Q: Will low-energy-density foods increase your satiety, make you feel full, and help you eat less?
A: Yes, it will.
Professor Barbara Rolls has led numerous studies looking at the effect of low-energy-density foods validating energy density as a beneficial parameter to increase satiety and help people eat less. She has also published several books on the topic, including the Ultimate Volumetrics Diet.
Q: But will those foods keep you full over the long term?
A: Maybe not so much!
As you’ll see, our analysis of more than one hundred thousand days of data from people using Nutrient Optimiser shows the effects of eating low-energy-density foods appear to be limited.
There are more valuable parameters we can use to help us optimise our foods and meals, like protein %, fibre, minerals, and vitamins. We appear to have an appetite for these nutrients, meaning our cravings will subside when we give our bodies what they need.
Additionally, prioritising nutrient-dense foods—or foods that contain more of the nutrients you require per calorie—will innately guide you to lower-energy-density foods. However, the converse is not always true. Simply prioritising lower energy-density foods doesn’t always help; in fact, it can decrease satiety!
In our previous article, Low Energy Density Foods and Recipes: Will They Help You Feel Full with Fewer Calories? I discussed energy density in detail. However, energy density often comes up when talking about satiety, and I’ve recently been discussing this with some people I respect.
So, I thought it would be helpful to dig back into the data in this article to:
- review the value of thinking in terms of energy density;
- show why energy density isn’t the most helpful strategy for Nutritional Optimisation in a comprehensive system; and
- identify other more valuable parameters.
In the first part of this article, we’ll dig into the data analysis. But this topic comes to life when you see what it looks like in foods and meals. So, if you want to see how real foods and meals compare in terms of energy density vs our satiety index, you can jump to the charts at the end.
- Satiation vs Satiety
- How Is the Energy Density of Food Measured?
- Benefits of Low-Energy Density Foods
- What Happens at The Energy Density Extremes?
- Does Adding Water Helps Us Eat Less?
- Protein % and Protein Leverage
- Good System Design
- Energy Density vs Protein %
- What Happens When We Add Fibre?
- There is No One-Size-Fits-All Solution
- Satiety Index Score vs Energy Density
Satiation vs Satiety
Before we even get started, let’s define satiation and satiety!
- Satiation is the short-term feeling of fullness and no longer wanting to eat immediately after a meal. So, while some bulky, high-volume, low-energy-density foods might make you feel ‘stuffed’ in the short term, there’s a good chance you’ll be seeking more food in a few hours.
- In contrast, satiety refers to the long-term feelings of fullness we experience when we obtain enough of all the essential nutrients we require. People who eat more satiating foods feel fuller for longer and can trust their appetite to guide them to what they should eat and when they need more nutrients and fuel.
Many experiments have measured the effect of specific food parameters on our propensity to eat more or less at the next meal. For example, in Rolls’ 1999 paper, Water incorporated into a food but not served with a food decreases energy intake in lean women, 24 women were given chicken and rice soup at an initial meal or chicken, rice, and a glass of water independent of one another. They then tested their hunger and the amount they ate at the next meal.
Rather than simply considering short-term fullness, our data from Optimisers represents multiple daily energy intakes of free-living people. This allows us to understand long-term satiety rather than short-term fullness. Because we have a substantial amount of data from a large population, we also can gain insight into which dietary patterns and nutrients align with eating less.
If you want to find a way to eat less, lose weight, and improve your metabolic health in the long run, you need to achieve and maintain long-term satiety. So it’s crucial to prioritise longer-term satiety over short-term fullness or satiation.
For more detail on satiety and some of its driving factors, check out The Cheat Codes for Nutrition for Optimal Satiety and Health.
How Is the Energy Density of Food Measured?
Now that you understand satiety and satiation, we can move on to define energy density.
Unfortunately, accurately measuring the energy density of food is challenging.
Outside the nutrition world, energy density is usually measured in terms of energy per unit volume. However, we never measure the volume of our food! To do so, we’d have to dunk it in water and measure the water displaced.
Instead, we usually use calories per unit of weight to measure the energy density of food. For reference:
- Foods and meals that consist of pure fat provide nine calories per gram and have an energy density of 222 grams per 2000 calories.
- Meanwhile, protein and carbs each provide four calories per gram and have an energy density of 500 grams per 2000 calories.
- Finally, alcohol contains seven calories per gram, giving it an energy density of 286 grams per 2000 calories.
We can calculate the weight, or mass, of food by summing:
- water, and
Measuring the weight of the various macros is easy, but measuring the water weight of our food is more challenging. Although it’s common to measure the water content of fruits and vegetables, the energy density of other food groups is measured less frequently.
