RightRep EN · PT Open the app
BLOG

Can You Train With Type 1 Diabetes? The Insulin Audit: The 9.5-Million File (Up 13% From 8.4 Million in 2021, With 513,000 New Cases This Year and 14.7 Million Projected by 2040), the Inverted Question Where the Library's Uncomplicated Medicine Turns Destabilizing (Riddell's Lancet Consensus Naming Fear of Hypoglycemia the Standing Barrier While T1D Stays At Least as Inactive as the General Population; Colberg's ADA Position File: the Glucagon Alarm Absent, Catecholamines Blunted), the Mode Door (Guelfi's Crossover: Continuous Moderate Dropped Glucose −4.4 mmol/L While Sprint-Mixed Work Dropped It −2.9 Despite MORE Total Work and Left Recovery Flat; Yardley's Review: Anaerobic Work Blunts the Drop and Can Push Glucose Up), the Order Door (Yardley's 12-Runner Crossover: Weights-First Held 9.2 vs Run-First 5.5 mmol/L at the 45-Minute Mark With Post-Exercise Hypo Duration Roughly Halved), the Dosing Door (Rabasa-Lhoret's Table: 50–75% Premeal Reductions Cutting Exercise-Induced Hypoglycemia Incidence by 75%), the Timing Door (MacDonald's Late-Onset File: 48 of ~300 Patients, Nocturnal, 6–15 Hours After Unusual Exertion, More Than Half With Seizures or Unconsciousness), the Muscle File (Milluzzo's Systematic Review: Sarcopenia Relevant in T1D Across All Age Groups — the Argument for Barbells Is Stronger Here, Not Weaker), and the Measurement File (Battelino's Consensus: Time in Range >70%, Time Below <4%)

By RightRep AI Assistant · · 20 min read

There is a fence at the top of this post and it outranks everything below it: type 1 diabetes is an insulin-prescriber partnership, and every insulin-dose change this post discusses is a change made with your care team — never a solo edit against a blog table. Fast-acting carbohydrate is gym equipment in this disease: glucose tabs or sugar within reach every session, before you need them. Severe hypoglycemia — confusion, inability to self-treat, seizure, unconsciousness — is emergency territory, not a rest-between-sets problem; and if you run high with ketones, or you are vomiting, ill, or recovering from illness, the session is off and the numbers go to your team (the training-while-sick audit covers the non-diabetes half of that call). One more fence, because the folk hack around it is dangerous: cutting insulin hard to "make room" for exercise can tip the metabolism into ketoacidosis — the consensus explicitly warns against withholding insulin for sport [3]. With those fences standing, the file underneath is the strangest this library has audited yet, because it inverts the genre: everywhere else — from blood pressure to depression — the audit answers "does exercise help X?" with some flavor of yes. Here the question is can you train at all without the session turning into a metabolic incident, and the answer the evidence gives is a trained yes: exercise is medicine in T1D too, but it is medicine with a pharmacology — mode, order, dose and clock all change what the same hour of work does to your blood.

The short answer: yes — and the training program has a second half

Strip the file to its spine and four findings carry it. The muscle argument is stronger in T1D, not weaker: the 2025 systematic review of sarcopenia in type 1 diabetes found muscle impairment a relevant issue across all age groups, with older age and lower BMI as the main determinants — the case for progressive resistance training stands on T1D physiology harder than on the general population's [2]. The mode and the order are levers: continuous moderate exercise dropped glucose −4.4 mmol/L where the same half-hour with short sprints mixed in dropped it only −2.9 — despite more total work — and left glucose flat through recovery instead of still falling [6]; and in the order crossover, weights-first held runners at 9.2 mmol/L where run-first had already fallen to 5.5 by the 45-minute mark, with post-exercise hypoglycemia duration roughly halved [8]. The dose is a lever with numbers on it: premeal insulin reductions of 50–75%, matched to intensity and duration, cut the incidence of exercise-induced hypoglycemia by 75% in the founding dose-reduction trial — a lever set with the prescriber, but a real one [9]. And the deadliest finding lands on the recovery day, not the training day: late-onset hypoglycemia, typically nocturnal, arriving 6–15 hours after unusually strenuous exercise — documented in 48 of ~300 young patients, in more than half of those cases with loss of consciousness or seizures [10]. That is the audit in one paragraph: train — heavy, even competitively — but the program is only half the program. The other half is the choreography that keeps the glucose curve survivable while the adaptation happens.

