--- title: "How Insulin Shapes Fat Burning: The Science Behind RealFoods" url: https://realfoods.app/science description: "How blood sugar and insulin, not only calories, shape whether your body burns stored fat. The research behind the Weight Impact Score, explained simply." last_modified: 2026-10-06T08:00:00Z --- 5 min read · Updated October 6, 2026 # The Science Behind RealFoods By [Dino Pohilj](https://realfoods.app/authors/dino-pohilj), Founder, RealFoods · Drafted with AI > Quick answer > > Blood sugar is the trigger and insulin is the actor. When a meal sends blood sugar up, insulin rises with it, and high insulin slows the release of stored fat from fat cells ([Jensen et al., Diabetes 1989](https://pubmed.ncbi.nlm.nih.gov/2573554/)). Calorie-matched portions of different foods can produce very different insulin responses ([Holt et al., AJCN 1997](https://pubmed.ncbi.nlm.nih.gov/9356547/)), and the carbohydrate-insulin model proposes that this changes how much fat the body keeps, though calorie-matched trials find only small differences. > > The Weight Impact Score estimates that insulin curve, not the calorie count. We argue that is a more useful second number for choosing what to eat, because it estimates what the body is likely to do with the meal. ## Why calories are only one number Calories measure energy. They establish nothing about what the body intends to do with that energy. A calorie is a physics unit: the heat required to raise one gram of water by one degree Celsius. Human biology is not a bomb calorimeter. Consider two 400-calorie meals. Meal A: a bowl of sugary cereal with skim milk. Meal B: salmon, avocado, and mixed greens. Both deliver roughly equal energy. Meal A is likely to send blood sugar up sharply within 30 minutes; Meal B raises it far less. We argue that the body's response, not only the calorie arithmetic, shapes whether it is more likely to keep fat or release it. Ludwig and colleagues, publishing in [*The American Journal of Clinical Nutrition* in 2021](https://pubmed.ncbi.nlm.nih.gov/34515299/), set out the carbohydrate-insulin model as a physiological framework for the obesity epidemic: dietary carbohydrate composition drives insulin secretion, which in turn pushes more of the energy from food into fat cells and raises hunger, and, they argue, the calorie-surplus model cannot explain the observed pattern of obesity without invoking hormonal mechanisms. The model remains debated, and RealFoods uses it as a framework, not as settled fact. ## What does insulin do after a meal? After a meal, insulin does three jobs at once. When a high-carbohydrate meal sends blood sugar up rapidly, the pancreas releases a surge of insulin, and insulin then performs these jobs simultaneously: 1. **Clears blood sugar** by routing it into muscle and liver cells, where it is stored as glycogen. 2. **Routes extra energy into fat cells** (lipogenesis) once glycogen stores are full, which can send excess sugar to fat tissue as triglycerides. 3. **Slows fat release** (lipolysis): elevated insulin inhibits hormone-sensitive lipase, the enzyme that releases stored fat for fuel, and fat-cell lipolysis is highly sensitive to insulin, as [Jensen and colleagues found (Diabetes, 1989)](https://pubmed.ncbi.nlm.nih.gov/2573554/). While insulin is high, the release of stored fat slows. In an exploratory analysis led by [Dr. Benjamin Bikman](https://realfoods.app/authors/benjamin-bikman), professor of cell biology and physiology at BYU, fat-tissue samples from 27 adults in a weight-maintenance feeding trial who had eaten meals with 60% of their calories from carbohydrate for 10 to 15 weeks showed lower mitochondrial respiration in the lab than samples from people eating 40% or 20% carbohydrate ([Bikman et al., 2022](https://pubmed.ncbi.nlm.nih.gov/35177807/)), an early signal rather than proof. RealFoods cites his work; Bikman is not affiliated with RealFoods and has not reviewed or endorsed the Weight Impact Score. A day of 1,600 calories from whole-wheat toast, fruit juice, flavoured yogurt and granola is likely to call in far more insulin than a day of 1,600 calories from eggs, salmon, vegetables and olive oil. We argue those two days will not play out the same way over time, though calorie-matched trials find only small differences in fat loss. The practical rule is short: [eat mostly real food](https://realfoods.app/real-food-weight-loss). In a [2025 randomized crossover trial](https://pubmed.ncbi.nlm.nih.gov/40760353/) of 55 adults with overweight or obesity in England (Dicken et al., Nature Medicine), people lost about twice as much weight over 