The food insulin index is a powerful tool that can help you make better food and meal choices. It measures how much a food causes your blood insulin levels to rise.
Foods with a high insulin index cause a rapid spike in blood insulin, while foods with a low insulin index cause a gradual rise.
Why is this important?
Insulin is a hormone that helps your body absorb glucose from the bloodstream. When your blood insulin levels are too high, it can lead to a number of health problems, including weight gain, type 2 diabetes, and heart disease.
The food insulin index can help you in a number of ways:
- Stabilize your blood sugar levels: Eating foods with a low insulin index helps to keep your blood sugar levels stable, which can improve your energy levels, mood, and overall health.
- Lose weight: Eating a diet rich in low-insulin-index foods can help you lose weight or maintain a healthy weight.
- Manage diabetes: For people with diabetes, the food insulin index can help them improve the accuracy of their insulin dosing calculations.
While the food insulin index is a valuable tool, it’s important to consider other factors when making food choices, such as satiety and nutrient density.
This article will show you how to use the food insulin index to tailor your food and meal choices to your individual goals and get the desired results.
- What Is the Food Insulin Index?
- Who Developed the Food Insulin Index?
- How Do You Calculate the Insulin Index of Food?
- Is the Food Insulin Index Important?
- Limitations of the Food Insulin Index
- What Makes a Food High on the Insulin Index?
- Low vs High-Insulin-Index Foods
- Low vs High Insulin Index Recipes
- What Is a Good Insulin Index Score?
- How Can I Use the Food Insulin Index to Dose Insulin Accurately?
What Is the Food Insulin Index?
The food insulin index is a measure of how much a particular food or meal raises insulin in the two hours after eating.
- Foods with a high-insulin index are typically high in refined carbohydrates like white bread, sugary drinks, and processed snacks that are absorbed quickly by the body.
- Conversely, foods with a low-insulin index tend to be higher in fat and don’t increase insulin significantly over the short term.
- Meanwhile, the food insulin index data shows that high-protein foods raise insulin about half as much as carbohydrates over the first two hours.
Who Developed the Food Insulin Index?
They researched the relationship between diet and insulin resistance, a significant risk factor for developing Type-2 Diabetes.
Their 1997 paper, An insulin index of foods: the insulin demand generated by 1000-kJ portions of common foods, detailed the concept and insulin responses to various common foods.
The chart below shows the insulin index values of the 38 foods tested. Here, researchers fed participants 1000 kJ (239 cal) portions of popular foods and their blood insulin responses were measured.
Foods with the highest insulin index scores from the original study included jellybeans, Mars Bars, potatoes and beans. Meanwhile, peanuts and eggs had the lowest insulin index scores.
For a printable version of the food insulin index chart, you can download the original paper here.
How Do You Calculate the Insulin Index of Food?
To determine a food’s insulin index score, researchers measured the change in insulin levels two hours after eating. The area under the insulin response curve was then calculated for each food. Pure glucose was given a score of 100, so any food with a food index score of less than 100 elicited a smaller insulin response than glucose (and vice versa).
Notice in the figure below how insulin levels peak quickly after eating pure glucose (black line) and then return to baseline at two hours. In contrast, insulin levels rise slowly for higher-fat foods like milk (aqua line) but remain elevated at two hours.
Is the Food Insulin Index Important?
I initially became interested in the food insulin index to help my wife Monica control her Type-1 Diabetes. The effect of accurately dosing for protein and avoiding food that required the most insulin was dramatic. She stabilised her blood glucose, reduced many adverse symptoms related to her diabetes, increased her energy levels, lost weight, and halved her daily insulin requirements!
The chart below shows her daily insulin dosing since she started using an insulin pump in 2008. Notice how her daily insulin dose dropped from 50 to 25 units per day in early 2011 when we transitioned to a lower-carb diet. That’s a 50% drop!
She reduced her bolus insulin requirements—or the insulin she requires at mealtimes—because her food choices improved. However, her basal insulin also dropped due to the weight she lost, as you can see from the photos of us below.
Many people have used the food insulin index to optimise their blood glucose management and insulin requirements. Other people following a keto diet have used it to identify more ketogenic foods.
However, I’ve since learned that the food insulin index can be counterproductive when taken to extremes (e.g. limiting protein because it has a short-term insulin response). As you will see, the food insulin index is not the only thing we need to consider.
Limitations of the Food Insulin Index
Before we get into the detailed food lists and charts, we should highlight several limitations of the food insulin index.
When trying to make sense of the food insulin index data, it’s critical to remember that the food insulin index tests were only undertaken over the first two hours after a meal.
Thus, the two-hour test captures the insulin response to refined carbohydrates. However, it does not measure the complete response to protein and fat which are metabolised over a longer timeframe.
