Energy density and weight loss without shrinking every meal

Lower energy density can mean fewer calories without less food. Trials show why feeling full, eating less energy, and losing weight differ.

日本語で読む（Japanese version）

You make dinner smaller, finish the plate, and still feel that something is missing. Later, you open the refrigerator. The portion you removed at dinner may find its way back into the evening somewhere else. A smaller meal can look like a complete plan on the table while leaving the rest of the night unresolved.

There is another question to ask before shrinking everything: must the amount of food and the energy it contains fall together? A meta-analysis of randomized controlled trials found that lowering the energy density of meals reduced energy intake without changing the amount of food eaten.[1] A satisfying portion and fewer calories are not necessarily incompatible.

That possibility comes with an essential distinction. Feeling fuller, consuming less energy that day, and losing weight over time are separate outcomes. In one trial, a lower-density ready meal increased fullness but did not significantly reduce energy intake at dinner.[5] The useful question is how to change the composition of a meal while keeping track of which outcome actually changes.

Energy density measures calories per gram of food

Energy density describes the energy contained in a given weight of food, expressed as kcal/g, or calories per gram. A review applying behavioral science to weight management identifies water and fat as especially important components: water provides 0 kcal/g and fat provides 9 kcal/g. Incorporating water-rich ingredients, including vegetables and fruit, can therefore lower a food’s energy density.[2]

The measure concerns the relationship between weight and energy. It does not belong permanently to a dish just because that dish has a familiar name. Its ingredients matter. The practical approach described in the review is to change the composition of meals so that people can eat satisfying portions with more water-rich foods. That could mean incorporating vegetables into a familiar dish or including a dish with more liquid as part of the meal.[2]

The review also discusses soup, salad, or fruit eaten as a first course. Studies described there found that this approach could enhance satiety, the feeling of being satisfied after eating, and reduce overall energy intake at the meal.[2] The point is how the food fits into the meal. It is not a rule that anything called soup will produce the same response or that an extra course automatically cancels calories elsewhere.

The phrase “amount of food” needs care. Food weight, the space food occupies on a plate, and its calorie content are different things. In the meta-analysis, food intake stayed constant while energy density and energy intake showed a positive linear relationship: higher density went with higher energy intake.[1] Serving a meal in a deeper bowl does not change its ingredients. The bowl cannot rewrite the food’s composition.

Energy density is also a guide to one aspect of food choice, rather than a complete score for nutritional quality. In the ready-meal trial discussed later, changing density changed intake of several nutrients, including salt.[5] Comparing density can be useful while still leaving a reason to read the rest of the label. A single number does not describe every difference between the foods being compared.

This topic sits beside portion size, whole grains, and nuts in the wider series. Portion size asks how much food is served; energy density asks what energy is contained in the weight being served. Questions about individual food groups belong in their own articles. Here, the evidence concerns experiments that changed the construction of meals, rather than a list of foods to ban.

Less energy does not always require less food weight

A meta-analysis of 38 studies in children and adults compared objectively measured energy intake from meals with different energy densities under controlled conditions. The median density was 1.1 kcal/g in the lower-density conditions and 1.5 kcal/g in the higher-density conditions. Lower density reduced energy intake by an average of 223 kcal, with a 95% confidence interval ranging from 259.7 to 186.0 kcal less.[1]

The 259.7 to 186.0 kcal reduction is the uncertainty range around the estimated average difference in these controlled meal comparisons.[1]

The studies varied substantially. They included children and adults, experiments using a food eaten before the main meal and experiments changing the main dish itself, and comparisons spanning a single meal or more than one meal. Differences between study results were large. When the researchers grouped studies by age and meal type together, that variation became smaller.[1] Those details help explain why the pooled estimate is useful without becoming a personal daily allowance.

The finding still changes the starting question at the dinner table. After a late journey home, dinner may be something you look forward to. Before deciding to make every part of it smaller, you can ask where water-rich ingredients could fit into the dishes you already eat. That is an example of the review’s approach to keeping a satisfying quantity while changing food composition.[2] It begins with the meal you want to continue eating.

