Made on
Tilda
Why the Food Insulin Index reveals what glycemic index misses — and how to use it to stabilize blood sugar, support weight loss, and protect long-term metabolic health.
The Food Insulin Index (FII) is a ranking system that measures how much a specific food raises blood insulin levels in the two hours after it is eaten. It was developed in 1997 by Dr. Susanne Holt, Professor Jennie Brand-Miller, and Professor Peter Petros at the University of Sydney, published in the American Journal of Clinical Nutrition.
Unlike carbohydrate counting or glycemic index, the FII captures the complete insulin response to a food — including the effects of protein, fat, and fiber, not just carbohydrates. This makes it a richer picture of how your body actually responds to what you eat.
All foods are scored relative to white bread (100). A food with a score of 40 triggers roughly 40% as much insulin as an equal-calorie portion of white bread. Glucose has a score of approximately 100 in studies that use it as the reference.
Scores are relative to white bread (100). Threshold ranges are approximate and used for educational purposes.
Most people are familiar with the Glycemic Index (GI). It is a useful tool, but it only measures how carbohydrates raise blood glucose. The problem is that insulin secretion is not driven by carbohydrates alone.
As Holt et al. demonstrated in the original research, protein-rich foods and fat-rich bakery products trigger insulin responses that are disproportionately higher than their glycemic responses. Your pancreas may secrete significant insulin even when blood glucose barely moves — something the glycemic index cannot capture.
The testing method is precise and consistent. Healthy participants consume 1,000 kilojoules (239 calories) of a test food. Blood insulin levels are then sampled regularly for 120 minutes. Researchers calculate the incremental area under the insulin response curve (iAUC) and compare it to the iAUC for the reference food — white bread, scored at 100.
A 2023 review published in Clinical Nutrition ESPEN compiled data from 80 distinct articles and 32 countries, cataloguing over 629 food and beverage items. This is the most comprehensive dataset available to date. Insulin index values ranged from 1 (acacia fiber, gin) to 209 (soy milk-based infant formula).
The following table presents foods with favorable Food Insulin Index scores, drawn from the Holt et al. (1997) study and subsequent published research. All scores are expressed relative to white bread = 100.
| Food | Category | II Score | Visual | Key notes |
|---|---|---|---|---|
| Peanuts | Nuts / Legumes | 20 | Among the lowest scores tested | |
| All-Bran cereal | Cereal | 32 | High fiber content blunts response | |
| Whole eggs | Protein | 31 | Low despite some protein-driven insulin release | |
| Porridge (oatmeal) | Cereal | 40 | Beta-glucan fiber moderates insulin release | |
| Muesli (unsweetened) | Cereal | 40 | Lower than processed cereals | |
| Brown pasta | Carbohydrate | 40 | Al dente cooking lowers the response further | |
| White pasta | Carbohydrate | 40 | Protein-starch matrix slows digestion | |
| Cheddar cheese | Dairy | 45 | High fat content blunts insulin response | |
| Lean beef | Protein | 51 | Zero carbs but amino acids stimulate insulin | |
| Lentils in tomato sauce | Legumes | 58 | Soluble fiber and plant protein moderate response | |
| White fish | Protein | 59 | Lean protein; some insulinogenic amino acids | |
| Apples | Fruit | 59 | Fiber and fructose moderate the response | |
| Oranges | Fruit | 60 | Whole fruit only; not fruit juice | |
| Grain bread | Carbohydrate | 60 | Whole grains slow starch digestion |
Scores are approximate means from published studies. Individual responses vary. Data rounded for readability.
Understanding which foods produce a larger insulin response is equally important. The following table covers moderate- to high-II foods from the original Holt et al. dataset and subsequent research.
