Exercise and appetite: why hunger and eating back calories differ

Exercise does not automatically make you eat back every calorie. Short-term appetite studies and months of training reveal different patterns.

日本語で読む（Japanese version）

On the way home from exercising, you stop to buy dinner. Alongside your usual meal, something sweet looks appealing. Are you hungry because you exercised, or does exercising feel like permission to add a treat? Those reasons can put the same item in your shopping basket, but they ask different questions about what happens after a workout.

The idea that the body immediately replaces every calorie used in exercise sounds plausible. Yet studies of the hours after exercise often find no clear increase in food intake. A trial lasting 24 weeks in adults with overweight or obesity found that increased intake could help explain why weight loss fell short of expectations. These results concern different periods of time, so they need to be read together rather than treated as contradictions.[3][6][16]

Hunger, the amount eaten, and long-term weight change are separate windows into the relationship between exercise and appetite. Looking through each window gives you a more useful way to think about dinner than assuming that a hungry evening makes exercise pointless. The research also leaves room for individual differences and for situations in which eating more after exercise really does occur.

Hunger and food intake answer different questions

Appetite studies may ask people how hungry or full they feel. They may measure the energy in the food people actually eat. Some also measure blood concentrations of hormones involved in appetite regulation. Feelings, food intake, and hormones are different outcomes. To investigate whether someone eats back the energy used in exercise, researchers need to consider expenditure alongside intake, rather than rely on a hunger rating alone.[4][7][14]

A comparison trial in 47 healthy men and women, including lean participants and participants with overweight or obesity, illustrates the distinction. Each person completed an exercise day and a resting day, with observation lasting 8 hours on each occasion. The exercise day began with 60 minutes on a treadmill. Participants then received standardized meals and, later, a buffet at which they could choose how much to eat. Exercise temporarily suppressed appetite, but energy intake at the buffet did not differ significantly between the conditions.[7]

The 95% confidence interval for the difference in intake was −391 to 346 kJ. For this buffet comparison, the interval spans both lower and higher intake after exercise, including no difference. It does not support a claim that everyone ate less after exercising, or that intake was proven identical. The study detected a reduction in appetite alongside no statistically significant difference in the meal outcome.[7]

The unit matters, too. Kilojoules, abbreviated kJ, and kilocalories, abbreviated kcal, both measure energy, but the numerical values cannot be compared as though the units were interchangeable. A number expressed in kJ should not be placed beside a number expressed in kcal and judged larger simply because its digits are larger. Keep the units attached when reading the findings.

Meal timing adds another layer. In the 47-person trial, standardized meals were served 1.5 and 4 hours after the experiment began, and the buffet was served at 7 hours. The treadmill exercise occupied the first hour. The food-intake comparison therefore included eating later in the observation period; it was not a measurement of the first bite immediately after getting off the treadmill.[7]

For an ordinary evening, the useful distinction is between how you feel and what you choose to eat. You might arrive home hungry and still eat your usual dinner. You might feel little hunger but add a dessert because it seems earned. Keeping a separate note of the sensation and the meal reflects the distinction researchers make. Neither entry can stand in for the other.

Researchers also distinguish absolute energy intake from relative energy intake. Absolute intake is the energy consumed in food. Relative intake subtracts the energy expended during exercise. A person can eat the same amount in both conditions while having a lower relative intake on the exercise day. This is why an unchanged meal and a short-term energy deficit can appear in the same study without any inconsistency.[6]

A single workout does not usually trigger immediate replacement

In research, acute exercise means a single bout of activity and the responses around it. A 2013 meta-analysis of food intake after a single bout of exercise included 29 studies and 51 trials. Exercise lasted 30–120 minutes, the experiments lasted 2–14 hours, and freely chosen test meals were offered 0–2 hours after exercise. These were measurements of a limited part of the day, rather than months of eating and weight change.[6]

The effect size for absolute energy intake was 0.14, with a 95% confidence interval of −0.005 to 0.29. The authors described this as a trivial effect. For relative intake, after accounting for exercise expenditure, the effect size was −1.35, with a 95% confidence interval of −1.64 to −1.05. That result pointed toward a short-term energy deficit.[6]

