AID systems

The IOB Trade-Off: How Four AID Systems Make Different Bets on Insulin

Every automated insulin delivery system must answer the same question: how much insulin is still working? They all answer it differently. These design choices shape everything from correction aggressiveness to exercise safety — and understanding them changes how you use your system.

AID systems Insulin on board Clinical practice

TLDR — the 60-second version

All four AID systems available in the UK try to solve the same problem: delivering enough insulin to lower high glucose without stacking too much and causing a low. They do this by tracking insulin on board (IOB) — how much insulin is still active in the body. But each system tracks IOB differently, and that single design choice creates different strengths, different blind spots, and different risks.

Control-IQ uses a fixed 5-hour decay model. IOB is visible on the pump. Conservative but transparent — what you see is roughly what you get.
MiniMed 780G uses a shorter 2–3 hour active insulin time. This lets it correct more aggressively, but device-reported IOB runs lower than the insulin actually circulating — a hidden risk during exercise.
CamAPS FX tracks IOB physiologically, closer to how insulin actually behaves in the body. No major visibility distortion. The most adaptable system, with flexible targets by time of day.
Omnipod 5 uses a similar short-AIT approach to the 780G. Tubeless and simple, but shares the same IOB visibility trade-off during activity.

The implication: there is no universally best system. The right choice depends on the person — their activity level, their variability, their need for transparency versus simplicity. And the only way to see these differences clearly is to model them side by side.

That is exactly what the GNL Explorers were built to do.

This content is for educational exploration only. It is based on clinical data and real-world patterns describing average responses across populations. It is not a prescription, not a medical device, and must not be used as either. It cannot replace individual clinical guidance from your diabetes care team.

The question every AID system must answer

When you take a bolus of rapid-acting insulin, it does not act instantly and it does not disappear on a schedule. It absorbs over minutes. It peaks. It tails off over hours. The shape of that curve varies by dose size, injection site, body temperature, and activity level.

An AID algorithm needs to know — at every five-minute CGM reading — how much of that insulin is still working. If it overestimates, it holds back too much and glucose stays high. If it underestimates, it stacks more insulin on top of what is already circulating and causes a low.

This is the IOB trade-off. Every system makes a bet about the shape of insulin action, and that bet defines how the system behaves.

Four systems, four different bets

Control-IQ: visible IOB, conservative decay

Control-IQ assumes insulin decays over a fixed 5-hour window. This creates a visible IOB number on the pump that users and clinicians can monitor. The trade-off is conservatism: the algorithm corrects less aggressively because it believes insulin is active for longer.

What this means in practice: Control-IQ is the most customisable system. The correction factor is the primary lever — a stronger correction factor makes the algorithm more responsive; a weaker one makes it more cautious. Basal rates, carbohydrate ratios, and sleep mode all interact. When settings are well tuned, it is excellent. When they are not, it can feel sluggish.

The IOB story: because the 5-hour model is longer than real-world peak insulin action (typically 1.5–2.5 hours), the algorithm tends to be conservative in its corrections. This is protective in some ways — less risk of stacking — but it also means the system can be slow to bring down persistent highs. The IOB number on the pump is visible and can be used for planning, but it does not perfectly match physiological reality either.

MiniMed 780G: aggressive corrections, shorter AIT

The 780G takes the opposite approach. By using an active insulin time of 2–3 hours (adjustable), the algorithm believes insulin clears faster than it physiologically does. This lets it deliver correction boluses more frequently — every 5 minutes if glucose remains elevated.

What this means in practice: the 780G is the most aggressive system for tackling high glucose. It auto-learns basal rates and sensitivity, requiring less manual tuning. Overnight control is typically excellent. The active insulin time setting is the primary responsiveness dial — shorter means more aggressive.

The IOB story: here is where the trade-off bites. By assuming insulin clears in 2–3 hours, the device reports IOB that is lower than the insulin actually circulating. At rest, this works — the algorithm compensates by adjusting the next correction. During exercise, it becomes a hidden risk. A person starting activity with device IOB showing 2 units may actually have 3 or more units of physiological insulin still working. The system cannot see this gap.

CamAPS FX: physiological tracking, flexible targets

CamAPS FX tracks IOB using a model closer to actual insulin pharmacokinetics. The algorithm does not publish the exact curve, but the behaviour indicates it accounts for the longer tail of insulin action more realistically than a fixed AIT model.

What this means in practice: CamAPS can safely deliver approximately 60% of total insulin algorithmically — more than any other system — because its IOB tracking does not create the visibility gap that shorter AIT models do. Targets are adjustable by time of day (as low as 4.4 mmol/L to as high as 11.0 mmol/L), making it the most adaptable system for variable needs: overnight tightening, pre-exercise loosening, pregnancy management.

The IOB story: no major trade-off. Because the IOB model is more physiological, there is less divergence between what the algorithm thinks is active and what is actually circulating. This makes exercise planning more intuitive and reduces the hidden IOB risk that affects shorter-AIT systems. The Ease Off feature and Slowly Absorbed Meal option add further flexibility.

Omnipod 5: tubeless simplicity, similar IOB trade-off

Omnipod 5 uses a responsiveness-level system (1–5) that maps to different AIT settings and glucose targets. At the most responsive setting, AIT is approximately 2 hours; at the most conservative, approximately 4 hours. The algorithm lives inside the Pod itself.

What this means in practice: Omnipod 5 is the simplest to operate. Tubeless design, minimal settings, algorithm on-device. For people who want AID without complexity, it delivers. The responsiveness level is the main adjustment — everything else is automated.

The IOB story: at responsive settings, Omnipod 5 shares the same IOB visibility trade-off as the 780G. Shorter AIT means device IOB runs lower than physiological reality. At more conservative settings (level 1–2, AIT 3.5–4 hours), the gap narrows. Exercise planning on responsive settings requires the same external awareness as the 780G — the device alone cannot show you the full insulin exposure.

