Type 1 Diabetes, Weight and Insulin Resistance
What insulin resistance is, why weight and type 1 interact, and what the evidence on the adjunct medicines really says. About health, not appearance. Read it plainly with Jude, earn your way into the evidence with Grace, then the full picture with John. Stop wherever you have enough.
How we teach: three rules, borrowed from Taleb
You earn each level by showing you understand it, not by scrolling past it. We only teach what we would use on ourselves and the people we love.
Understanding beats memory and luck, so the checks reshuffle every time you retry. A pass means you got it, not that you guessed it. And we teach you to tell a trend (signal) from one reading (noise).
We give you the scaffolding and get out of your way. Roam where your curiosity leads, go as deep as you want, and ask Grace anything. We will not teach a bird how to fly.
Want to talk this through for your own pattern, your own units and your own insulin trend? Ask Grace, then take it to your care team.
One page, three depths
This guide compounds: each layer rests on the one beneath it. Read Jude’s plain version, then pass a short understanding check to open Grace, then another to open John. You can roam freely within a layer; you cannot skip ahead a layer, because the next one would not make sense and you would be standing on a gap.
The whole thing, in plain words
In type 1 diabetes the body makes little or no insulin, so insulin is taken from outside. Insulin resistance means the body answers that insulin less well than it used to, so the same job needs a bit more insulin. It is common in type 1, and it tends to creep up slowly over the years. None of this is a personal failing; it is biology, and biology can be nudged.
There is a simple, kind way to track it. Add up all the insulin you use in a day, then divide by your weight in kilograms. A lower number means the insulin is working easily; a higher number means it is working harder. Watching that number drift over months tells you more than any single day ever could.
Weight and type 1 are linked, but not in the way people sometimes assume. This is about health and how hard insulin has to work, never about appearance, and never about blame. Carrying extra body fat, especially around the middle, tends to make insulin work harder, and taking more insulin can in turn make weight harder to shift; the two can feed each other. The good news is that the same handful of habits help both at once.
Insulin used per kilogram is a friendly gauge, not a verdict. These are bands, not exact lines, and they shift with how much insulin your own body still makes. Read the trend over months, set the aim with your team, and be kind about it.
What helps? The same things that help most people feel well: moving your body, sleeping enough, eating in a way you can keep up, and looking after stress. Where extra body fat is part of the picture, losing a little can ease things a lot; you do not need to lose a great deal for insulin to start working better. There are also medicines that can help some people, and we cover those honestly further down, with the evidence, not the hype.
Does this match the life of the person living it? If a number on a scale or a dose ever makes you feel ashamed, the number is being used wrongly. This is about how hard insulin has to work, not about how you look. Aim for steady and liveable, and be on your own side.The Pemberton lens, lived recognisability, one of the four GNL appraisal lenses.
Why it happens, and the seven levers
Type 1 carries a built-in metabolic disadvantage
In a body without diabetes, insulin is released into the portal vein and reaches the liver first, where it switches off glucose output directly. In type 1, insulin is injected under the skin and reaches the rest of the body before the liver, so peripheral insulin levels run roughly four to eight times higher than they would otherwise.1 That chronic peripheral over-exposure is itself a driver of insulin resistance, which is why type 1 is metabolically more complex than glucose alone. Higher insulin resistance in type 1 also tracks with higher cardiovascular and all-cause risk at the population level (the so-called double-diabetes phenotype), which is why this is a health matter, not an appearance one.2
It works at eight sites at once
Ralph DeFronzo’s “ominous octet” names eight places insulin resistance and high glucose are driven from. In type 1 several of them compound each other, so it is rarely a single-cause problem.3
| Site | What goes on |
|---|---|
| Muscle | Fat stored inside muscle cells blunts insulin signalling, so muscle takes up less glucose. |
| Liver | The liver keeps releasing glucose after meals despite rising insulin. |
| Fat cells | High insulin locks fat inside adipose tissue, making fat harder to mobilise. |
| Glucagon | Too much glucagon keeps the liver pushing glucose out. |
| Gut hormones | A weaker GLP-1 and GIP response blunts the post-meal insulin signal. |
| Kidney | The kidney reabsorbs more glucose back into the blood. |
| Brain | Altered appetite signalling nudges energy intake up. |
| Beta cells | In type 1 these are largely gone, removing the body’s own fine adjustment. |
Seven levers, lifestyle first, medicines on top