What the disease is: the 9.5-million file, and why the pancreas changes the question

The mechanics in a paragraph: in type 1 diabetes the immune system destroys the pancreatic β-cells, the body's insulin production collapses toward zero, and insulin becomes an injected drug dosed by estimate against food, activity, stress and sleep. In the sugar audit — type 2, insulin resistance — exercise raises sensitivity and the numbers improve almost mechanically. In T1D there is no insulin production left to sensitize in the moment's favor: the muscle's glucose uptake still rises steeply during work (largely insulin-independent at contraction), while injected insulin keeps absorbing from the depot under the skin — two clocks that don't know about each other. That is why the same hour of training can end 4 mmol/L down, flat, or up. The scale is the hook: the IDF Atlas 11th edition and T1D Index version 3.0 estimate 9.5 million people living with type 1 diabetes globally in 2025, up 13% from 8.4 million in 2021 — 1.0 million of them children under 15, another 0.8 million aged 15–19 — with 513,000 new cases this year (incidence rising 2.4% in the last year alone), 174,000 premature deaths, 17.2% of them from never being diagnosed in time, and a projection of 14.7 million people by 2040 [1]. This is not a rare pediatric disease; it is a lifelong condition of millions of adults standing in gyms, and the consensus file says they train less than everyone else: people with T1D tend to be at least as inactive as the general population, with fear of hypoglycemia, loss of glycaemic control, and inadequate knowledge around exercise management named as the standing barriers [3].

The broken alarm: why "listen to your body" fails here

The deepest problem is not the insulin math — it is that the body's own low-glucose alarm is compromised. In people without diabetes, falling glucose triggers a precisely sequenced hormonal rescue: glucagon and epinephrine release within minutes, liver glucose pouring out, symptoms — shakiness, sweating, pounding heart — that scream eat now. The ADA position statement files what T1D does to that sequence: the glucagon response to falling glucose is absent in T1D, catecholamine responses are blunted, and both get worse after recent hypoglycemia — the hypoglycemia-associated autonomic failure that turns one bad low into a week of not feeling the next one [4]. In practice: in T1D, the meter and the continuous glucose monitor are the alarm; the feeling is an unreliable narrator, especially overnight and especially after recent lows [4]. This single file explains why the rest of this audit is written the way it is — measured, not felt; pre-planned, not improvised; and always carrying the treatment within arm's reach, because by the time the body speaks clearly it may already be too late to self-treat [10].

The mode door: why lifters have it easier than steady-state runners, acutely

The first lever is the shape of the session itself. The review file puts it plainly: moderate-intensity aerobic exercise increases the risk of hypoglycemia in T1D, while brief high-intensity — anaerobic — work causes smaller declines and can drive glucose up after exercise, so folding small amounts of anaerobic work (short sprints, or resistance exercise) into an aerobic session blunts the drop [5]. The crossover that quantified it took seven participants with T1D through two matched half-hours: continuous moderate exercise at 40% of peak aerobic capacity dropped blood glucose −4.4 ± 1.2 mmol/L, while the same base load with a 4-second sprint every 2 minutes — simulating team-sport patterns — dropped it only −2.9 ± 0.8 mmol/L (P < 0.05), despite the sprint version doing MORE total work [6]. Recovery told the same story from the other side: after the sprint-mixed session glucose held stable for the full hour of recovery; after the continuous session it kept falling [6]. The mechanism file closes the loop — during intermittent high-intensity work the liver keeps producing glucose (driven by the lactate, catecholamine and growth-hormone surge of the hard efforts), while continuous moderate work suppresses production and amplifies muscle uptake at once [6][7]. None of this says steady state is bad; it says the costume of the session has a glycemic signature lifters can use: hard efforts built into the work buy back glucose stability for free [5][6].

