Section 1: What the Omnipod 5 is

The Omnipod 5 is a tubeless, waterproof patch pump system that combines the Omnipod 5 Pod with a Dexcom G6 or G7 continuous glucose monitor (CGM). What makes it architecturally distinctive among AID systems is that the closed-loop algorithm runs on the Pod itself — not on a companion phone or controller. This means the system continues to dose insulin and respond to glucose trends even when the phone is out of range, switched off, or left in another room.

The Pod adheres directly to the skin and contains both the insulin reservoir and the infusion mechanism. There is no tubing. This design removes the physical tether common to other pump systems and is often cited by users as a reason they chose this platform.

What makes Omnipod 5 distinctive

• Tubeless, on-body design. The Pod adheres directly to skin — no tubing connects pump to cannula. The Pod is waterproof.
• Algorithm on the Pod, not the phone. Closed-loop dosing continues autonomously even when the phone is not nearby. The phone app is used for bolusing and monitoring, but is not required for the algorithm to function.
• Personalised glucose target. The user sets a glucose target in the app. This is the primary mechanism through which responsiveness is adjusted. A lower target instructs the algorithm to work more aggressively toward a tighter glucose level; a higher target is more conservative.
• Algorithm-led at lower responsiveness settings. When the target is set higher and bolusing behaviour is less consistent, the system takes on more of the management. This can be appropriate for periods when active engagement with the system is lower.
• User behaviour matters more at higher responsiveness settings. When the target is set lower, the algorithm works harder to tighten glucose — but it relies on accurate carb counting and consistent meal bolusing to perform at its best. At the most responsive settings, missing boluses or inaccurate carb entries are more likely to result in visible glucose excursions.
• Compatible CGMs. Omnipod 5 is currently approved for use with Dexcom G6 and Dexcom G7.

Section 3: The five educational calibration band levels

The table below summarises the five levels used in the GNL framework. Each level reflects a different point along the responsiveness spectrum. All descriptions are educational approximations of average tendencies — individual experience varies considerably, and the GNL explorer outputs are a starting point for exploration, not a clinical recommendation.

Reminder: calibration band is an educational label, not a device setting.

Level 5 — Very high responsiveness

At this level, the educational calibration band sits at 150–200 mg/dL and the active insulin time is set to 2 hours. The target style is the tightest in the framework, implying the algorithm is working hard to keep glucose close to the lower end of its operating range. The shorter AIT means the system treats insulin as clearing quickly, allowing more frequent corrections. The carb ratio expression is very strong, meaning the system’s on-board carb logic is most sensitive to meal entries. The behaviour requirement is correspondingly very high: this level tends to describe scenarios where the user is consistently counting carbohydrates accurately and bolusing reliably at each meal. The correction factor used in the GNL framework for this level is 90 mmol (1600 mg/dL).

This level is an educational approximation of an engaged user working at the most responsive end of the Omnipod 5 operating range. It is not appropriate for all users or all periods — the higher behaviour requirement means that inconsistent bolusing is more likely to produce visible glucose excursions.

Level 4 — High responsiveness

The calibration band here sits at 175–225 mg/dL and the active insulin time is 2 hours 15 minutes. The target style is tight and the carb ratio expression is strong. The behaviour requirement is high — consistent bolusing and reasonably accurate carb counting tend to characterise users who achieve results toward this end of the spectrum. The correction factor is 95 mmol (1700 mg/dL). The system style leans user-tighter: the algorithm is responsive, but the user’s bolusing behaviour remains a material input to the overall result.

Level 3 — Medium responsiveness

The calibration band sits at 200–250 mg/dL and the active insulin time is 2 hours 30 minutes. Target style is balanced, carb ratio expression is mid-band, and the behaviour requirement is moderate. The correction factor is 100 mmol (1800 mg/dL). This level represents a midpoint on the responsiveness spectrum — the system is neither operating at its most aggressive nor its most tolerant. Many people find that this band reflects their everyday experience on Omnipod 5 when managing meals with reasonable consistency.

Level 2 — Low responsiveness

At this level, the calibration band sits at 225–275 mg/dL and the active insulin time is 2 hours 45 minutes. The target style is conservative and the carb ratio expression is weak, meaning the algorithm’s on-board carb logic has less influence on the output. The behaviour requirement is low. The correction factor is 105 mmol (1900 mg/dL). The longer AIT means the system is more cautious about stacking corrections. The system style is algorithm-led: the system takes more of the management responsibility and is less dependent on consistent user bolusing. This level may reflect periods of lower engagement, illness, or significant life disruption — or simply a preference for more system-led management.

