Shana B Traina, Susan D Mathias, Hilary H Colwell, Ross D Crosby, Charles Abraham

This study assessed measurement properties of the 17-item Diabetes Intention, Attitude, and Behavior Questionnaire (DIAB-Q), which measures intention to engage in self-care behaviors, including following a diabetes diet and engaging in appropriate physical activity. The DIAB-Q includes questions based on the Theory of Planned Behavior. Items were developed using published literature, input from health care professionals, and qualitative research findings in patients with and without type 2 diabetes mellitus (T2DM). In Stage I of the study, 23 adults with T2DM were interviewed to evaluate the content and clarity of the DIAB-Q. In Stage II 1,015 individuals with T2DM completed the DIAB-Q and supplemental questionnaires, including the Short Form-36 acute (SF-36), section III of the Multidimensional Diabetes Questionnaire, the Summary of Diabetes Self-Care Activities questionnaire, and self-administered items relevant to the treatment and management of T2DM (eg, blood pressure and glycated hemoglobin [HbA1c]) at baseline and 3-7 days later. Once the DIAB-Q scale structure was determined, its test-retest reliability, construct validity, and known-groups validity were evaluated, and minimal clinically important change was estimated. In Stage I, the 23 respondents surveyed generally reported that the DIAB-Q was clear and comprehensive and endorsed questions as relevant to their intentions to engage in diabetes-related self-care activities. Most subjects in Stage II were male, Caucasian, and married. Mean age was 63 years. Factor analysis revealed six psychological constructs (Behavior, Planning, Intention, Perceived Behavioral Control, Attitude, and Subjective Norm). Test-retest reliability was acceptable (≥0.70) for all scales, except Perceived Behavioral Control. Construct validity was demonstrated based on correlations with diabetes-specific items/scales and the SF-36. Known-groups validity was confirmed for Behavior, Planning, and Intention when respondents were categorized into groups that differed based on body mass index, disease severity, and HbA1c. Item scores were transformed to a 100-point scale, and minimal clinically important change estimates ranged from 6-11 points, representing the change that would be considered important to a respondent. The DIAB-Q is a brief, psychometrically sound, patient-reported outcome that can be used among individuals with T2DM to evaluate intention to engage in self-care behaviors.