Subsequently, the information we have on energy density isn’t consistent for many foods. As you’ll see, this makes using energy density as a parameter challenging with available food databases.
Benefits of Low-Energy Density Foods
To date, we have 108,895 days of food logs from 2,972 people that contain information on all parameters necessary to calculate energy density. The distribution chart below shows the range of energy densities amongst our Optimiser data.
The median energy density of food consumed by our Optimisers is 386 grams per 2000 calories. The 15th percentile is 340 grams per 2000 calories, and the 85th percentile is 1279 grams per 2000 calories.
Towards the far right of the chart above, we can see some people logged water, soups, drinks, shakes, coffees, and other liquid-based foods, which lowered the energy density of their diet. We’ll touch on the relevance of this shortly, but we want to dive into the good news about energy density first.
The satiety response chart below shows that people consuming foods with a lower energy density—at least in the normal range—consume 39.7% fewer calories than those who consume very energy-dense, high-fat meals.
The average Optimiser gets 46% of their calories from fat. So, the effects of lower energy density towards the right of the chart above may align with people consuming less fat and more of their energy from protein.
The relationship between energy density and calories is the inverse of the chart below shows that moving from high to low fat aligns with a 37% reduction in calories. However, while lower energy-density foods contain less fat, energy density does not differentiate between carbohydrates and protein, which each provide four calories per gram.
From our satiety analysis, we know that protein has a more substantial influence than fat or carbs on how full you feel over the long term. We also know that the average Optimiser only gets around 16% of their energy from non-fibre carbs, so most of the shift in energy density they experience comes from reducing fat.
For more details on the satiety response to fat vs protein, see High Protein vs High Fat: What’s Ideal for YOU?
The chart below shows the average satiety responses to all of the macronutrients together on one chart. Importantly, increasing protein % has the most significant impact on satiety.
As you will see further down, protein % is a much better predictor of satiety than energy density.
What Happens at The Energy Density Extremes?
Things tend to go south when we swing to extremes.
- At one end, we have an energy-dense, pure-fat diet.
- On the other, we have a low-energy-density, watered-down diet.
Fat is a great source of energy that comes with protein. However, pure fat—like olive oil, coconut oil, butter, lard, tallow, and avocado oil—contains little in the way of nutrients per calorie and is energy-density, so it provides lower satiety.
But what happens when we move to the other extreme and add water to dilute our energy density? Looking at the bigger picture, we see that we tend to eat more when foods’ energy densities exceed 550 grams per 2000 calories.
Thus, the data shows that very low-fat meals or watery foods and drinks are not very satiating. In other words, the volume may fill you up in the short term, but your appetite will seek out the protein and other nutrients you require over time.
Unfortunately, it’s impossible to know if the people consuming the most volume were eating watery soups, soda sweetened with high-fructose corn syrup, or Starbucks Frappuccinos with extra simple syrup, whipped cream, and sprinkles.
Energy density alone doesn’t help us differentiate the quality of the food, much less the composition. However, unbound water found in drinks, coffee, smoothies, and soups is less likely to elicit the same beneficial effects as it might when incorporated into the food matrix of minimally processed fruits and vegetables.
Does Adding Water Helps Us Eat Less?
As a general rule, drinking your calories in shakes or smoothies where the food matrix requires less effort to digest and assimilate is not ideal. Rather than eating pulverised and essentially pre-digested food, you experience greater satiety when you chew your food, and your stomach has to do some work to break it down.
It’s also worth noting that free-living people tend to gravitate towards more energy-dense foods and meals that combine fat and carbs when they’re available. Few people in the real world can sustain themselves on watery soups designed for their diluted energy values. Because their nutrient content is also diluted, they provide little in the way of long-term satiety.
We’ve tried to paint energy density in the best light in the following analysis. To do so, we culled all data with an energy density greater than 700 grams per 2000 calories to show the range in which a lower energy density appears beneficial. This left us with 101,464 days of data, which we will analyse below.
Protein % and Protein Leverage
If you’ve been following the Optimising Nutrition blog, you’ll know that our analysis has identified protein % as the dominant factor influencing satiety.
Rather than eating more protein in terms of absolute quantities, we experience greater satiety when we dial back non-protein energy from carbs and fat while prioritising protein, thus increasing our per cent of total calories from protein or our protein %.
We created the satiety response chart below using our Optimiser data analysis. Here, we can see that people consuming the highest protein % eat 61% fewer calories.
For more details, check out this article on the Protein Leverage Hypothesis.
Counterintuitively, this does not necessarily require a very high protein intake in absolute terms (i.e., grams); instead, it usually calls for someone to dial back their fat and carb intake while marginally increasing their protein consumption.
