Abstract
AIMS: To explore the association between using automated insulin delivery (AID) systems and physical activity (PA) levels among people with type 1 diabetes. METHODS: Cross-sectional study including 1156 participants from the BETTER Type 1 Diabetes registry. INCLUSION CRITERIA: type 1 diabetes; aged ≥18 years old; reported PA with a validated questionnaire. PA is defined as activities lasting at least 10 consecutive minutes. Regression models (with and without adjustment for imbalanced variables) were used to compare PA levels among treatment options, with pairwise comparisons between the AID group and each group (i.e. Pump + continuous glucose monitoring (CGM), Injections + CGM, and non-CGM group). RESULTS: In the AID group, 87% of participants reported performing PA during the past 7 days (primary outcome), compared to 79% in the Pump + CGM group, 77% in the Injections + CGM group and 77% in the non-CGM group (p = 0.134 without adjustment; and p = 0.316 with adjustment for sex, age, duration of diabetes, highest educational level and annual household income). Only 28% of study participants met the target of performing at least 150 min of moderate to vigorous PA per week. CONCLUSIONS: Compared with other treatment modalities, AID users within the BETTER T1D registry tended to engage in more frequent and longer PA. However, these differences may be attributed to sociodemographic factors rather than the technologies themselves. Our study also shows a low percentage of participants meeting PA recommendations. Continued research and development on diabetes technologies, more personalized interventions and public health strategies are needed.