Unfortunately, we don’t know much about the longer-term insulin response to our food because no studies (to date) have looked at it. The two-hour tests from the food insulin index are the best we have.
However, a recent study on people with Type-1 Diabetes showed that dietary fat initially lowered blood glucose in the short term when insulin wasn’t administered. Still, greater fat intake caused a significant increase in blood glucose three to eight hours after consumption.
So, dietary fat appears to elicit a smaller yet still significant insulin response over a longer timeframe.
It Only Shows the Change in Insulin Levels
When trying to make sense of the food insulin index, it’s important to note that the tests measured the change in blood insulin levels after food (i.e., bolus insulin).
An underappreciated fact is that your pancreas still produces insulin to keep your energy reserves locked in storage even when you’re not eating. This is known as basal insulin.
If your body didn’t produce basal insulin, your body would effectively disintegrate, and all the fat, protein, and glucose in your body would flow into your bloodstream, similar to a person with uncontrolled Type-1 Diabetes. So while insulin is often viewed as an anabolic hormone, most of the time, insulin works as an anti-catabolic hormone to prevent your body from plunging into uncontrolled catabolism.
For most people, most of their daily insulin is basal insulin. We know from people with Type 1 Diabetes that basal insulin makes up about half the daily insulin requirements for people on a typical western diet. However, for people on a lower-carb diet, 70 to 80% of their daily insulin requirement is basal insulin and not related to food.
Simply eating high-fat foods with a lower insulin index will not turn off your pancreas. Reducing basal insulin requires weight loss, which requires higher-satiety foods and meals. For more on this, see:
- Insulin is NOT Making You Fat (and Here’s Why)
- What Does Insulin Do in Your Body?
- High-Satiety Index Foods: Which Ones Will Keep You Full with Fewer Calories?
The 1997 Food Insulin Index study only tested 38 foods, so applying the data to other foods and meals is difficult.
Fortunately, the food insulin index database has been expanded since the original study and now contains over one-hundred foods. Dr Kirstine Bell’s 2014 thesis, Clinical Application of the Food Insulin Index to Diabetes Mellitus, expanded the Food Insulin Index database, which now contains 147 foods. You can view the complete updated Food Insulin Index database in Appendix 3 of the thesis here.
The chart below plots the data in terms of carbohydrates (%) vs food insulin index.
You can dive into all the data in the interactive Tableau version of this chart here. It’s best viewed on your computer screen, not your phone. You can mouse over each of the points to see more data.
Overall, we can see that while carbohydrates correlate with the food insulin index, there is a lot of scatter. Carbohydrates are clearly not the only factor at play.
- Towards the right, we can see that fruits (like peaches, raisins, and oranges) tend to sit below the trend line. This is because fructose (the form of sugar found in fruit) tends to be converted to fat in the liver and thus doesn’t raise insulin or glucose as much.
- Towards the left, we can see that the foods with the lowest insulin index are high-fat foods (like butter and olive oil).
- However, we can also see that low-carb, high-protein foods (like white fish, tuna and steak) have a significant insulin response).
What Makes a Food High on the Insulin Index?
While 147 foods is an improvement from the original 38 foods in the original study, it’s still hard to make much sense of such a limited dataset and apply it to other foods and meals.
So I ran a multivariate regression analysis on the 147 data points, which showed:
- Carbohydrates raise insulin the most;
- Protein raises insulin about half as much as carbohydrates (46%) over the first two hours;
- The fibre in food reduces the insulin impact of carbohydrates (by 78%); and
- Fat raises insulin by 15% as much as carbohydrates over the first two hours after eating.
Low vs High-Insulin-Index Foods
The regression coefficients from the multivariate analysis of the food insulin index data have allowed us to predict the two-hour insulin response to any food or meal.
Net Carbs vs Food Insulin Index vs Satiety Index
To demonstrate, the chart below shows net carbohydrates vs food insulin index. The colouring is based on our satiety index score. Foods in green provide more satiety per calorie, whereas foods in red tend to be much easier to overeat.
Again, you can dive into the detail in the interactive Tabluea version of this chart here. Make sure you hover your mouse over each food to see all the details.
In the top right of the chart, we can see how high-carb foods like rice, dates, and watermelon elicit the most significant food insulin index responses. However, things get more interesting when we look in the bottom left corner.
- Some foods like butter, olive oil and pork fat have the least amount of carbs and the lowest food insulin index. However, they also have meagre satiety index scores.
- Meanwhile, some low-carb, high-protein foods like turkey breast, tuna and cod that elicit a moderate insulin response have a very high satiety score.
So, what does this mean in practice?