In this simple head-to-head comparison, we see that the positive influence of protein % far exceeds what we get when we shift our focus to energy density.
- Focussing on protein % results in a 61% calorie reduction, whereas
- Lowering energy density has us eating 40% fewer calories.
Nonetheless, this doesn’t necessarily mean these two parameters wouldn’t work in a system of Nutritional Optimisation. Perhaps we could combine them to get make even better food choices than using one or the other alone?
Good System Design
Nutrition and our appetite are fascinating, complex, and mysterious. They are anything but simple and come with many moving parts that are constantly interacting!
I’ve spent a lot of time looking at how all the various nutrients correlate to improved satiety. Several have risen to the top and appear to have the greatest predictive influence, including:
But as I’ve continued to analyse the data, I’ve realised that these dominant parameters are also proxies for other underlying parameters. In other words, nutrients tend to come clustered with one another, meaning lower-priority vitamins, minerals, essential fatty acids, and amino acids likely increase the satiety factor of those nutrients we listed above.
Additionally, foods and meals containing more nutrients per calorie also tend to have a lower energy density.
A range of factors aligns with satiety. However, it’s critical to understand which ones are the most important and how they interact to know what to prioritise and what is more an effect of focusing on the things that matter (i.e., energy density when consuming high-protein, nutrient-dense foods).
For an example of how a poorly designed system can lead you to be more confidently wrong, see Is Tufts University’s Food Compass Nutrient Profiling System ‘Broken’?
Google is a complex multi-criteria artificial intelligence algorithm that utilises hundreds of factors to provide the best search results. However, the weighting of each factor is calibrated with a massive amount of data. It’s also tailored to the searcher’s context and goals.
When optimising our nutrition, we can’t simply throw all our favourite parameters into a black box and hope for the best! Fortunately, as you will see below, we can get close to identifying the critical parameters of a large dataset and how to weight them.
Energy Density vs Protein %
So far, we see that energy density and protein % are both useful when we consider them separately. However, we are still left with two questions:
- Which parameter is most helpful in predicting satiety?
- Can they be combined together?
The table below shows results from a multivariate analysis where protein and energy density were considered together. Here, we see that:
- Moving from low to high protein % (i.e., 19 to 44%) aligns with a 29.5% decrease in overall calorie consumption, whereas
- Moving from a low to high-energy-density diet only provides a 2.8% reduction in calories when the effect of protein % is also considered.
|Energy Density (g/2000 cal)||5.3E-26||335||429||-39||-2.8%|
Overall, the takeaways from this multivariant analysis are that:
- protein % has a stronger relationship with energy intake, and
- energy density adds little value when we’ve already considered protein %.
What Happens When We Add Fibre?
Our previous analysis also showed that fibre helps increase satiety. As the chart below shows, people consuming the most fibre tend to eat 39% fewer calories than those who consume the least. For more on fibre, see Dietary Fibre: How Much Do You Need?
But it’s likely not just fibre that has this positive effect. Foods high in naturally-occurring fibre tend to be rich in other essential nutrients like potassium, manganese, iron, folate, and vitamins C and K1. For more details, see Nutrient Leverage Theory – Cluster Analysis.
Moreover, natural fibre is found in minimally processed carbohydrates free from added sugars and processed grains. So, prioritising fibre generally eliminates ultra-processed foods.
The following table shows what happens when we add fibre in the multivariate analysis.
|energy density (g/2000 cal)||4.44969E-10||338||429||25||1.8%|
So, what does this mean for energy density?
As we can see in the percentage column (%), it is positive 1.8%. Low-energy-density foods align with an increased calorie intake after fibre and protein are considered.
So, if you’re already prioritising protein and fibre, prioritising lower-energy-density foods will cause you to eat more!
It’s also important to consider that fibre is one of the most commonly measured constituents in food, and it’s easier to conceptualise than energy density. Thus, we don’t need to worry about energy density with fibre and protein % in the system. Adding energy density degrades its effectiveness.
Aside from protein and fibre, many other nutrients align with greater satiety. For more details, see The Cheat Codes for Nutrition for Optimal Satiety and Health.
There is No One-Size-Fits-All Solution
Unfortunately, there is no one nutritional solution for everyone. However, our analysis shows that protein % always dominates. If someone’s goal is fat loss, most people will benefit from increasing the protein % of their diet.
While protein leverage is simple in theory, it’s often hard to implement; when we have access to energy-dense foods that combine fat and carbs, we tend to overeat them. But swinging too far to the high protein % extreme tends to end badly, often resulting in intense cravings and rebound binging as the body tries to compensate for the easy energy it’s used to getting.