The order door: the single most gym-actionable trial in the file

If you train both modalities in one session — and most lifters eventually do conditioning work — the order crossover is the finding to tape inside your locker. Twelve physically active adults with T1D did 45 minutes of running at 60% of peak capacity and 45 minutes of resistance work (three sets of eight across seven exercises) in both orders, on separate days, with glucose tracked during, after, and around the clock by continuous monitoring [8]. Running first: glucose fell through the whole first bout, landing at 5.5 ± 0.7 mmol/L. Weights first: glucose rose through the first bout, sitting at 9.2 ± 1.2 mmol/L at the same 45-minute mark (P = 0.006 between orders) [8]. By the end of both sessions the gap had closed — the running half of the weights-first day did its work afterwards — but the post-exercise accounting still favored the weights-first day: hypoglycemia in the hours after lasted 48 minutes after weights-first vs 105 after run-first, at roughly half the severity (area under the curve 59 vs 112 units·min) — differences the small sample couldn't carry to significance individually, all pointing the same direction [8]. The physiology is the mode door wearing a stopwatch: resistance work is anaerobic work, and its catecholamine/lactate signature holds glucose up while it lasts [5][8]. Protocol: the barbell before the treadmill — same session, calmer curve.

The dosing door: the lever with numbers on it

The founding trial on insulin adjustment for exercise is still the table clinicians work from: eight men with well-controlled T1D on basal-bolus therapy did standardized post-meal exercise of different intensities and durations, with their premeal rapid-acting insulin cut by graded amounts, in randomized crossover [9]. The findings, in the paper's own arithmetic: the full premeal dose raised hypoglycemia risk at every intensity; at light intensity for an hour, a 50% reduction ended exercise essentially flat (−0.62 mmol/L); at moderate intensity, 50–75% reductions kept the end-of-exercise change between −0.39 and +0.49 mmol/L; at 75% of aerobic capacity, a 75% reduction was what it took to stay safe (+0.71 mmol/L) — and across the protocol, these matched reductions cut the incidence of exercise-induced hypoglycemia by 75% [9]. Two honest caveats belong to the finding: the trial's basal-bolus regimen was an older ultralente-plus-lispro design, so the precise percentages are a clinician's starting frame rather than a copy-paste for modern pumps and ultra-rapid analogs; and the fence from the top of this post applies in full — the percentages are set with your prescriber, from your own logged responses, not read off a blog [9]. The point of the trial for a lifter is not the exact table; it is proof the lever exists and moves the outcome by three-quarters.

The timing door: the finding that kills at 3 a.m.

The most sobering file in this audit is also the oldest. In 1987, a prospective case-finding study described postexercise late-onset (PEL) hypoglycemia as its own clinical entity: among roughly 300 young patients with T1D monitored for up to two years, 48 experienced it — typically nocturnal, arriving 6–15 hours after unusually strenuous exercise or play, and in more than half the cases producing loss of consciousness or seizures requiring glucagon, intravenous glucose, or professional care [10]. The pattern details are the ones lifters should hear verbatim: it was not limited to patients in tight metabolic control; it often followed a single bout of exercise; and it showed a taste for exactly two populations — people unaccustomed to exercise doing an unusual session, and trained athletes in transition from an untrained to a trained state (a hard description of anyone starting a new program block); a dozen of the overnight cases occurred in patients whose insulin wasn't even peaking at night [10]. The modern reading is the same as the 1987 reading: after a hard, long, or brand-new session, the muscle keeps refilling its glycogen for hours and the injected insulin keeps absorbing — the recovery day is a hazard window, and the bedtime glucose check after big sessions is not optional diligence but the documented difference-maker [3][10]. Program your training log accordingly: the session entry is not closed when the last set ends.

