Exercise Dose and HbA1c in Youth With Type 1 Diabetes

The biggest HbA1c changes came with longer, combined programs

For children and adolescents with type 1 diabetes, exercise training appears to lower HbA1c—but the average result hides an important pattern. In the best available randomized-trial synthesis, programs lasting at least 24 weeks, with sessions lasting at least 60 minutes and combining aerobic and resistance exercise, produced the largest observed reductions. Under those combined conditions, the estimated HbA1c reduction reached as much as 1.30 percentage points.

That is a clinically meaningful amount if it holds up. But it is not a precision prescription carved into a treadmill console. The evidence comes from small subgroup analyses, and the 24-week and 60-minute cutoffs were categories in the meta-analysis, not experimentally proven biological thresholds.

For families and clinicians, the practical message is narrower and more useful than "exercise helps." Sustained, supervised, mixed-mode training looks more promising than brief or single-mode programs for glycemic control in youth with type 1 diabetes. High-intensity exercise also requires active glucose management, because a lower HbA1c average does not guarantee a predictable glucose response after any individual session.

What the youth trials found

A 2023 systematic review and meta-analysis by García-Hermoso et al. pooled 14 randomized controlled trials involving 509 participants aged 6 to 18 with type 1 diabetes. Across all exercise interventions, HbA1c fell by a mean difference of 0.62 percentage points versus control. The standardized effect was Hedges' g = −0.38 (95% confidence interval, −0.66 to −0.11), which was statistically significant because the confidence interval excluded zero.

That average is modest. The subgroup results suggest that program design may account for much of the difference:

• Programs lasting at least 24 weeks had an HbA1c effect of g = −0.92 (95% CI, −1.44 to −0.40).
• Sessions lasting at least 60 minutes had an effect of g = −0.71 (95% CI, −1.05 to −0.08).
• Concurrent aerobic-plus-resistance training had the largest modality-specific estimate: g = −0.63 (95% CI, −1.05 to −0.21).
• High-intensity exercise had an effect of g = −0.43 (95% CI, −0.83 to −0.03).

All four estimates favor exercise, but their confidence intervals are wide. The interval for the 60-minute subgroup, for example, extends from a small effect to a much larger one. These numbers show a plausible dose-response pattern, not an exact forecast of what one teenager's HbA1c will do.

Why combined training is appealing—and what the study cannot prove

Concurrent training means including both aerobic and resistance exercise in the same program. It had the largest modality-specific effect in the youth meta-analysis. That result fits with findings in other populations: a network meta-analysis in people with prediabetes found an advantage for concurrent exercise on glycemic control and insulin resistance, while a 2024 meta-analysis in sedentary adults reported improved glucose-metabolism markers with combined training.

Those studies support plausibility, not a direct transfer of effect size. Prediabetes and type 2 diabetes retain endogenous insulin secretion; type 1 diabetes does not. Exercise can improve peripheral insulin sensitivity in all of these groups, but insulin dosing and the risks of hypo- and hyperglycemia make the practical physiology of type 1 diabetes different.

The youth analysis also did not report separate effect sizes for aerobic-only and resistance-only training. So it cannot establish that combining the two modalities caused the larger effect. Concurrent programs may have differed in total exercise volume, supervision, baseline HbA1c, or insulin-management support. "Aerobic plus resistance" is the most promising format in this evidence, but the additive contribution of each component remains untested.

The 24-week and 60-minute numbers are guideposts, not cliffs

The strongest-looking estimates came from programs of at least 24 weeks and sessions of at least 60 minutes. It would be a mistake to infer that a 23-week program or a 55-minute session suddenly fails. The meta-analysis grouped trials into those bins; it did not compare 20 versus 24 weeks or 50 versus 60 minutes in a dose-response experiment.

Still, the direction makes practical sense. HbA1c reflects roughly the prior three months of glycemia, so very short interventions have limited time to change it. More time spent training may also improve fitness, confidence, routine, and the ability to manage exercise safely. But the available trials do not identify the minimum effective duration or session length.

The evidence supports building a sustained program. It does not support treating 24 weeks and 60 minutes as biological switch points.

Adults with type 1 diabetes should not borrow the youth effect size

A 2024 meta-analysis by Lazić et al. found that high-intensity interval training in adults with type 1 diabetes improved cardiorespiratory fitness and produced small favorable effects on HbA1c and 24-hour mean glucose. The direction agrees with the youth findings, but the glycemic effect in adults was smaller. The review concluded that limited research leaves the most effective glycemic-control protocol unclear.

That gap matters. The large youth estimates may reflect developmental differences in muscle insulin sensitivity or glucose handling, but neither review experimentally isolated those mechanisms. Older-adult exercise literature describes training-related changes in GLUT4 translocation, mitochondrial biology, and insulin signaling, yet a 2025 review of older adults addresses a population with age-related muscle loss and different metabolic constraints. It cannot validate a youth type 1 diabetes prescription.

Evidence from type 2 diabetes also cannot settle the question. A 2024 meta-analysis of HIIT protocols in type 2 diabetes supports favorable glycemic effects of high-intensity training, but type 2 diabetes has a different disease mechanism and different exercise-related insulin dynamics.

HbA1c misses the part that can make exercise difficult

HbA1c is an average, not a session-by-session safety report. A 2023 systematic review by McClure et al. found inconsistent acute blood-glucose responses to high-intensity interval exercise in adults with type 1 diabetes. Some studies suggested less hypoglycemia than moderate continuous exercise; others reported post-exercise hyperglycemia, consistent with catecholamine-driven hepatic glucose release.

That variability is not a footnote. Insulin on board, carbohydrate intake, exercise timing, prior activity, and the individual response can change the result. A lower HbA1c over months can coexist with difficult glucose swings around workouts. Continuous glucose monitoring and individualized insulin and carbohydrate planning are therefore part of implementation, especially for high-intensity sessions.

Adherence also matters when the useful program length may be measured in months. In a small randomized trial of 19 sedentary adults with type 1 diabetes, Alarcón-Gómez et al. reported that six weeks of HIIT improved health-related quality of life, sleep quality, exercise motivation, and enjoyment compared with control. Those outcomes do not prove a long-term HbA1c benefit, but they may matter for completing a program long enough to test one.

What to do with this evidence

For youth with type 1 diabetes, the current randomized evidence favors a program that combines aerobic and resistance exercise and continues for months rather than weeks. Sessions of at least 60 minutes and programs of at least 24 weeks produced the largest observed HbA1c effects, with an estimated maximum reduction of 1.30 percentage points under the combined favorable conditions.

Use those figures as an evidence-based ceiling, not a promise. The full analysis includes only 14 trials and 509 young people, the subgroup estimates are imprecise, and the apparent thresholds may partly reflect how researchers grouped studies. For adults with type 1 diabetes, expect smaller average HbA1c effects from HIIT and no validated equivalent dose target. The program that lowers HbA1c is useful only if its glucose management is safe enough to repeat next week.