While reducing carbohydrates is a great way to reduce your insulin response to food and stabilise your blood glucose, it’s crucial not to avoid high-satiety, high-protein foods just because they elicit an insulin response, particularly if weight loss is a goal.
Food Insulin Index vs Nutrient Density vs Satiety Index
To illustrate further, the chart below compares the food insulin index vs nutrient density (per serving). Again, the colouring is based on our Satiety Index Score. Notice how most nutritious foods have a moderate insulin response (e.g., 20 to 35% food insulin index).
Again, you can dive into the detail in the interactive Tableau version here.
In addition to these charts, we have prepared a range of simple printable food lists tailored to a range of goals using the food insulin index as one of the ranking factors, along with satiety and nutrient density.
Low vs High Insulin Index Recipes
While it’s easy to look at the insulin index data of one food, we usually don’t eat things solo. Instead, we combine them into meals and recipes.
To help you identify recipes with a lower food insulin index, the chart below shows our 1000+ NutriBooster recipes regarding food insulin index vs Diet Quality Score. Again, the colouring is based on our Satiety Index Score.
There’s a lot of data, so it will be best to dive into the detail in the interactive Tableau version here.
The critical thing to note is that the recipes with the greatest nutrient density scores and satiety responses tend to have lower—but not very low—food insulin index scores. In contrast, the recipes in the bottom left have considerably low food insulin index and nutrient density scores.
You can learn more about our suite of NutriBooster recipe books here. If you’re already part of our Optimising Nutrition community, you can download samples of each NutriBooster recipe book here.
What Is a Good Insulin Index Score?
If you’re insulin resistant and your blood glucose rises by more than 30 mg/dL (1.6 mmol/L) after eating, choosing foods with a lower food insulin index score can be beneficial.
Large rises in glucose tend to lead to crashes a few hours later that lead to increased hunger and overeating. For more on this, see our article on reactive hypoglycemia here.
However, you don’t need to push below 40% as it will force you to exclude many nutrient-dense, high-satiety foods.
If you require therapeutic ketosis and elevated ketones to manage a condition like epilepsy, dementia, Parkinson’s, or major depression, you may benefit from choosing foods with an even lower insulin index. However, you should be mindful that these foods and meals aren’t as satiating, so they are not ideal if fat loss is a goal.
How Can I Use the Food Insulin Index to Dose Insulin Accurately?
Despite its limitations, the food insulin index data is particularly helpful for people with diabetes who need to inject insulin. Accurately dosing insulin for carbohydrates and protein allows them to achieve better blood glucose control.
Unfortunately, the insulin requirement for protein is often overlooked because the amount people eating a higher-carb diet require is smaller than the insulin needed for carbohydrates.
However, following a lower-carbohydrate diet will likely push you to consume more protein. Hence, accurately calculating the insulin required for protein becomes more critical as it will make up most of your bolus insulin requirement.
Per usual practice, the insulin required for carbohydrates is calculated by multiplying net carbohydrates by the insulin:carb ratio (ICR). This is covered with a bolus dose of rapid-acting insulin.
As noted above, protein requires about half as much insulin as carbohydrates per gram over the first two hours. To calculate your insulin for protein, take the grams of protein in the meal, multiply it by the ICR and divide by two.
Rather than administering this as a bolus of rapid-acting insulin, many people on multiple daily insulin doses (i.e., insulin pens) use R-insulin, which works over a more extended period than rapid-acting insulin. This tends to match the protein digestion profile, which requires insulin over around five hours.
Alternatively, people using an insulin pump often use the ‘extended bolus’ (e.g. on Medtronic Pumps) or a ‘slowly absorbed meal’ (e.g. on CamAPS).
Because the response is more gradual, most people don’t dose insulin for dietary fat. Instead, the insulin is often covered by their basal insulin.
If you use a closed-loop artificial pancreas system, the algorithm will respond by increasing your basal insulin requirements to cover the dietary fat you consume. However, you may need to reduce your basal insulin if you are fasting.
For more tips on managing insulin-dependent diabetes, see How to Optimise Type-1 Diabetes Management (Without Losing Your Mind).
- The food insulin index measures the average insulin responses to a list of foods.
- Non-fibre carbohydrates spike insulin the most. Meanwhile, protein increases insulin to rise about half as much as carbs in the first two hours following a meal. Fat has a smaller response over a longer timeframe.
- Understanding how our bodies respond to food allows us to dose insulin and select foods that stabilise blood glucose more accurately.
- A diet with a very low food insulin index may be appropriate if you require therapeutic ketosis. However, these foods tend to sacrifice nutrient density, which is critical for satiety. Thus, a moderate insulin index diet will be more appropriate for most people as it provides for greater satiety and nutrient density.