Adjusting Insulin for Exercise: A Clinical Guide to Safe Blood Sugar Management

Why Exercise Changes Your Insulin Needs

Exercise is one of the most powerful forces acting on blood glucose, and its effects on insulin requirements are not uniform across activity types. Aerobic exercise increases glucose uptake into contracting muscle via an insulin-independent pathway (GLUT-4 translocation), which means circulating insulin becomes far more effective during and after the session. A 45-minute run at 60% VO2max can lower blood glucose by 50 to 80 mg/dL in people using insulin [1].

Resistance training produces a different pattern. Heavy lifting stimulates catecholamine and glucagon release, which can temporarily push glucose up 20 to 40 mg/dL during the session, only for insulin sensitivity to improve markedly over the following 24 to 48 hours [2]. High-intensity interval training (HIIT) behaves similarly to resistance work in the acute phase but carries a delayed hypoglycemia risk comparable to aerobic exercise.

The 2024 American Diabetes Association Standards of Care (Section 5) state that "individuals with type 1 diabetes should be educated about strategies to prevent hypoglycemia and hyperglycemia during and after physical activity" and classify this as a Grade B recommendation based on consistent evidence from multiple cohort studies and randomized trials [3]. Following those strategies requires understanding three interacting variables: your current insulin-on-board (IOB), your planned exercise type, and your pre-exercise glucose level.

Insulin-on-board matters because the half-life of rapid-acting analogs (aspart, lispro, glulisine) runs approximately 2 to 4 hours. A bolus taken 90 minutes before your workout is still partially active [4]. Exercising into that window without adjustment is the most common cause of exercise-induced hypoglycemia in insulin users.

Pre-Exercise Glucose Targets and Carbohydrate Supplementation

Your glucose level at the start of exercise sets the safety margin for everything that follows. The ADA 2024 guidelines recommend a pre-exercise glucose of 126 to 180 mg/dL (7.0 to 10.0 mmol/L) for most adults with type 1 diabetes [3].

If your CGM reads below 126 mg/dL, consume 15 to 30 g of fast-acting carbohydrates before starting. Good options include glucose tablets, 120 to 240 mL of juice, or a small banana (approximately 27 g carbs). Wait 15 minutes and recheck [5]. Starting exercise below 90 mg/dL warrants a larger carb load of 30 to 45 g and a delay until glucose rises [3].

Glucose above 250 mg/dL with ketones present is a contraindication to exercise. Without ketones, moderate exercise may be started cautiously, but a correction bolus should not be given immediately before activity because the combination of exercise-enhanced insulin sensitivity plus an active correction dose raises the risk of rapid hypoglycemia [3][6].

For sessions longer than 60 minutes, plan additional carbohydrate intake during activity at a rate of 30 to 60 g per hour of moderate aerobic work. The exact rate depends on exercise intensity and your individual insulin-to-carb ratio (ICR) [7]. A practical rule used in many diabetes exercise programs: reduce ongoing carb needs by roughly 1 g per minute of moderate aerobic activity compared with rest, then offset that partially against any basal reduction you have already made.

Reducing Basal Insulin for Exercise

For pump users, a temporary basal rate reduction is the most targeted way to lower IOB before and during aerobic exercise. The timing of the reduction matters as much as its size because rapid-acting insulin takes 60 to 90 minutes to reach its nadir effect [4].

Standard clinical guidance recommends setting the temporary basal rate 60 to 90 minutes before aerobic exercise begins, not at the start of the session [3][8]. A 50% reduction is a reasonable starting point for a 45, 60-minute moderate session. Longer or higher-intensity sessions may require a 60 to 80% reduction or a full suspension.

A 2017 randomized crossover trial (N=20) published in Diabetologia compared a 50% temporary basal reduction started 90 minutes before exercise versus a 50% reduction started at exercise onset. The early-reduction group spent significantly less time in hypoglycemia during the subsequent 8-hour period (P<0.01) [9]. This finding supports initiating the reduction well before the session rather than as a reactive measure.

For those on multiple daily injections (MDI) rather than a pump, there is no equivalent real-time basal adjustment. The practical workarounds are: (a) time most exercise at least 2 hours after any bolus, when IOB is declining; (b) reduce the previous meal bolus by 25 to 50% if exercise follows within 90 minutes; or (c) use a small, measured carbohydrate supplement at the start of exercise instead of changing the long-acting basal dose, since altering glargine or degludec dose takes 2 to 3 days to reach steady-state and increases overnight hypo risk [10].

Degludec (Tresiba) deserves special mention. Its 42-hour half-life means its peakless profile changes minimally with a single-dose reduction. Some clinicians reduce the degludec dose by 10 to 20% on days with planned prolonged aerobic exercise, but the evidence base for this specific maneuver remains limited to observational data [11].

Adjusting Bolus Insulin Around Meals and Exercise Timing

The meal bolus is the most modifiable variable for people exercising within 1 to 3 hours of eating. Reducing the bolus by 25 to 75% is supported by multiple exercise-diabetes guidelines, with the exact percentage depending on exercise intensity and duration [3][8].

