Glucose in the Dynamic Psychosocial Environment: Findings From Adolescents With Type 1 Diabetes

This study aimed to explore how real-time fluctuations in psychosocial experiences impact short-term glycemic outcomes among adolescents with type 1 diabetes.

Adolescents with type 1 diabetes (N = 88) completed 8 days of ecological momentary assessments (EMA), reporting 8 times per day on their negative and positive mood, diabetes self-care, and social interactions. Glucose was measured with participants' personal continuous glucose monitoring devices. Dynamic structural equation modeling was used to examine links of EMA-assessed psychosocial factors to glycemic outcomes (mean, standard deviation, time in range, and time above range) over the 2 hours following each EMA survey. Concurrent links between the psychosocial variables and glucose were also assessed.

Diabetes self-care predicted lower mean glucose over the next 2 hours but was more consistently linked to glucose in concurrent analyses. Mood was unrelated to glucose outcomes. Experiencing a social interaction predicted worse outcomes over the next 2 hours for 3 of the 4 glycemic metrics (standard deviation: CI [.32,1.88]; TIR: CI [-0.22, -0.01]; TAR: CI [-0.23, -0.03]). Additional analyses showed this effect held across both family and peer interactions and were generally unaffected by interaction quality.

Results suggest that social interactions-regardless of with whom one interacts-are highly distracting to adolescents with diabetes and may detract from one's diabetes management. Further, adolescents' perceptions of diabetes self-care may miss important aspects of disease management, or their self-care efforts may be overshadowed by the variability and unpredictability of type 1 diabetes. Results are exploratory and should be replicated in a confirmatory setting.