Summary: In a crossover trial in type 1 diabetes, enabling the Omnipod 5 Activity feature 30 or 60 minutes before exercise reduced insulin delivery and blunted the exercise-induced glucose drop compared with automatic mode, although the rate of hypoglycemia during exercise did not significantly differ.
PICO Summary
| Element | Detail |
|---|---|
| Population | 38 adults and adolescents with type 1 diabetes on the Omnipod 5 system; three-way crossover, 70-min treadmill sessions. |
| Intervention | Activity feature enabled 60 minutes (AF-60) or 30 minutes (AF-30) before exercise. |
| Comparison | Standard automated mode (Auto). |
| Outcome | AF-60 and AF-30 reduced insulin delivery before and during exercise (p<0.001) and attenuated the glucose drop (Auto -57, AF-60 -44 [p=0.02], AF-30 -36 mg/dL [p=0.01]). Hypoglycemia during exercise did not differ significantly (Auto 42%, AF-60 29%, AF-30 24%). Post-exercise metrics were comparable. |
Omnipod 5 Activity feature and exercise glucose
Crossover RCT · type 1 diabetes · 70-min treadmill
Enabling the Activity feature before exercise cut insulin delivery and softened the glucose fall, with the largest effect at 60 minutes. Hypoglycemia rates trended lower but were not significantly reduced.
Expert Commentary
This is a practical, well-designed crossover study addressing a question that genuinely vexes patients on automated insulin delivery: how to exercise without crashing. The mechanism worked as intended, enabling the activity feature ahead of time cut insulin delivery and meaningfully softened the glucose fall, with a clear dose-of-timing effect. But the honest and important finding is the one the authors highlight rather than bury: despite these favourable changes, the actual incidence of hypoglycaemia was not significantly lower across conditions, with the numerical trend not reaching significance in this sample. That gap between an improved glucose trajectory and unchanged hypoglycaemia rates is the real clinical lesson, the feature helps but does not by itself solve the problem. The study is also modest in size and conducted in a controlled treadmill setting rather than messy real life. Can I use this with my patients? Yes, concretely. For a patient on this system, it supports enabling the activity feature well before exercise, ideally the full 60 minutes, while making clear that this is one layer of protection, not a guarantee, and that pre-exercise carbohydrate and glucose monitoring remain essential to avoid lows.
References
Turner LV, Sherr JL, Zaharieva DP, et al. Use of the Omnipod 5 automated insulin delivery system activity feature reduces insulin delivery and attenuates the drop in glycemia associated with exercise in a randomized controlled trial. Diabetes Care. 2025;48(9):1598–1606. doi:10.2337/dc25-0141