Why this matters: the exercise litmus test

Exercise is where the IOB trade-off becomes visible — sometimes urgently. During aerobic activity, insulin action is amplified. Muscle glucose uptake increases through pathways that are partly independent of insulin. If insulin on board is higher than expected, the combination can cause rapid glucose drops.

This is why different systems create different exercise risks:

System	IOB visibility	Exercise implication
Control-IQ	Visible on pump (5-hour model)	IOB can be monitored before activity. Conservative model is somewhat protective, though still imperfect.
MiniMed 780G	Lower than physiological (2–3 hour AIT)	Device IOB underestimates true exposure. External tools or experience needed to gauge real risk before activity.
CamAPS FX	Physiological tracking	Least IOB distortion. Ease Off mode reduces corrections. Most intuitive for exercise planning.
Omnipod 5	Lower than physiological at responsive settings	Similar risk to 780G at high responsiveness. Activity mode should be started 90 minutes before exercise.

The Activity to Lower Highs Explorer models exactly this interaction — showing how glucose is expected to fall over 10, 20, and 30 minutes of moderate activity based on true insulin exposure, not device-reported IOB.

Settings are levers, not prescriptions

Each system gives clinicians and patients different levers to pull. Understanding which lever does what — and why — is where optimisation begins.

System	Primary responsiveness lever	Secondary levers
Control-IQ	Correction factor	Basal profile, carb ratio, sleep mode
MiniMed 780G	Active insulin time (2–3 hours)	Glucose target, carb ratio
CamAPS FX	Glucose target (by time block)	Carb ratio, Ease Off, Slowly Absorbed Meal
Omnipod 5	Glucose target / responsiveness level	Carb ratio, activity mode

The AID System Explorer lets you enter age, weight, and total daily dose and see exactly how each system responds at five different responsiveness levels — including the algorithm-calculated basal rates, correction doses, and glucose targets. The differences are often striking.

No clinician can do this maths in their head

This is the uncomfortable truth behind AID optimisation. The interactions between IOB decay curves, correction factors, basal modulation, meal absorption, and activity are too complex for mental arithmetic. A 15-minute clinic appointment cannot model how a correction factor change on Control-IQ will alter the stacking risk compared to shortening AIT on the 780G.

This is not a criticism of clinical practice — it is a recognition that the systems themselves are more sophisticated than the tools traditionally available to discuss them. The algorithms run continuous calculations every five minutes. The conversations about them happen every three months.

This is exactly why the GNL Explorers exist. They bridge that gap. They let clinicians and patients see — visually and numerically — what changing a setting actually does to the system’s behaviour. Not as a prescription, but as a starting point for an informed conversation.

When a young person asks “Why does my pump keep giving me corrections at night?” the answer is different on every system. On Control-IQ, it may be a basal profile issue. On the 780G, it may be the AIT setting. On CamAPS, it may be the overnight target. Without a tool that shows these differences, the conversation defaults to generic advice.

With the explorers, the conversation becomes specific: “Here is what your system is doing at this responsiveness level. Here is what it would do if we adjusted this setting. Which direction makes sense for your life?”

Matching the system to the person

There is no universally best AID system. The evidence shows all four improve time in range and quality of life compared to non-AID therapy. The question that matters is: which system’s design philosophy matches this person’s life?

If transparency and control matter most

Control-IQ gives the most visible IOB and the most user-adjustable settings. It rewards understanding and responds well to careful tuning. Many people find it works best when they engage actively with the settings and use sleep mode strategically.

If aggressive correction and automation matter most

The 780G delivers the most automated correction logic. It learns continuously, corrects every 5 minutes, and requires the least manual tuning. It tends to produce excellent overnight results with minimal input. The trade-off is IOB visibility during activity.

If adaptability and physiological accuracy matter most

CamAPS FX offers the widest range of target customisation, the most physiological IOB tracking, and features specifically designed for variable needs — pregnancy, exercise, high-fat meals. It tends to suit people with unpredictable schedules or those who exercise frequently.

If simplicity and discretion matter most

Omnipod 5 is tubeless, requires minimal settings, and runs the algorithm inside the Pod itself. For people who want effective AID without complexity — or who need a discreet system for sport, work, or personal preference — it delivers with less overhead.

Explore the differences yourself

The GNL Explorers were built to make these invisible differences visible. Each tool addresses a different aspect of the IOB trade-off — from algorithm behaviour to exercise risk to glucose recovery.

AID System Explorer

Enter age, weight, and total daily dose. See how all four systems behave at five responsiveness levels — including basal rates, correction doses, and targets.

Open the AID System Explorer

Activity to Lower Highs Explorer

Model how glucose responds to moderate activity based on true insulin exposure — not device-reported IOB.

Open the Activity Explorer

Exercise Carbohydrate Calculator

Estimate carbohydrate needs for 30 minutes of activity based on insulin on board and body weight.

Open the Exercise Carb Calculator

Hypo and Hyperglycaemia Explorer

Map the mechanisms behind glucose going below or above range — identify which variables are most likely at play.

Open the Hypo and Hyper Explorer

The bottom line

AID systems are remarkable. They have changed what is achievable for hundreds of thousands of people with type 1 diabetes. But they are not identical, and treating them as interchangeable misses the point.

The IOB trade-off — the design choice each system makes about how to track active insulin — shapes everything downstream. It determines how aggressively the algorithm corrects, how visible the remaining insulin is to the user, and how safe it is to start exercising after a meal bolus.

Understanding these differences does not require an engineering degree. It requires the right tools to see them. That is what the GNL Explorers provide — not answers, but a framework for asking better questions with your diabetes care team.