Introduction and Objective: DD- the emotions, stress, and worries associated with diabetes— is associated with glycemic control and severe hypoglycemia in younger individuals with T1D, but little is known about DD in older adults. Methods: We recruited OA (65+) with T1D from the T1D Exchange national registry (08/27/2024-11/05/2024) to complete an electronic survey including the Type 1 Diabetes Distress Assessment System (T1DDAS), which includes a core score and 10 subscores for sources of DD. We assessed univariate linear associations between other self-reported sociodemographic and diabetes-related variables and T1DDAS core scores (clinically significant ≥2.0). Results: Among OA with T1D (n=337; 70.7±4.6 years; 63% female; 96.4% White; median (IQR) T1D duration 45 (25) years; 97% CGM users; 66% HCL users; HbA1c 6.6±0.7%; 36% elevated DD; median (IQR) DD core score 1.75 (1.125), significant positive associations with core scores included sex, HbA1c, and history of 1+ ER visits, while T1D duration and food security status were negatively associated (Figure). Conclusion: Among a self-selected sample of OA with T1D, we corroborated known clinical associations with DD (HbA1c, T1D duration, ER visits). Female sex and food insecurity also emerged as potentially novel risk factors for DD in this age group. A. Wisniewski: None. J.B. DeMonte: None. A. Cristello Sarteau: None. A. Fruik: None. A. Kahkoska: None. NIH/NIDDK (P30 DK124723); NCDRC Ignition Award
This book shows how dbt is used to build data transformation pipelines that enable dependency management, version control, and automated testing
Introduction and Objective: Autoantibody testing informs diabetes classification. However, result interpretation is nuanced, and incomplete testing may lead to incorrect diagnoses, inappropriate treatment, and adverse outcomes. We aimed to characterize differences in autoantibody evaluation across clinical settings and quantify rates of incomplete T1D evaluation based on practice setting. Methods: Using a single-center academic electronic medical record, we evaluated people who had GAD65, IA-2A, insulin, and ZnT8 autoantibodies drawn for diabetes classification in July-December 2024. Study personnel collected demographic information, diabetes history, and laboratory results. Incomplete evaluation was defined as a person with phenotypic features of T1D with ≥ 1 negative antibody who did not have a complete autoantibody panel. Results: During the reporting period, 315 people were tested for islet autoantibodies across various practice settings: endocrinology clinic (41%), primary care (27%), inpatient admission (21%), and the emergency department (ED) (10%). GAD65 was ordered most frequently (92%), followed by IA-2A (61%), insulin (50%), and ZnT8 antibodies (37%). By practice setting, incomplete evaluations were identified in 31% (ED), 26% (inpatient), 20% (primary care), and 8% (endocrinology) of workups. Incomplete evaluation led to misclassification of diabetes and delayed diagnosis in multiple cases. Conclusion: Nearly 25% of patients with features of T1D and ≥ 1 negative autoantibody undergoing pancreatic autoantibody evaluation had an incomplete work-up. This is likely an underestimate, as one limitation of this data is that it does not capture people with phenotypic characteristics of T1D who are not being evaluated for autoantibodies. Development of a T1D autoantibody panel may prevent misclassification of diabetes and complications. Understanding practice patterns will inform tools to improve consistency of autoantibody ordering across care settings to improve T1D evaluation. Disclosure M.C. Abram: None. B.C. Matson: None. R.J. Vitale: None. K.R. Klein: Consultant; Novo Nordisk, Roche Pharmaceuticals.
As with any new technical project, time must be spent on the setup and configuration to establish a solid foundation that will enable the contributors to the project to be successful. dbt is no different in this respect, and there is more than one path available to you for successfully setting up a new project.
In this chapter, we will explore the concept of sources and seeds in dbt. These two items are mostly independent of each other, and we combined them into the same chapter as they are smaller, but very important, subjects. Sources refer to the original data sources, such as data from databases, that are used as the input for dbt. Source tables are referred to in dbt models via a Jinja function, which we will discuss more later in the chapter. Seeds, on the other hand, refer to the data that is used to populate the initial sets on small unchanging data, typically used for reference data or lookup tables. Seeds are still referenced within models utilizing a Jinja function, but use a different command than sources.