It takes time, practice, and often a period of food tracking to form new habits and increase your food’s satiety value. This is the exact process we guide people through in our four-week Macros Masterclass.
Most people eat TO satiety. But by taking the bull by the horns and changing WHAT you eat, you can eat FOR satiety. That is, you can change what you eat so your body is satisfied with more or less energy. When you do this, you will be able to break free of weight gain, poor metabolic health, and falling victim to your appetite!
While we don’t have randomised controlled trials (yet!), this principle works in practice when people apply it. Rather than having someone deprive themselves or try to dilute their food with water, prioritising what your body needs changes the game! The comment below from a Macros Masterclass participant is pretty standard feedback.
Once you’ve got your foundations of adequate protein in place, dialling in your priority minerals and vitamins is the next step on your journey towards Nutritional Optimisation. Your unique nutritional solution depends on your current baseline diet and the nutrients it’s lacking most.
In the Micros Masterclass, Optimisers track their current intakes in Cronometer, which is synced with our Nutrient Optimiser software. This allows them to identify foods and meals that provide more of their priority nutrients they aren’t getting enough of so they can satisfy their cravings.
If you’re interested in seeing what Nutrient Optimiser is all about, you can use our Free 7-Day Nutrient Clarity Challenge to get a glimpse of your priority nutrients and the foods and meals that contain them.
Satiety Index Score vs Energy Density
Satiety and energy density are fairly abstract concepts until you see what it looks like in terms of foods and meals.
To make these ideas applicable, we’ve used our detailed understanding of how various parameters interact to formulate a satiety index score that estimates how many calories an average person will eat of a certain food or meal if that’s all they eat in a day. Additionally, we used it to calculate a satiety index score, which allows us to rank foods and meals based on how satiating they are.
In the following sections, we’ll discuss the outcomes of our satiety analysis of various recipes and foods.
The chart below brings the abstract theory to life and shows our six hundred NutriBooster recipes plotted in terms of energy density vs satiety index score.
The colouring is based on nutrient density; those in green contain more nutrients per calorie and have a greater satiety index, whereas those in dark red are the most energy dense.
You can dive into the detail in the interactive Tableau version of this chart here.
There is some alignment between energy density and our satiety index score, which we have highlighted below.
- Foods like salads, soups, and smoothies found towards the right have the lowest energy densities.
- However, foods and recipes towards the bottom right of the chart have a low energy density. However, they have lower nutrient densities and, thus, lower satiety scores. We might be full for a few hours if we only ate these meals. However, we would likely find ourselves super hungry later on as our appetites upregulated to get us the nutrients we didn’t get from our soup or smoothie.
- Interestingly, there is a large bunch of very nutrient-dense meals with a moderate energy density in the middle of the chart towards the top. We would miss these robust protein-focussed recipes if we only focussed on low-energy-density meals shown toward the right.
Finally, the chart below shows a range of popular foods plotted on a graph with their energy density vs satiety score. Unfortunately, the USDA database only contains water data—the amount of water that adds weight to food—for 45% of its entries, so the list of foods is a little shorter than usual.
If you’re interested, you can dive into all the details in our Tableau chart here. Alternatively, you can download one or all of our optimised food lists, which are tailored for various goals and preferences, here.
Below, we’ve outlined a few trends we saw in the recipes.
- The foods towards the far right are non-starchy green veggies. These foods are super nutrient-dense and help fill in the gaps missing in your micronutrient profile. However, most people don’t eat a lot of them or can’t because they are so filling.
- Towards the bottom, we see watery foods like watermelon, pears, and apples. These foods have low energy densities but don’t contain much protein and lack a broad spectrum of nutrients. Although you may feel full after eating these foods in the short term, they don’t provide the protein most people require for long-term satiety.
- Again, we see a range of protein-rich foods in the middle and towards the top of the chart with moderate energy densities and high-satiety ratings because they provide a solid dose of all the nutrients your body needs. Again, these foods may be overlooked if you only focus on energy density.
- Foods with lower energy densities tend to align with lower-calorie intakes. However, this may be a proxy for less fat.
- Once we consider protein and fibre, energy density becomes irrelevant.
- Free-living humans tend to gravitate towards more energy-dense foods.
- Low-energy-density foods may provide short-term satiation, fullness and reduced intake at the next meal. However, they seem less likely to provide long-term satiety.
- Multivariate regression analysis enables us to develop a satiety index score. Foods and meals with high satiety index scores tend to have moderate energy densities and provide more protein and micronutrients that will keep us satisfied for longer.