Guide series · Part 2 of 5

The two levers – insulin and carbs

Bolus reduction and carbohydrate timing are the two independent levers that most exercise strategies rest on. Knowing how each one works, and when to reach for which, changes what is possible.

Bolus reduction Carbs Pump

Part 2 · Anchor thesis

Two levers, used together, beat either one alone.

Every modern exercise recommendation for type 1 diabetes rests on two independent levers: adjusting the insulin dose around planned activity, and timing or sizing carbohydrate intake before, during and after. West et al (2011) demonstrated that combining the two produces more stable glycaemia than either on its own.

This Part sets out the framework, the evidence, and the decision points. It is not a prescription. It is a map of starting points that survive contact with CGM and real life – to refine with your diabetes team over time.

Major in the majors

Three variables still decide almost everything

1 · Most important

Insulin on board

How much recent bolus insulin is still active when exercise starts.

2

Starting glucose

Where you begin shapes which lever you reach for first.

3

Trend arrows

Direction and speed of change before exercise begins.

Lever one – bolus reduction

Rabasa-Lhoret et al (2001) established the 25 / 50 / 75 percent pre-meal bolus reduction scale for exercise of increasing intensity and duration. That scale is the foundation cited in every subsequent consensus statement, including the EASD/ISPAD 2025 position paper (Moser, Zaharieva and Pemberton).

Exercise typeTypical starting reduction of the meal bolus
Low intensity, 30 minutes~25%
Moderate intensity, 30-60 minutes~50%
Moderate-to-vigorous, 60+ minutes~75%

Evidence suggests most people find these percentages a reasonable starting point when exercise is planned within the 90-minute window after a meal. Individual response varies. Many people experiment with the framework and iterate based on CGM feedback over weeks.

This is educational, not prescriptive. Changes to insulin doses should be agreed with your diabetes team. The percentages above are the scale used in published consensus statements; they are not a recipe.

West, Moser and the combined approach

West et al (2011) showed that bolus reduction combined with carbohydrate feeding produced more stable glycaemia than either intervention alone. Moser and colleagues have extended this into the AID era, with explicit guidance for pump, AID, and multiple daily injection regimens separately. The pattern is consistent: the two levers work together.

Lever two – carbohydrate timing and amount

Riddle et al (2000) provides the approximate upper bound for glucose utilisation during sustained exercise in adolescents – around 1.5 g/kg/hr. That figure sets the ceiling for how much carbohydrate might be needed for long sessions. Most people, most of the time, use considerably less.

Shetty et al (2016) refined the dose-response by intensity: higher intensity pushes carbohydrate demand higher within the session, and the requirement is not linear. Campbell et al (2015) added the bedtime snack evidence – a ~0.4 g/kg carbohydrate snack after daytime exercise tends to reduce overnight hypoglycaemia. Kalergis et al (2003) refined that further: composition (not only quantity) affects the overnight trajectory.

Timing beats amount in most windows

McCaughan, Shetty, and the practical pattern from Moser all point the same way: carbohydrate timed to the window where hypoglycaemia is most likely tends to outperform a larger dose given at the wrong time. A common starting point is a small amount of carbohydrate just before aerobic exercise when IOB is still active, with a bedtime snack after longer sessions.

Fasted versus fed exercise

A different planning problem

McCaughan et al (2021) characterises the fasted-state exercise response in type 1 diabetes. With minimal IOB, the glucose-lowering effect of moderate exercise is attenuated. This is a different planning problem from fed-state exercise: the risk is often post-exercise hyperglycaemia (counter-regulation dominates) rather than hypoglycaemia during the session.

Many people find fasted morning exercise simpler to manage once they know this – low IOB, stable glucose, and exercise type becoming the main variable. Evidence suggests morning fasted sessions are a common way to reduce the mental load of the IOB calculation.

Pump disconnection

For pump users, disconnecting during exercise is a specific sub-question – for how long, with what basal deficit, and with what glycaemic consequence. The evidence base here is smaller and more contested than for bolus reduction, with reasonable trials on either side.

Short disconnections (up to an hour) during aerobic exercise are a pattern many people use. Longer disconnections need closer attention to the basal deficit that builds up and the post-session rebound. This is very much a conversation for your diabetes team rather than a rule.

Decision heuristic – which lever first?

The shortest usable heuristic:

Exercise within 90 minutes of a bolus

IOB is the dominant driver. Bolus reduction usually matters more than extra carbohydrate. Reach for lever one.

Exercise more than 3 hours after a bolus

The Three-Hour Rule has reduced IOB. Small carbohydrate adjustments often do more work than insulin changes. Reach for lever two.

Long sessions (60+ minutes, moderate or higher)

Both levers, together. Bolus reduction on the preceding meal plus carbohydrate during the session, with a bedtime snack for longer efforts.

Without extra carbs – the sprint trick

The 10-second sprint as a non-carb counter-regulator

Bussau et al (2006) demonstrated that a brief maximal sprint after moderate exercise attenuates the post-exercise glucose drop without requiring extra carbohydrate. Iscoe and Riddell showed the same principle for short high-intensity bursts inside continuous moderate sessions. A useful tool when carbohydrate intake is not wanted (weight management, fasted training) or available.

Explorers that pair with this Part

Survive and Thrive – Exercise resources

Three one-page A4 resources, one each for pump, MDI, and AID users.

AID PDF Pump PDF MDI PDF

Back to the guides

Exercise Guide hub CGM Guide hub

This guide is educational. It describes average responses and general principles. It is not medical advice and cannot replace individual clinical guidance from your diabetes care team.