The muscle file: why the barbell argument is stronger in T1D

Because everything above reads like a hazard list, the honest counterweight is the reason to run this gauntlet at all — and it is stronger in T1D than out of it. The 2025 systematic review (15 studies, PRISMA-conducted) concluded that sarcopenia and skeletal-muscle impairment are a relevant issue in type 1 diabetes across all age groups, with older age and lower BMI the main determinants [2]. Muscle is this disease's metabolic buffer and its insurance policy: the largest glucose-disposing tissue in the body, the storage the recovery-day glycogen refill draws on, and — as the general-population files on weekly set volume and progressive overload lay out — trainable by exactly the same stimulus rules as everyone else's [2]. What changes in T1D is not the stimulus; it is the fuel choreography around it. The practical translation: don't let the hypoglycemia file argue you out of the weight room — let it argue you into doing it first in the session [8], with carbs in reach, and into treating the training log as the second half of the prescription [11].

The measurement file: the numbers that make it safe

The consensus on continuous-glucose-monitoring metrics gave T1D the vocabulary this whole audit runs on: targets are judged on the curve, not the feeling — time in range above 70% (3.9–10.0 mmol/L), time below 3.9 mmol/L under 4%, and time below 3.0 mmol/L under 1% [11]. For training specifically, the consensus frame runs the other direction too: the pre-exercise reading is a gate, not a curiosity — a working corridor in the neighborhood of 5.0–13.9 mmol/L for starting a session, below which you treat with carbohydrate first and above which you pause and assess (with ketones on the agenda when high), the precise bands per exercise mode set out in the consensus and individualized with your own team [3][11]. The measurement file is also what turns the other levers from folklore into protocol: mode [6], order [8] and dose [9] were all measured effects — and your own version of them only exists if the session outcomes are logged next to the glucose curve, which is the exact intersection this app's session log is built for.

The program-proof: EXTOD, and why education is the intervention

That training itself is achievable and worth engineering for is not just an inference from physiology — it has trial infrastructure behind it. The EXTOD pilot randomized adults within three months of T1D diagnosis to usual care or a year of monthly exercise consultation, specifically to test whether exercise preserves the surviving β-cell function of the honeymoon period — feasibility first, with recruitment, dropout and adherence measured as the primary outcomes, on the path to a definitive trial [12]. And when the qualitative file asked why so few people with T1D exercise, the same group built the answer into an intervention: EXTOD Education, a structured education programme for exercising with T1D developed through the UK Medical Research Council framework from the evidence review, behavior-change taxonomy, and stakeholder interviews — because the barrier file says knowledge around exercise management, not motivation, is the missing piece [3][13]. The barrier audit and the education build are the same fact from two sides: the reason most people with T1D don't train is not that they can't — it is that nobody taught them the choreography [3][13].

The counterfile: what exercise is not, in this disease

Honesty section, four entries. Exercise is not insulin: no dose of training replaces the drug, and the EXTOD β-cell hypothesis is a pilot-stage question, not a promise [12]. The T2D remission file does not transfer: the sugar audit's weight-loss-driven remission ladder belongs to insulin resistance; established T1D has no β-cell mass left to rescue by lifestyle, and training's proven seats here are fitness, muscle, cardiovascular risk and quality of life — real ones, different ones [1][3]. "Exercise stabilizes everything" is false in the acute window: the first weeks of any new program are typically the rockiest glycemic period a T1D lifter lives through — new sessions are precisely the "unusually strenuous exercise" MacDonald's late-onset file fed on [10] — so the correct response to starting a program is more checking, not less, until the pattern is learned [3][11]. And the extremes fail in both directions: moderate continuous work with full insulin on board is the classic hypo setup [5][9], while cutting insulin hard to avoid it opens the DKA door the consensus explicitly warns about [3] — the safe zone is engineered in the middle, with carbs, dose adjustments and measurement, not at either pole.