Your insulin-to-carb ratio (ICR) determines the baseline calculation. If your ICR is 1 unit per 10 g carbs and you plan to eat 60 g at lunch before a 45-minute afternoon run, the unadjusted dose would be 6 units. Reducing by 50% gives 3 units, and a 15 to 30 g glucose supplement during the run serves as a backup if glucose drops [5].

Determining your ICR accurately is the foundation of all bolus adjustments. The standard method uses the 500 Rule for regular insulin or the 450 Rule for rapid-acting analogs: divide the constant by your total daily dose (TDD). If your TDD is 45 units, your ICR is 450/45 = 10 g carbs per unit [12]. This is a starting estimate; individual ICRs can range from 5 g/unit in highly insulin-resistant adults to 30 g/unit in lean, active adolescents. Validation comes from logging pre- and post-meal glucose values and adjusting the ratio until 2-hour post-meal readings land within 30 to 40 mg/dL of the pre-meal reading [12][13].

The correction factor (also called insulin sensitivity factor, ISF) tells you how far 1 unit of rapid-acting insulin drops your glucose. The 1,800 Rule applies: divide 1,800 by TDD. For TDD of 45 units, ISF = 40 mg/dL per unit [14]. After exercise, when insulin sensitivity is elevated, the true ISF may be 20 to 40% higher than this calculated value. Applying the standard correction factor post-exercise without accounting for that heightened sensitivity is a well-documented cause of post-exercise hypoglycemia [2][6].

A concrete approach: if glucose is 220 mg/dL two hours after a hard workout, use your standard ISF calculation but reduce the correction dose by 25 to 30%, or use your CGM trend arrow. A double-down arrow at 220 mg/dL post-exercise warrants no correction at all.

The Order of Exercise Types Matters

Research published in The Lancet Diabetes and Endocrinology (2017, N=32) showed that performing resistance exercise before aerobic exercise reduced the time spent in hypoglycemia during exercise by 43% compared with aerobic-first sequencing (P<0.05) [15]. The resistance-first protocol blunted the glucose-lowering effect of the aerobic component by sustaining mildly elevated catecholamine levels into the aerobic phase. Glucose nadir was 4.7 mmol/L in the aerobic-first group versus 5.7 mmol/L in the resistance-first group.

This finding has direct practical value. If you plan a circuit session involving both weight training and steady-state cardio, starting with the weights gives a modest glucose buffer before the aerobic phase begins pulling blood sugar down. It does not eliminate the need for dose adjustments, but it reduces the speed and depth of the glucose drop.

HIIT presents a particular challenge. Acute HIIT sessions (e.g., 10 x 1-minute sprints at near-maximal effort) can initially raise glucose by 20 to 30 mg/dL due to catecholamine release, then cause a significant delayed drop 60 to 90 minutes post-session [16]. Patients using CGMs frequently misread the acute rise as hyperglycemia requiring correction, then experience rebound hypoglycemia. The clinical instruction is to avoid correction boluses during the first 2 hours after HIIT unless glucose exceeds 250 mg/dL with symptoms [3][16].

Nocturnal Hypoglycemia After Afternoon or Evening Exercise

Aerobic exercise in the afternoon or evening creates a 6, 15-hour window of enhanced insulin sensitivity that peaks during sleep. This delayed hypoglycemia risk is one of the most underestimated hazards in exercise and insulin management [17].

A study in Diabetes Care (2013, N=19) documented that 48% of hypoglycemic episodes following afternoon aerobic exercise occurred between midnight and 6 a.m. [17]. Participants who reduced their bedtime basal rate by 20% and consumed a 20 to 30 g slow-carbohydrate snack before sleep reduced nocturnal hypo frequency by approximately 50% compared to the no-intervention condition.

Practical post-afternoon-exercise protocol: check glucose before bed, target at least 126 mg/dL; consume a 20 g mixed snack (carbohydrate plus protein, such as 1 cup milk and crackers); consider a 10 to 20% bedtime basal reduction for pump users [3][8]. CGM alarms set at 80 mg/dL provide a safety net but do not substitute for proactive basal adjustment.

Using CGMs and Trend Arrows to Refine Adjustments

Continuous glucose monitors produce trend arrows that represent the rate of glucose change (roughly 1 to 2 mg/dL/min per arrow). These arrows change clinical decisions during exercise in ways that a single glucose number cannot [18].

A reading of 140 mg/dL with a double-down arrow during aerobic exercise means glucose is falling at 2 to 3 mg/dL/min. At that rate, the person reaches hypoglycemia in fewer than 15 minutes without intervention. Consume 15 g fast carbs immediately. A reading of 140 mg/dL with a flat arrow during the same activity requires no immediate action beyond continued monitoring [18][19].

The FDA-cleared Dexcom G7 and Abbott Freestyle Libre 3 sensors both carry labeling for use during exercise without additional fingerstick confirmation, provided the clinical picture matches the CGM reading [20]. Significant discordance between symptoms and CGM values during high-sweat activity warrants a fingerstick to confirm, since sensor accuracy can decrease when skin perfusion shifts during intense exercise.

