Diabetes care behaviors and glycemic control in adolescents with Type 1 diabetes: An individual growth curve approach

Individual growth curve modeling was applied to an accelerated longitudinal data set to characterize change in actual diabetes care behaviors and glycemic control in adolescents over three years; to examine the complex interrelations between diabetes care behaviors and glycemic control; and to distinguish growth trajectories by inter-individual characteristics that may explain variability in change. At study entry, participants (n = 238; M age = 13; range 10-17; 50% male; 71% Caucasian) recruited from tertiary care clinics in two major metropolitan areas demonstrated moderate glycemic control (M = 8.3%), wide range in the performance of diabetes care behaviors, and a predominantly conventional treatment (CT ≤ 2 daily injections; 55%) rather than an intensive treatment (IT ≥ 3 daily injections/pump) approach to insulin therapy. Analyses revealed an overall significant decline in eating frequency over time, t = -11.0, p < .001, related to poorer glycemic control and IT. A decline in blood glucose monitoring over time, t = -6.5, p < .001, was related to CT and nonwhite ethnicity. A decline in exercise frequency over time, t = -6.6, p < .001, related to IT, lower SES, and female gender. In contrast, more appropriate dietary composition, according to ADA guidelines, was related to higher SES and unlike other self-care behaviors, diet composition and exercise duration were some of the few stable disease care behaviors that did not significantly change throughout adolescence. Unexpectedly, glycemic control did not deteriorate during adolescence; rather visual inspection suggests a worsening in early adolescence was followed by modest improvement and/or stabilization in later adolescence. Better glycemic control over time related to higher SES and CT. Evaluation of different treatment regimens revealed that youth who use IT concomitantly engage in more appropriate blood glucose monitoring and demonstrate better dietary composition. However, IT also was related to other poorer disease care behaviors and to poorer metabolic control over the course of the study. A heterogeneous IT group may explain these inconsistent results. In sum, results support a multidimensional conceptualization of diabetes care and complex associations between diabetes care behaviors and glycemic control.