So what do you actually do? The six-rule protocol

Assemble the file and the protocol writes itself — as a stack, with the care team in the loop.

1. Keep training — and keep lifting. The sarcopenia file says T1D needs resistance work more than the general population [2]; the stimulus rules (sets per week, progressive overload) are unchanged — only the fuel choreography is new [2][5].

2. Lift before you run. In combined sessions, weights-first held glucose at 9.2 vs 5.5 mmol/L at the 45-minute mark and roughly halved post-exercise hypoglycemia duration [8] — the single cheapest glycemic upgrade in the file.

3. Engineer the session, not just the effort. Hard efforts built into the work — sprints, heavy sets, intervals — blunt the glucose drop (−2.9 vs −4.4 mmol/L with more work done) and flatten the recovery fall [5][6]; steady continuous work earns a plan, not a ban.

4. Learn your insulin levers with your prescriber. Premeal and basal adjustments around training are validated, quantified tools — 50–75% matched reductions cut exercise hypos by 75% in the founding trial [9] — set with your care team from your own logged responses, never solo-edited [3].

5. Carry fast carbs every session — and watch the second half. Glucose tabs are equipment [4]; treat early, treat measured, re-check. And the session stays open for 6–15 hours: hard, long, or brand-new sessions get a bedtime glucose check, because the deadliest low in this file arrived the night after [10].

6. Judge months on the curve and the log together. Time in range >70%, time below range <4% [11], read next to the training log — because the insulin adjustments that make rule 4 safe are data-driven, and the pattern only exists if it's recorded. Illness pauses the program, always — when to train sick has the general file, and in T1D illness gets the meter and the care team involved first.

How RightRep fits into this

The choreography above runs on records — which is the shape this app is built around:

  • The session log as the insulin file's data half. Mode, order, load, duration, and how each session ended — logged per set, per session — is the dataset rule 4's adjustments are made from; the difference between guessing and titrating is the record [9][11].
  • Program structure you can hold constant. The sarcopenia file's answer is a progressive program [2]; this app's structured programming holds the stimulus side steady while you and your team tune the fuel side — one moving variable at a time.
  • The between-visits ledger. The care team's view of your training is the export; the tracking argument applies double when the log feeds dose decisions [11].

The bottom line

  • 9.5 million people live with type 1 diabetes in 2025, up 13% from 8.4M in 2021 — 1.0M of them children; 513,000 new cases this year; 14.7 million projected by 2040 [1].
  • T1D trains less than the general population, with fear of hypoglycemia the named standing barrier — a knowledge problem with an education-shaped answer [3][13].
  • The body's low-glucose alarm is broken in T1D — glucagon absent, catecholamines blunted, worse after recent lows: the meter is the alarm [4].
  • Mode: continuous moderate −4.4 vs sprint-mixed −2.9 mmol/L (more total work, flat recovery) [6]; mechanism — the liver keeps producing during intermittent hard effort [7].
  • Order: weights-first 9.2 vs run-first 5.5 mmol/L at 45 min (P = 0.006), post-exercise hypo duration 48 vs 105 min [8].
  • Dose: matched 50–75% premeal reductions cut exercise-induced hypoglycemia incidence by 75% [9].
  • Timing: late-onset hypoglycemia in 48 of ~300 patients — nocturnal, 6–15 h after unusual exertion, over half with seizures or unconsciousness [10].
  • Muscle: sarcopenia relevant in T1D across all ages — the resistance-training case is stronger here [2].
  • Measurement: time in range >70%, below range <4% and <1% under 3.0 mmol/L — the curve, not the feeling [11].
  • Program-proof: EXTOD's pilot and structured-education build put trial infrastructure behind trainable T1D [12][13].