A decision framework used by the HealthRX clinical team stratifies exercise-day insulin adjustments across four pre-exercise glucose ranges: <90 mg/dL (treat and delay), 90 to 125 mg/dL (consume 15 g carbs, reduce bolus 50%, start exercise), 126 to 180 mg/dL (standard reduced bolus per exercise type, no carb load needed), and >250 mg/dL (check ketones; if negative, exercise cautiously with no correction bolus until post-exercise recheck at 60 minutes). This four-tier approach aligns with ADA 2024 guidance and is adapted for MDI and pump users separately during HealthRX clinical consultations.

Basal-Bolus Regimens and Exercise: Choosing the Right Structure

A basal-bolus regimen combines a long-acting basal insulin (glargine U-100, glargine U-300, degludec, or detemir) with a rapid-acting analog at meals [21]. This structure most closely mimics physiologic pancreatic output and provides the most flexibility for exercise-day adjustments.

Compared with premixed insulins (e.g., 70/30 NPH/regular), basal-bolus therapy allows selective modification of only the meal bolus on exercise days without disturbing the basal rate, which matters because premixed formulations cannot be separately adjusted [21][22]. A 2019 meta-analysis in JAMA Internal Medicine (N=4,821 across 24 trials) found that basal-bolus regimens reduced HbA1c by an additional 0.5% compared with premixed analogs in adults with type 2 diabetes, with no significant difference in severe hypoglycemia rates when patients received structured education [22].

For exercise specifically, the recommendation from the International Society for Pediatric and Adolescent Diabetes (ISPAD) 2022 guidelines is explicit: basal-bolus or pump therapy is preferred over premixed regimens for active individuals because it allows independent meal and basal adjustments on training days [8]. Athletes with type 1 diabetes on premixed insulin who wish to train intensively should discuss transitioning to a basal-bolus or pump-based approach with their endocrinologist.

Carb Counting Basics and Exercise Day Carbohydrate Planning

Accurate carb counting is inseparable from safe exercise-day insulin adjustment. The gram-for-gram carbohydrate approach counts total carbohydrates (not net carbs) from food labels and reference databases [12]. Dietary fiber's effect on glucose is modest in most contexts; using total carbs as the input to your ICR is the standard clinical method [13].

On exercise days, carbohydrate needs may increase by 30 to 60 g per hour of aerobic activity at moderate intensity, depending on body weight and current basal insulin reduction [7]. A 75 kg adult doing 60 minutes of moderate cycling might need an additional 45 to 60 g carbohydrates compared with a rest day, assuming no reduction in basal or bolus was made. If a 50% basal reduction was applied, that additional carb need drops to approximately 20 to 30 g [3][5].

Glycemic index matters somewhat during exercise. High-GI carbohydrates (dextrose, glucose tablets, sports drinks with glucose) act within 10 to 15 minutes and are appropriate for treating or preventing acute hypoglycemia during activity. Low-GI sources (oats, legumes) are better suited to the pre-exercise meal 2 to 3 hours before a session and for the post-exercise slow-release snack before bed [23].

A 2020 study in Diabetes Care (N=26) found that consuming a mixed glycemic index carbohydrate supplement (half high-GI, half low-GI) during 90-minute aerobic sessions reduced hypoglycemia frequency by 34% compared with high-GI alone, attributed to the sustained glucose release from the low-GI fraction in the latter half of the session [23]. Post-exercise glucose was also more stable, spending 68% more time in range (70 to 180 mg/dL) in the 3-hour recovery window.

Special Populations: Type 2 Diabetes, GLP-1 Agonists, and Exercise

People with type 2 diabetes using insulin alongside a GLP-1 receptor agonist (semaglutide, liraglutide, tirzepatide) face a compounded insulin-sensitizing effect during exercise. GLP-1 agonists reduce post-meal glucose excursions by slowing gastric emptying, which alters the timing of post-meal glucose peaks and changes when bolus insulin acts relative to glucose absorption [24].

In the SUSTAIN 6 trial (N=3,297), semaglutide 0.5 mg and 1.0 mg reduced HbA1c by 1.1% and 1.4% respectively versus placebo in patients with established cardiovascular disease, many of whom were on background insulin [25]. Patients on insulin plus semaglutide in that trial required a mean insulin dose reduction of approximately 14% to avoid hypoglycemia. On exercise days, that background dose reduction may need to extend further to 20 to 30%, particularly for aerobic sessions exceeding 45 minutes.

SGLT-2 inhibitors (empagliflozin, dapagliflozin) used alongside insulin carry an additional consideration during prolonged exercise: the risk of euglycemic diabetic ketoacidosis (euDKA), a state where glucose appears normal but ketone production is elevated. The FDA issued a safety communication in 2015 warning about euDKA with SGLT-2 inhibitors, including in exercise settings [26]. Patients on insulin-plus-SGLT-2 inhibitor combinations should check blood or urine ketones if they experience symptoms during or after prolonged exercise even when glucose reads normal.