The honest organising principle: most of the everyday shift comes from movement, fat loss where it applies, and reducing glucose exposure. Medicines can be powerful, but they tend to work best layered onto those foundations, not used in place of them.4
| Lever | Mainly acts on | What the evidence suggests |
|---|---|---|
| 1. Activity | Muscle, liver, appetite | The single most powerful lever; addresses more sites than any one drug. Often needs planned insulin reductions so it does not become a hypo generator. |
| 2. Fat loss where it applies | Liver, muscle (ectopic fat) | Even 5 to 10% can ease insulin resistance meaningfully; the aim is fat loss with muscle kept, via protein and resistance training. |
| 3. Nutrition | Liver, fat cells | Limiting liquid sugar and heavy saturated fat reduces the glucose and fat load that feed resistance. |
| 4. GLP-1 medicines | Appetite, glucagon, liver | Off-label in type 1; meaningful effects on insulin need, time in range and weight (see below). |
| 5. SGLT-2 inhibitors | Kidney | Lower glucose by excreting it; carry a real risk of ketoacidosis in type 1, so used only with strict ketone rules and care-team oversight. |
| 6. Pioglitazone | Fat distribution | Moves fat out of liver and muscle into safer stores; slow onset, fluid-retention trade-offs. |
| 7. Metformin | Liver | Mainly reduces liver glucose output; insulin-lowering effect in type 1 is modest, commonly cited around 5%.5 |
What the adjunct medicines really show
This is the part where headlines outrun evidence, so here is the measured version. GLP-1 receptor agonists (such as liraglutide and semaglutide) and the dual GLP-1/GIP agonist tirzepatide are not licensed for type 1; they are reviewed here, not endorsed, and every decision is one for the care team. In the randomised trials, the average effects are real but moderate, and they come with a ketone signal that has to be respected.6
| Trial | Drug + setting | Average effects | Grade |
|---|---|---|---|
| ADJUNCT ONE (2016) | Liraglutide + insulin, 1,398 adults | HbA1c about -0.20% at the top dose; weight -2.2 to -4.9 kg; dose-related ketosis and hypoglycaemia signals | A |
| ADJUNCT TWO (2016) | Liraglutide + capped insulin, 835 adults | HbA1c about -0.33% (insulin held, so the effect is the drug’s); weight -2.5 to -5.1 kg | A |
| ADJUST-T1D (2025) | Weekly semaglutide + automated insulin delivery, 72 adults | Time in range +8.8 points; weight -8.8 kg; insulin -22 units per day; composite goal met by 36% versus 0% on placebo; no ketoacidosis | A |
| Park 2024 meta-analysis | 24 trials, 3,377 people | Per mg of liraglutide: HbA1c -0.09%, weight -2.2 kg, insulin -4.3 units per day; consistent dose-response | A |
| Tirzepatide observational (Akturk and Shah 2025) | 26 adults, single centre, no control arm | Suggested larger weight and insulin effects, but uncontrolled and not comparable to the trials above | C |
When these medicines cut insulin need quickly, insulin must be reduced proactively or hypoglycaemia follows; and a real signal of euglycaemic ketosis (raised ketones with normal glucose) means ketone monitoring stays standard of care. SGLT-2 inhibitors carry the sharper version of this, the risk of diabetic ketoacidosis, which is why they are used only under strict protocols. None of this is a do-it-yourself exercise.
A trial that sounds exciting is not the same as a large effect. When you read “improves time in range” or “leads to weight loss”, ask the two questions that keep everyone honest: by how much on average, and at what cost in side effects? The honest GLP-1 story in type 1 is moderate average benefit with a ketone signal to manage, not a miracle.The Goldacre lens, evidence-grade discipline, one of the four GNL appraisal lenses.
The mechanism, the measure, and the limits
The gold standard you cannot use, and the proxy you can
The reference method for insulin resistance is the hyperinsulinaemic-euglycaemic clamp, which nobody runs in clinic. The workable proxy is total daily insulin divided by body weight in kilograms. Maahs and colleagues offer reference bands: below 0.4 units per kilogram is insulin-sensitive, 0.5 to 0.7 is mild resistance, 0.7 to 1.0 is moderate, and above 1.0 is high.7 These are population reference points, not personal verdicts; residual C-peptide shifts a person down the scale, because their own pancreas is still doing some of the work.
A worked illustration, not your data: the slow upward creep over years is the thing worth watching, and it is the thing the levers can bend. Bands are reference points, shifted by residual C-peptide; the slope tells the story.
Why the levers compound, and why order matters
Because insulin resistance in type 1 operates across muscle, liver, fat, gut, kidney and brain at once, the most effective approach addresses several sites together. Activity is upstream and touches the most sites; pharmacology is downstream amplification. A GLP-1 added to a sedentary week tends to disappoint; the same drug paired with movement, adequate protein and resistance training tends to compound, partly because muscle preserved during weight loss is itself a glucose sink. That is the case for lifestyle-first, not as a moral position but as a mechanistic one.