Keep the prescriber in the loop, keep the glucose tabs in reach, put the barbell before the treadmill — and train like the 9.5-million file says you can, at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Ogle GD, Wang F, Haynes A, et al. Global type 1 diabetes prevalence, incidence, and mortality estimates 2025: Results from the International Diabetes Federation Atlas, 11th Edition, and the T1D Index Version 3.0. Diabetes Research and Clinical Practice, 2025;225:112277. pubmed.ncbi.nlm.nih.gov
  2. Milluzzo A, et al. Sarcopenia in type 1 diabetes mellitus: a systematic review. Diabetes Research and Clinical Practice, 2025;225:112399. pubmed.ncbi.nlm.nih.gov
  3. Riddell MC, Gallen IW, Smart CE, et al. Exercise management in type 1 diabetes: a consensus statement. The Lancet Diabetes & Endocrinology, 2017;5(6):522–536. pubmed.ncbi.nlm.nih.gov
  4. Colberg SR, Sigal RJ, Yardley JE, Riddell MC, Dunstan DW, Dempsey PC, Horton ES, Castorino K, Tate DF. Physical Activity/Exercise and Diabetes: A Position Statement of the American Diabetes Association. Diabetes Care, 2016;39(11):2065–2079. pubmed.ncbi.nlm.nih.gov
  5. Yardley JE. Resistance exercise in type 1 diabetes. Canadian Journal of Diabetes, 2013;37(6):435–438. pubmed.ncbi.nlm.nih.gov
  6. Guelfi KJ, Jones TW, Fournier PA. The decline in blood glucose levels is less with intermittent high-intensity compared with moderate exercise in individuals with type 1 diabetes. Diabetes Care, 2005;28(6):1289–1294. pubmed.ncbi.nlm.nih.gov
  7. Guelfi KJ, Ratnam N, Tweedie MG, Cameron F, Smith G, Jones TW. Effect of intermittent high-intensity compared with continuous moderate exercise on glucose production and utilization in individuals with type 1 diabetes. American Journal of Physiology-Endocrinology and Metabolism, 2007;292(3):E865–E870. pubmed.ncbi.nlm.nih.gov
  8. Yardley JE, Kenny GP, Perkins BA, Riddell MC, Malcolm J, Boulay P, Khandwala F, Sigal RJ. Effects of performing resistance exercise before versus after aerobic exercise on glycemia in type 1 diabetes. Diabetes Care, 2012;35(4):669–675. pubmed.ncbi.nlm.nih.gov
  9. Rabasa-Lhoret R, Bourque J, Ducros F, Chiasson JL. Guidelines for premeal insulin dose reduction for postprandial exercise of different intensities and durations in type 1 diabetic subjects treated intensively with a basal-bolus insulin regimen (ultralente-lispro). Diabetes Care, 2001;24(4):625–630. pubmed.ncbi.nlm.nih.gov
  10. MacDonald MJ. Postexercise late-onset hypoglycemia in insulin-dependent diabetic patients. Diabetes Care, 1987;10(5):584–588. pubmed.ncbi.nlm.nih.gov
  11. Battelino T, Danne T, Bergenstal RM, et al. Clinical Targets for Continuous Glucose Monitoring Data Interpretation: Recommendations From the International Consensus on Time in Range. Diabetes Care, 2019;42(8):1593–1603. pubmed.ncbi.nlm.nih.gov
  12. Narendran P, Jackson KC, Solomon TPJ, et al. Exercise to preserve β-cell function in recent-onset Type 1 diabetes mellitus (EXTOD) — a randomized controlled pilot trial. Diabetic Medicine, 2017;34(11):1521–1531. pubmed.ncbi.nlm.nih.gov
  13. Narendran P, and the EXTOD Education study group. Development of a group structured education programme to support safe exercise in people with Type 1 diabetes: the EXTOD education programme. Diabetic Medicine, 2020;37(6):957–964. pubmed.ncbi.nlm.nih.gov