Abstract 4139423: Glycemic Changes After Peer Group Exergaming and Step Goal Activities Among Adolescents With Type 1 Diabetes

Adolescents with type 1 diabetes (T1D) experience both psychosocial and self-management barriers to physical activity (PA). We previously reported on a peer support intervention with group exergaming and step-counting goals for adolescents with T1D. The intervention yielded high participation, motivation, perceived group cohesion, and participant satisfaction. The purpose of the present sub-study was to assess 24hr continuous glucose monitoring (CGM) changes following the biweekly 20-40min group exergaming sessions (Nintendo Ring Fit Adventure) across the 6-week intervention, adjusted for other daily PA tracked by smartwatch (Fitbit Inspire 2). Participants (n=15) were mean age 15.6 [SE=0.4] years, 7 non-Hispanic white, 6 female, 1 non-binary, mean A1c 8.9%±0.6%. They provided 75.3% of possible 24hr CGM readings (n=12 Dexcom G6, n=3 Abbott Libre 2) and 70.5% of possible 24hr Fitbit readings, thus 1,008 person-hours with CGM and Fitbit. In a 2-level linear model (i.e., within-person biometrics from CGM and Fitbit, after adjustment for between-person HbA1c and age), hours ≤24hr after exergaming sessions - compared to hours >24hr after them – had lower average glucose (185mg/dL [95% CI 173,196] vs 192mg/dL [177,214], t=-2.66, P <.01) and higher probability of glucose being in the clinical target range of 70-180mg/dL (OR=1.18 [95% CI 1.06,1.32], t=2.98, P <.01). Increased hourly Fitbit steps further contributed to lowering average glucose (standardized β=-0.65 (SE=0.11), t=-2.37, P <.01) and increasing probability of being in the clinical target glucose range (β=0.021±0.003), t=6.26, P <.01). Whereas, increased hourly Fitbit heart rate contributed to elevating average glucose (β=0.52±0.08), t=6.52, P <.01) and decreasing likelihood of being in the clinical target glucose range (β=-0.012±0.002), t=-5.20, P <.01). In conclusion, both peer group exergaming and Fitbit steps were followed by glucose being modestly lower and more often in the clinical target range, suggesting our intervention was an implementable strategy to aid in glucose management but may require increasing the volume of exercise. Fitbit heart rate elevations independent of PA may signal sympathetic nervous system activation (e.g., from mental stress) and partially predict elevation of glucose. Further research should test dose response to varying amounts of exergaming, refine the optimal time window duration after exergaming to observe glucose effects, and examine other CGM metrics.