Reading the adjunct evidence the way a reviewer would
The pivotal liraglutide trials, ADJUNCT ONE and TWO, showed HbA1c effects below the regulatory threshold alongside dose-related ketosis and symptomatic hypoglycaemia, and the FDA did not approve liraglutide for type 1 on efficacy and safety grounds.6 The newer story is the automated-insulin-delivery era: ADJUST-T1D paired weekly semaglutide with an AID system in 72 adults with type 1 and obesity, and a composite goal (time in range above 70%, time below range under 4%, and at least 5% weight loss) was met by 36% versus 0% on placebo, with no ketoacidosis.8 Tirzepatide in type 1 rests on a single-centre observational cohort of 26 with no control arm; the numbers look larger but the design cannot support a head-to-head claim.9 In paediatric type 1 there are no Phase 3 trials at all, a gap ISPAD 2024 flags explicitly.6
Longer, larger AID-era trials with hard outcomes, paediatric data, and head-to-head tirzepatide RCTs would all strengthen it. Until then the honest summary is: moderate average benefit on insulin need, time in range and weight, a real ketone signal, and a paediatric evidence hole. Sell the evidence, never the hype.
It is the rare, large harm that matters most, not the average week. With these medicines that harm is ketoacidosis, low-probability but high-consequence. The robust move is to protect hardest against the catastrophic event you cannot afford: keep ketone monitoring in place, reduce insulin proactively, and never run an off-label adjunct without a structured protocol and your care team.The Taleb lens, robustness to outliers, one of the four GNL appraisal lenses.
A model is only as honest as its weakest input. The insulin-per-kilogram proxy is a clean, useful gauge, but it is a proxy for a clamp it cannot reproduce, and it is blind to residual C-peptide unless you account for it. The tirzepatide signal rides on an uncontrolled cohort. Name the limits out loud, and never sell the proxy or the cohort as the territory.The Hayes lens, technical and methodological rigour, one of the four GNL appraisal lenses.
The whole guide, summarised
Glucose never lies, and neither does the insulin trend: both keep an honest record over months. Read them kindly, address health not appearance, and choose the smallest set of levers that bends the slope.
This page is the taster. The full journey, three modules and their 30 questions, with your progress saved, lives in Learn with Grace. Glucose never lies; come and learn to read it.
References
Evidence grades A (strongest) to D (editorial or working analysis).
- Portal-peripheral insulin mismatch in T1D: subcutaneous insulin produces peripheral hyperinsulinaemia roughly four to eight times physiological. GNL working synthesis from the insulin-resistance concept base; mechanism per DeFronzo. B
- Sun et al. Insulin resistance and cardiovascular disease and all-cause mortality in type 1 diabetes. 2024 (full bibliographic detail under MA verification). B
- DeFronzo RA. From the triumvirate to the ominous octet: a new paradigm for the treatment of type 2 diabetes mellitus. Diabetes. 2009;58(4):773-795. (Eight-site framework applied to T1D by GNL.) B
- GNL insulin-resistance guide, seven-lever synthesis (movement, fat loss, nutrition, GLP-1, SGLT-2, pioglitazone, metformin). Working clinical synthesis. D
- Metformin in T1D: modest insulin-dose reduction, commonly cited around 5% in practice summaries; mechanism mainly hepatic. GNL working synthesis. D
- de Bock M, et al. ISPAD 2024 Clinical Practice Consensus Guidelines (adjunct therapy and paediatric gap); ADA Standards of Care 2025 §9; GLP-1/GIP adjunct evidence pack (ADJUNCT ONE Mathieu 2016; ADJUNCT TWO Ahrén 2016; FDA non-approval of liraglutide for T1D). A
- Maahs DM, et al. Insulin dose per kilogram as a practical proxy for insulin resistance in T1D (reference bands). B
- Shah V, et al. ADJUST-T1D: weekly semaglutide added to automated insulin delivery in adults with type 1 diabetes and obesity. NEJM Evidence. 2025. A
- Akturk HK, Shah VN, et al. Tirzepatide in type 1 diabetes: single-centre observational cohort (N=26), 2025. C
- Park J, et al. GLP-1 analogues as adjunctive therapy for type 1 diabetes: systematic review and meta-analysis (24 RCTs, N=3,377). 2024. A
One page, three voices: Jude, Grace, John. Population-average, not personalised. About health, never appearance.
Related guides: Diabetes, the foundations, Exercise and type 1, Nutrition and type 1, Mealtime insulin, HbA1c and Time in Range, Automated insulin delivery (AID), Prevention and cure.