Disagree, 2 = Disagree, 3 = Neither Agree nor Disagree / Neutral, 4 = Agree, 5 = Strongly Agree).
• Scoring Rules:
  • Knowledge Section: Dichotomously scored. Each correct response is awarded 1 point. Each incorrect response and each “I do not know” response is awarded 0 points. Total knowledge raw scores range from 0 to 12 points, with higher scores reflecting greater objective diabetes and lifestyle literacy.
  • Attitudes, Beliefs, and Intent Section: Scored continuously from 1 to 5 for each item. Positively keyed items are scored directly (1 = Strongly Disagree to 5 = Strongly Agree), where higher scores indicate more positive health beliefs, stronger health self-efficacy, elevated perceived risk awareness, and greater intention to adopt preventive health behaviors. Negatively phrased/reverse items are inverted prior to composite calculation.
  • Subscales: The survey generates profile scores across specific subdomains: (1) Diabetes Knowledge, (2) Exercise and Healthy Lifestyle Knowledge, (3) Risk Awareness, (4) Attitudes and Beliefs, and (5) Intent to Change Behavior.

11. Permissions & Fee and Test Year

• Test Year: 2023.
• Copyright / Ownership: Pari Mokhtari, Kristine C. Jordan, Andrew Wilson, Brittany Sisco-Taylor, and Julie M. Metos; published by Elsevier Inc. on behalf of the Society for Nutrition Education and Behavior.
• Commercial Status: Non-commercial.
• Fee: Free of charge ($0) for non-commercial educational, academic research, public health, and school health assessment purposes.
• Permissions & Licensing: The CDKA may be utilized for academic research, health education curriculum assessment, and teaching. Researchers and educators wishing to reproduce, adapt, or deploy the instrument should cite the original validation study published in the Journal of Nutrition Education and Behavior (Mokhtari et al., 2023). Formal permission for widespread commercial redistribution or inclusion within commercial software must be requested through Elsevier’s permissions clearinghouse.

12. References

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall, Inc.

Bandura, A. (2004). Health promotion by social cognitive means. Health Education & Behavior, 31(2), 143–164. https://doi.org/10.1177/1090198104263660

Mokhtari, P., Jordan, K. C., Wilson, A., Sisco-Taylor, B., & Metos, J. M. (2023). Development and validation of a diabetes questionnaire for middle school students. Journal of Nutrition Education and Behavior, 55(2), 135–144. https://doi.org/10.1016/j.jneb.2022.11.004

Rosenstock, I. M., Strecher, V. J., & Becker, M. H. (1988). Social learning theory and the Health Belief Model. Health Education Quarterly, 15(2), 175–183. https://doi.org/10.1177/109019818801500203

13. Items of the Scale