Lowering the density of a meal and adding a lower-density food are different actions. The review describes strategic incorporation of foods into the diet.[2] Eating the usual dinner and then adding another dish afterward is a different situation from altering the composition of the dinner itself. “I added vegetables” tells you what was added; it does not establish that the day’s total energy intake fell.

In a controlled experiment, researchers adjust the density of the food that is served. At home, you can change both the meal and what you choose afterward. A useful translation of the research is therefore concrete: identify the dish you changed, the ingredients it replaced, and any food added later. These questions keep the comparison visible. Otherwise, a meal can acquire a lower-density label while its place in the whole day’s eating remains unclear.

Does the reduction survive the rest of the day?

A lower-calorie lunch does not settle the daily total if dinner makes up the difference. A systematic review and meta-analysis of 31 experimental studies examined food-density changes while measuring energy intake for at least a full day. It included 4 studies in children and 27 in adults. Serving lower-density food was associated with a large reduction in daily energy intake.[6]

The pooled standardized mean difference was −1.002.[6] This is a way of expressing results from different studies on a common scale. It is neither a calorie difference nor a kilogram change. The researchers found little compensation through eating more at other meals. Results were consistent in studies that changed macronutrient composition to alter density and in those that did not.[6] That gives support to examining the day, rather than assuming a meal-level reduction must immediately disappear.

There is also an instructive exception. A randomized controlled crossover trial in 20 healthy men examined foods relevant to military rations, where limited food volume can make it difficult to consume enough energy. The foods began at moderate or higher energy density, at least 1.6 kcal/g. Researchers manipulated both energy density and volume density, meaning the weight of food contained in the same volume, in selected parts of the diet.[15]

Under the lower-energy-density condition, intake from the experimental foods was 291 kcal lower, but intake from other, self-selected foods was 203 kcal higher. The overall difference was −88 kcal, with a 95% confidence interval of −282 to 105 kcal. Total energy intake was not significantly different.[15] Looking only at the experimental food would have missed the compensation occurring elsewhere.

The men experienced different conditions, each lasting 2 days, in random order. Experimental foods were held to the same volume and engineered using leavening, physical compression, and fat manipulation.[15] This design distinguished a change in calories per gram from a change in how much food weight occupied that volume. Those are separate properties, even if both can be described casually as making food “denser.”

When less food weight occupied the same volume, energy intake from the experimental foods was 417 kcal lower and total energy intake was 276 kcal lower. The confidence interval for the total difference was −470 to −83 kcal, and the difference was statistically significant.[15] Compensation therefore differed depending on the property being manipulated. A calorie change in selected food items and a calorie change in the complete diet did not always move together.

This was also a particular eating arrangement. Consumption of the experimental foods was compulsory at specified eating occasions, while the remaining food was self-selected. The experimental items were incorporated into meals and snacks.[15] That differs from simply allowing people to eat freely from lower-density meals. Where participants have freedom to adjust their intake is part of the study design, and helps explain what the results can tell us.

If you change dinner at home, consider the food eaten afterward as part of the same experiment. An evening addition is information about how the changed meal worked in your routine. There is no need to make it a verdict on your willpower. The pooled evidence found little compensation, but this individual trial found no significant overall energy difference after compensation.[6][15] Both results belong in a fair account of the evidence.

Numbers: energy density and the outcomes it changes

Across 38 studies, lower-density meals reduced energy intake by an average of 223 kcal; the 95% confidence interval was 259.7 to 186.0 kcal less.[1]

Across 31 studies measuring at least a full day, compensation through eating more at other meals was small.[6]

In 26 women, equal-calorie ready meals increased fullness in the lower-density condition without significantly changing dinner energy intake.[5]

In a year-long trial of 97 women, completers lost 7.9 kg with fat reduction plus more fruit and vegetables, versus 6.4 kg with fat reduction alone.[3]

In a long-term trial of 2,718 women, lower dietary energy density persisted for 4 years, but body weight did not differ between groups at that point.[11]