8 weeks on minimally processed meals as on ultra-processed meals, with both sets of meals following UK healthy-eating guidelines. The key insight: Blood sugar is the trigger and insulin is the actor. When a meal sends blood sugar up, insulin is called in, and while insulin is high the release of stored fat slows. That is why we look at a meal's estimated insulin impact as well as its calories. Ready to stop guessing what to eat? ![Download on the App Store](https://realfoods.app/download_appstore.png) ## Does the same food affect everyone the same way? No. Zeevi and colleagues at the Weizmann Institute (Cell 2015) recruited 800 participants, fitted them with wearable blood sugar sensors (CGMs), and tracked 46,898 meals. Their finding: identical meals produced highly variable blood sugar responses between individuals. Standard nutritional guidelines, which assign a single glycaemic value to a food regardless of who is eating it, cannot capture this variation. We argue this research is important not as a counsel of despair but as a call for meal-level rather than food-level scoring, and for reading a score as an estimate: the same meal can land differently in different people, so a score reads best as a pattern across your week, not a single exact number. ## How does the Weight Impact Score work? Weight Impact is RealFoods' 1 to 10 score for how strongly a meal is likely to raise blood sugar and call in insulin, where lower is better: 1 to 2 is Burning Fat, 3 to 4 Mostly Neutral, 5 to 6 Fat-Burning Paused and 7 to 10 Keeping Fat. It is a modelled estimate from the meal's foods, not a medical measurement. In general, a few food traits shape how fast a meal sends blood sugar up: - **Refined starches and added sugars** send blood sugar up steeply. Intact whole grains and fibre-dense carbohydrates do so more slowly. White bread and lentils with the same carbohydrate grams behave very differently. - **Fibre-rich whole foods** slow how fast sugar is absorbed. The fibrous matrix of whole vegetables, legumes and intact grains is a main reason these foods tend to give a gentler rise despite containing carbohydrate. - **Portion size** matters too. 300g of white rice is likely to have a higher impact than 100g of the same food. The output: a 1 to 10 score per meal, shown with its zone. A day that averages in the Burning Fat or Mostly Neutral zones (below 5) is intended to reflect a pattern with a lower estimated insulin load, which we argue leaves more room for stored fat to be used. The feeding studies behind the meal-level scoring are summarised in our [score explainer](https://realfoods.app/how-weight-impact-score-works). How RealFoods turns research like this into a score is its own modelling, and none of the study authors has reviewed or endorsed it. ## Common questions ### How do blood sugar and insulin affect fat burning? When a meal sends blood sugar up, the pancreas releases insulin to bring it back down, and high insulin slows the release of stored fat from fat cells (Jensen et al., Diabetes 1989). Two meals with identical calorie counts can produce very different insulin curves (Holt et al., American Journal of Clinical Nutrition 1997), and the carbohydrate-insulin model proposes that this shifts how much fat the body keeps, though the model is still debated. ### What is the Weight Impact Score based on? Weight Impact is RealFoods' 1 to 10 score for how strongly a meal is likely to raise blood sugar and call in insulin, where lower is better: 1 to 2 is Burning Fat, 3 to 4 Mostly Neutral, 5 to 6 Fat-Burning Paused and 7 to 10 Keeping Fat. It is a modelled estimate from the meal's foods, not a medical measurement. It draws on the glycemic index, the insulin index of foods (Holt et al., 1997) and the carbohydrate-insulin model, which is an active and debated area of peer-reviewed research. ### Is this approach supported by peer-reviewed research? The approach draws on peer-reviewed research, but the carbohydrate-insulin model behind it is still debated. Sources cited on the RealFoods science and studies pages include Ludwig and Ebbeling (2018) and Ludwig (2023) on the model itself, and Zeevi et al. (2015) on how blood sugar responses to the same food vary between people. Controlled feeding trials such as Hall et al. (2021) are part of the debate, so RealFoods uses the model as a framework for its estimates, not as proven fact. ## Sources 1. Ludwig DS, Aronne LJ, Astrup A, et al. The carbohydrate-insulin model: a physiological perspective on the obesity pandemic. *Am J Clin Nutr.