Level 1 — Very low responsiveness

The calibration band sits at 250–300 mg/dL and the active insulin time is 3 hours. The target style is at its highest tolerance and the carb ratio expression is very weak. The behaviour requirement is very low — the algorithm is working at its most autonomous. The correction factor is 110 mmol (2000 mg/dL). The longest AIT in the framework means the system is at its most cautious about correction stacking. The system style is firmly algorithm-led. This level is an educational representation of the system operating in its most conservative mode, with the user providing minimal active input beyond the pump being worn and the CGM functioning. It tends to describe scenarios where the glucose target set in the app is at the more conservative end, or where bolusing is infrequent.

Section 5: How to use the explorer output

The GNL Multi-System Explorer generates exploratory output ranges — not prescriptions, not targets, and not device settings. Understanding what the output is and what it is not helps make the most of it.

What the output describes

The output describes a range of estimated insulin parameter values — correction factor, carb ratio, and basal reference — based on the inputs entered and the calibration band level selected. These values represent educational approximations of what those parameters might look like in a person with broadly similar characteristics who is using Omnipod 5 at the corresponding responsiveness level. They are derived from population-average tendencies and well-established insulin dosing relationships.

The output is most useful as a structured starting point — a way to put approximate numbers on a conversation about whether current settings feel broadly appropriate, and to explore how they might shift if responsiveness were changed. Many people find the output helpful for framing a question to bring to a clinic appointment, rather than as an answer in itself.

What the output is not

• It is not a recommendation to change any device setting.
• It is not a calculation of your individual insulin requirements.
• It is not validated against clinical trial data for any individual.
• It is not a substitute for assessment by a diabetes care team.
• It does not account for CGM accuracy, infusion site variation, carbohydrate absorption differences, or any of the many individual factors that shape real-world glucose outcomes.

Uncertainty is built in

The output is presented as a range rather than a single number deliberately. The range reflects the genuine uncertainty in applying population-average relationships to any individual. Someone at the lower end of the output range and someone at the upper end may both be entirely reasonable — the spread is an honest acknowledgement that there is no single correct answer.

This is consistent with how AID systems tend to behave in practice: two people with similar characteristics using the same system at the same responsiveness level often end up at different calibrated values because of the many variables the algorithm cannot fully account for at a population level. Your CGM data, reviewed over time with your care team, is the most reliable guide to what is working for you.

Section 6: Limitations and disclaimer

Calibration band: educational representation only

The “calibration band” is an educational expression layer created by GNL. It is not a real parameter in the Omnipod 5 device or app. It is not a term used by Insulet in their clinical documentation, training materials, or device software. It exists solely within the GNL educational framework to describe, at a conceptual level, the responsiveness spectrum the algorithm can operate across. Any reference to “calibration band” in GNL content — including this page, the explorer, and any associated resources — should be understood in this context.

If you encounter the term “calibration band” in any other context in relation to Omnipod 5, it is not connected to the GNL framework and may have a different meaning. Always refer to official Insulet documentation and your diabetes care team for device-specific guidance.

Educational purpose only

This page, and all GNL content relating to the AID system explorer, is for educational exploration. It describes average responses and general principles drawn from insulin dosing literature and general AID system behaviour. It does not constitute individual clinical guidance, and it does not incorporate your personal clinical history, medical records, or device data.

GNL is an educational resource for people living with type 1 diabetes. It is not a clinical decision support tool and does not output clinical recommendations. Nothing on this page should be used as the basis for changing any insulin dose, device setting, or management approach without discussion with a qualified diabetes care team.

No proprietary claims

The GNL framework uses publicly available information about AID systems combined with established insulin dosing principles. It makes no claim to represent the proprietary algorithm logic of Omnipod 5, to have access to Insulet’s internal parameters, or to replicate the behaviour of the on-device algorithm. The five calibration band levels are GNL constructs, not Insulet categories.

Individual variation

Glucose outcomes in type 1 diabetes are shaped by an exceptionally wide range of factors: insulin sensitivity that varies hour to hour and day to day, carbohydrate absorption that varies by food composition and gut transit, physical activity, sleep quality, stress, illness, hormonal cycles, infusion site condition, CGM sensor accuracy, and much more. No population-average framework can capture this variation at an individual level. The explorer output acknowledges this by presenting ranges rather than single values — and by consistently pointing to CGM data and care team review as the appropriate next step.