The Acute Effects of Real-World Physical Activity on Glycemia in Adolescents with Type 1 Diabetes: The Type 1 Diabetes Exercise Initiative Pediatric (T1DEXIP) Study

Objective: Data from the Type 1 Diabetes Exercise Initiative Pediatric (T1DEXIP) study were evaluated to understand glucose changes during activity and identify factors that may influence changes. Research Design and Methods: In this real-world observational study, adolescents with type 1 diabetes self-reported physical activity, food intake, and insulin dosing (multiple-daily injection users) using a smart phone application. Heart rate (HR), continuous glucose monitoring data were collected, as well as pump data downloads. Results: 251 adolescents (age=14±2 years [mean±SD]; HbA1c=7.1±1.3% [54±14.2mmol/mol]; 42% female) logged 3,738 activities over ~10 days of observation. Pre-activity glucose was 163±66 mg/dL (9.1±3.7mmol/L), dropping to 148±66 mg/dL (8.2±3.7mmol/L) by end of activity; median duration of activity was 40 minutes (20, 75 [IQR]) with a mean and peak HR of 109±16 bpm and 130±21 bpm. Drops in glucose were greater in those with lower baseline HbA1c levels (P=0.002), shorter disease duration (P=0.02), less hypoglycemia fear (P=0.04), and those with a lower BMI (P=0.05). Event-level predictors of greater drops in glucose included self-classified "non-competitive" activities, insulin on board >0.05 U/kg body mass, glucose already dropping prior to the activity, pre-activity glucose >150 mg/dL (>8.3mmol/L) and time 70-180 mg/dL >70% in the 24 hours before the activity (all P<0.001). Conclusions: Participant-level and activity-event level factors can help predict the magnitude of drop in glucose during real-world physical activity in youth with type 1 diabetes. A better appreciation of these factors may improve decision-support tools and self-management strategies to reduce activity-induced dysglyemia in active adolescents living with the disease.