* 2021;114(6):1873-1885. [PMID 34515299](https://pubmed.ncbi.nlm.nih.gov/34515299/) 2. Jensen MD, Caruso M, Heiling V, Miles JM. Insulin regulation of lipolysis in nondiabetic and IDDM subjects. *Diabetes.* 1989;38(12):1595-1601. [PMID 2573554](https://pubmed.ncbi.nlm.nih.gov/2573554/) 3. Holt SH, Miller JC, Petocz P. An insulin index of foods: the insulin demand generated by 1000-kJ portions of common foods. *Am J Clin Nutr.* 1997;66(5):1264-1276. [PMID 9356547](https://pubmed.ncbi.nlm.nih.gov/9356547/) 4. Dicken SJ, Jassil FC, Brown A, et al. Ultraprocessed or minimally processed diets following healthy dietary guidelines on weight and cardiometabolic health: a randomized, crossover trial. *Nat Med.* 2025;31(10):3297-3308. [PMID 40760353](https://pubmed.ncbi.nlm.nih.gov/40760353/) 5. Bikman BT, Shimy KJ, Apovian CM, et al. A high-carbohydrate diet lowers the rate of adipose tissue mitochondrial respiration. *Eur J Clin Nutr.* 2022;76:1339-1342. The authors include Bikman, who discloses royalties from a book about insulin resistance. [https://pubmed.ncbi.nlm.nih.gov/35177807/](https://pubmed.ncbi.nlm.nih.gov/35177807/) Continue reading: [how the Weight Impact Score works in detail](https://realfoods.app/how-weight-impact-score-works), [how blood sugar drives weight gain](https://realfoods.app/blood-sugar-and-weight-gain), or [the peer-reviewed studies behind RealFoods](https://realfoods.app/studies). Continue learning ## Metabolic Science How blood sugar and insulin shape whether your body is more likely to keep fat or burn it. - [Does High Blood Sugar Cause Weight Gain?](https://realfoods.app/blood-sugar-and-weight-gain) How blood sugar, insulin and weight gain connect in the carbohydrate-insulin model, and what to try. - [What is Insulin Resistance?](https://realfoods.app/what-is-insulin-resistance) Signs, mechanism, and how what you eat and activity can improve insulin sensitivity and lower type 2 diabetes risk. - [What is the Glycemic Index?](https://realfoods.app/what-is-the-glycemic-index) A practical guide to GI, glycemic load, and why meal-level scoring beats single-food rankings. - [How to Lower Insulin Naturally](https://realfoods.app/lower-insulin-naturally) Five food and meal-timing habits aimed at a lower estimated insulin load, and how they differ from GLP-1 drugs like Ozempic. - [PCOS and Insulin Resistance: What the Research Says About Food](https://realfoods.app/pcos-insulin-resistance) What the research says about PCOS, insulin resistance and weight, and how a lower estimated insulin impact per meal fits in: protein and fibre, fewer sugary drinks, walks after eating. - [Prediabetes: What to Eat for Steadier Blood Sugar](https://realfoods.app/prediabetes-diet) Prediabetes explained: blood sugar ranges, food swaps, sugar limits and what the Finnish and U.S. prevention trials found. - [What to Eat on Ozempic and GLP-1](https://realfoods.app/what-to-eat-on-glp-1) 45 foods scored 1 to 10 by insulin impact for people on semaglutide or tirzepatide, plus why food quality still matters when appetite is suppressed. - [Signs Your Body May Be Burning Fat](https://realfoods.app/how-to-know-if-your-body-is-burning-fat) Stored fat is released more readily when insulin is low. What you can feel, what devices read, and how to estimate it from what you ate, without a CGM. - [Eating for Insulin Resistance: What to Eat and What to Limit](https://realfoods.app/insulin-resistance-diet) No single official eating plan exists for insulin resistance. Here is the eating pattern research supports, scored food by food on the 1 to 10 Weight Impact Score. - [Low-Insulin Breakfast Foods: A Scored List](https://realfoods.app/low-insulin-breakfast-foods) Breakfast foods scored by estimated insulin impact: eggs and salmon at 1, dairy anchors and strawberries at 2, plus why food order matters as much as selection and what trials of eating breakfast versus not eating it show on both sides. - [What Actually Breaks a Fast (and What Doesn't)](https://realfoods.app/what-actually-breaks-a-fast) Which drinks keep a fast and which end it, ruled item by item under two definitions: calories end a strict fast, while an insulin rise slows the release of stored fat. Sweeteners and gum land in the gray zone. - [Why Your Weight Stalled on GLP-1 Medication](https://realfoods.app/why-your-weight-stalled-on-glp-1) A stall during GLP-1 treatment usually reflects water shifts, changing body composition, narrowing appetite effects, and metabolic adaptation, not failure. > ![Dino Pohilj](https://realfoods.app/dino-pohilj-112.webp) > > Written by > > [Dino Pohilj](https://realfoods.app/authors/dino-pohilj) > > Founder, RealFoods > > Updated and reviewed October 6, 2026