21-LB: Development and Validation of the Barriers and Supports Evaluation for Working-Age Adults with Type 1 Diabetes

• Published in: Diabetes (New York, N.Y.) Vol. 70; no. Supplement_1
• Main Authors: COX, ELIZABETH D., PLANALP, ELIZABETH M., KLIEMS, HARALD, PALTA, MARI, LECAIRE, TAMARA J., CHEWNING, BETTY A.
• Format: Journal Article
• Language: English
• Published: New York American Diabetes Association 01-06-2021
• Subjects: Cognition & reasoning; Cognitive ability; Content analysis; Demography; Diabetes; Diabetes mellitus (insulin dependent); Medical records; Patient-centered care; Quality of life
• ISSN: 0012-1797; 1939-327X
• DOI: 10.2337/db21-21-LB

Objective: To optimize type 1 diabetes (T1D) self-management, experts recommend tailoring care to meet people’s needs and preferences. To facilitate this tailoring, we developed and validated the Barriers and Supports Evaluation (BASES) tool to identify working-age adults’ T1D self-management barriers and supports.
Methods: Participants were 25-64 year old adults with T1D recruited from clinics and a community-based registry. Guided by Social Cognitive Theory, content analysis of 33 semi-structured interviews was used to create a comprehensive item pool of 136 items, further refined to 70 candidate items on a 5-point Likert scale through cognitive interviewing and pilot testing. To develop and validate the tool, exploratory and confirmatory factor analyses were applied to 392 participants’ survey responses to the candidate barrier and support items. Additional survey data included demographics and the Diabetes-Specific Quality of Life Scale-Revised. To evaluate concurrent validity, A1c values from medical records and quality of life (QoL) scores were regressed on domain scores.
Results: Factor analyses yielded 5 domains with a total of 30 items: (1) Cognitive Processes, (2) Costs and Insurance, (3) Learning Opportunities, (4) Family and Friend Support, and (5) Diabetes Provider Interactions. Models exhibited good to adequate fit, with Comparative Fit Index (CFI)>0.88 and Root Mean Squared Error of Approximation (RMSEA)<0.06. All domains demonstrated significant associations with A1c and QoL in the expected direction, with the exception of the Family and Friend Support domain. The Cognitive Processes domain exhibited the strongest associations with both A1c (β=0.7; 0.4-0.9) and QoL (β=-19; -21 to -17).
Conclusions: The BASES tool is brief, patient-centered, and demonstrates construct and concurrent validity. The tool could be used in clinical practice to tailor diabetes support and education to individual’s needs, improving effectiveness of services.