| Food | Category | II Score | Visual |
|---|---|---|---|
| Brown rice | Carbohydrate | 62 | |
| Grapes | Fruit | 74 | |
| French fries | Carbohydrate | 74 | |
| Cornflakes | Cereal | 75 | |
| White rice | Carbohydrate | 79 | |
| Croissant | Bakery | 79 | |
| Bananas | Fruit | 81 | |
| Low-fat strawberry yogurt | Dairy | 84 | |
| Whole-meal bread | Carbohydrate | 96 | |
| White bread (reference) | Carbohydrate | 100 | |
| Cookies (sweet) | Bakery | 92 | |
| Mars Bar | Confectionery | 112 | |
| Jellybeans | Confectionery | 117 | |
| Baked beans (tomato sauce) | Legumes | 120 | |
| Potatoes (boiled) | Carbohydrate | 121 |
The Food Insulin Index has produced several counterintuitive findings that challenge popular nutrition beliefs:
Lean beef has an insulin index score of approximately 51, despite containing essentially zero carbohydrates. This is because several amino acids — particularly leucine, lysine, and isoleucine — are potent direct stimulators of pancreatic beta cells. A zero-carb meal is not the same as a zero-insulin meal, and the FII makes this clear in a way the glycemic index cannot.
Both white and brown pasta score around 40 on the insulin index — roughly the same as oatmeal and much lower than white bread (100) or white rice (79). The protein-starch interaction in pasta creates a more slowly digestible food matrix that moderates insulin release. Cooking pasta al dente (slightly firm) lowers the response further.
Low-fat strawberry yogurt scores around 84 — significantly higher than plain whole-milk dairy. When fat is removed and sugar is added (a common formulation for flavored low-fat yogurts), the insulin demand increases substantially. Full-fat plain yogurt has a much more favorable profile.
Boiled potatoes score approximately 121 — higher than white bread itself (100). Potato starch, when fully gelatinized through boiling, is rapidly digested and generates a large insulin response. Cooling cooked potatoes increases their resistant starch content and lowers the insulin response somewhat.
"Diets that provoke less insulin secretion may be helpful in the prevention and management of diabetes. A physiologic basis for ranking foods according to insulin demand could therefore assist further research."
Chronically elevated insulin levels — a condition called hyperinsulinemia — is one of the earliest signs of metabolic dysfunction. Research links persistently high insulin not just to type 2 diabetes, but to a broad range of conditions.
A 2024 scoping review in Nutrients analyzed 25 peer-reviewed studies and concluded that a high dietary insulin index and insulin load are associated with the development of metabolic syndrome, insulin resistance, and type 2 diabetes. The review also found that FII can predict postprandial insulin response more accurately than carbohydrate counting alone.
Insulin is the primary anabolic hormone regulating fat storage. When insulin is consistently elevated, fat breakdown (lipolysis) is suppressed and fat storage is promoted. Diets that chronically spike insulin make it physiologically harder to mobilize stored body fat, even in a caloric deficit.
A 2024 cohort study in Scientific Reports found that individuals in the highest dietary insulin load quartile had significantly higher odds of metabolic syndrome compared to those in the lowest quartile. Research has also linked higher dietary insulin load to adverse lipid profiles, including elevated triglycerides and reduced HDL cholesterol.
For people with type 1 diabetes using insulin pumps or multiple daily injections, the FII offers a more precise basis for meal-time dosing than carbohydrate counting alone. The FII accounts for the insulin demand of protein and fat — components that carbohydrate-only counting systematically ignores.
Translating insulin index data into daily eating is easier than it sounds. The core principle is simple: pair protein and fat with fiber-rich carbohydrates, and choose whole, minimally processed foods.
The Food Insulin Index is a valuable tool, but it has real limitations that are important to understand:
All FII measurements capture the insulin response over 120 minutes. This window suits carbohydrate-driven responses, which peak quickly. However, the insulin response to high-fat meals extends well beyond two hours. Studies in type 1 diabetes patients show that dietary fat can cause a significant rise in blood glucose three to eight hours after consumption — a phenomenon the two-hour FII window does not capture.
Insulin responses vary considerably between individuals, based on gut microbiome composition, insulin sensitivity, body composition, meal history, sleep quality, and stress levels. FII scores are averages from small test groups (typically 10–11 healthy participants per food). They are useful population-level guides, not precise individual predictions.
Most FII data comes from testing single foods in isolation. When foods are combined in a meal, the interactions between macronutrients can change the insulin response in complex ways. A 2009 study by Bao et al. showed that FII-based predictions of composite meal responses were strong but not perfect, highlighting the complexity of real-world eating.
Despite the 2023 compilation cataloguing over 629 items, many everyday foods have not yet been formally tested. Data for many vegetables, specialty grains, and modern processed foods is absent or estimated.