Five heterogeneous HbA1c trajectories from childhood to adulthood in youth with type 1 diabetes from three different continents: A group‐based modeling approach. Issue 7 (28th August 2019)

• Record Type: Journal Article
• Authors: Clements, Mark A.; Schwandt, Anke; Donaghue, Kim C.; Miller, Kellee; Lück, Ursula; Couper, Jennifer J.; Foster, Nicole; Schröder, Carmen; Phelan, Helen; Maahs, David; Prinz, Nicole; Craig, Maria E.
• Is Part Of: Pediatric diabetes. Volume 20:Issue 7(2019)
• Page Start: 920
• Page End: 931
• Publication Date: 2019-08-28
• Subjects: childhood diabetes -- diabetes -- HbA1c -- pediatric diabetes -- type 1 diabetes; Diabetes in children -- Periodicals; 616.462
• DOI: 10.1111/pedi.12907
• Languages: English
• ISSNs: 1399-543X

Abstract

Abstract: Objectives: Only a fraction of youth meet established targets for glycemic control; many experience deteriorating control over time. We compared trajectories of hemoglobin A1c (HbA1c) among youth from three trans‐continental type 1 diabetes (T1D) registries and identified clinical variables associated with the odds of following increasing vs stable trajectories. Research design and methods: Analyses included longitudinal data from 15 897 individuals age 8 to 18 with T1D for at least 2 years and HbA1c measurements in at least 5 years during the observation period. Cohorts were selected from Australasian Diabetes Data Network (ADDN; Australia), German/Austrian/Luxembourgian Diabetes‐Patienten‐Verlaufsdokumentation initiative (DPV; Germany/Austria/Luxembourga), and the T1D Exchange Clinic Network (T1DX; US) clinic registries. Group‐based trajectory modeling and multivariable logistic regression identified unique HbA1c trajectories and their predictors. Results: Five heterogeneous trajectories of glycemic control in each registry were identified: low, intermediate, high stable; intermediate and high increasing. The overall HbA1c level for each trajectory group tended to be lowest in the DPV, higher in the ADDN, and highest in the T1DX. The absolute level of HbA1c and the proportion of individuals within each trajectory varied across registries: 17% to 22% of individuals followed an increasing trajectory. Compared with maintaining a stable trajectory, following an increasing trajectory was significantly associated with ethnic minority status, lower height z‐score, higher BMI z‐score, insulin injection therapy, and the occurrence of severe hypoglycemia; however, these factors were not consistent across the three registries. Conclusions: We report the first multinational registry‐based comparison of glycemic control trajectories among youth with T1D from three continents and identify possible targets for intervention in those at risk of an increasing HbA1c trajectory.