An effect size helps compare the magnitude of changes across studies. It is not the number of calories eaten or burned. In particular, −1.35 should not be read as a calorie allowance or a prediction of weight loss. Its role in this analysis is to describe the strength and direction of the pooled result for relative intake.[6]

Despite variation among studies, the authors concluded that people tended not to compensate for exercise expenditure by changing food intake in the immediate hours afterward. The observation period is part of that conclusion. A result obtained over several hours does not show what a participant ate the next day, and it cannot settle what happens after continuing an exercise program.[6]

A 2025 systematic review and meta-analysis focused specifically on people living with overweight or obesity. It included 19 studies and 34 trials. After a single bout of exercise, hunger and the perceived amount people wanted to eat decreased, and absolute intake also decreased. The effect size for absolute intake was −0.19, smaller than the −0.54 effect size for relative intake.[2]

The older analysis found almost no change in the amount eaten; the newer analysis, with a more specific participant group, found a small reduction. Those findings should retain their differences. They draw on different sets of studies and participants. Neither supports a blanket statement that exercising immediately causes overeating sufficient to replace all the energy used.[2][6]

For someone exercising after work in Tokyo, a normal dinner afterward is therefore not evidence that the session was wasted. The relevant comparison in these studies is intake after exercise versus intake without exercise. Participants were allowed meals under both conditions. Eating after activity is not itself the outcome researchers call compensation; the question is whether intake changes in response to the activity.[6][7]

Hormone changes do not determine the size of dinner

Acylated ghrelin often appears in explanations of exercise and appetite. Under resting conditions, this hormone is associated with stimulating food intake. Peptide YY is among the hormones associated with suppressing intake. These signals can help describe responses to exercise, but a change in a blood measurement does not, by itself, establish how much someone will eat.[4]

A 2014 meta-analysis of appetite-related hormones included 20 studies, 28 trials, and 241 participants. Men made up 77.6% of the sample. Acylated ghrelin decreased with exercise, with a median decrease of 16.5%. Peptide YY increased, with a median change of 8.9%. Publication bias was detected in these analyses, meaning that the available published research may give a selective picture of the evidence.[4]

The finding for glucagon-like peptide-1, usually shortened to GLP-1, was less definite. Its effect size was 0.275, with a 95% confidence interval of −0.031 to 0.581. The estimate pointed toward an increase, but the interval included no difference. The 2025 analysis in people with overweight or obesity also found that the increase in this hormone was not statistically significant. The certainty of the results differed across hormones.[2][4]

A 2023 review described appetite and hormone responses to moderate-to-vigorous land-based exercise as temporary. Hormone concentrations usually returned to resting values in the hours after exercise. That return was not necessarily accompanied by a compensatory increase in appetite or intake on the same day. How much these short-term responses matter for long-term weight management remained uncertain.[8]

A trial in 24 physically inactive adults compared resistance exercise, aerobic exercise, and a sedentary condition. Participants exercised after breakfast and later received a lunch at which they could choose how much to eat. Lunch intake was 991 kcal after resistance exercise, 937 kcal after aerobic exercise, and 944 kcal in the sedentary condition. Those numerical differences were not statistically significant.[14]

In that trial, resistance exercise produced lower measurements of both appetite-stimulating and appetite-suppressing hormones than aerobic exercise. Despite those differences, subjective appetite ratings and lunch intake did not differ significantly among conditions. Nor were the hormone measurements significantly different when exercise conditions were compared with the sedentary condition. The comparison between exercise types and the comparison between exercising and sitting therefore need to remain separate.[14]

The composition and methods of the hormone meta-analysis matter when applying it to women readers around Tachikawa. The 2014 analysis was predominantly male. It included studies that summarized hormone responses across multiple measurements during and after exercise, and excluded studies reporting only before-and-after values. Both who was studied and how the response was measured shape the scope of the result.[4]

Numbers: appetite after exercise

29 studies and 51 trials: the scale of the meta-analysis of food intake following a single bout of exercise.[6]

0.14: the effect size for absolute energy intake, with a 95% confidence interval of −0.005 to 0.29; the authors judged the change trivial.[6]

−0.19: the effect size for reduced absolute intake in the analysis focused on people with overweight or obesity.[2]

90.7 kcal/day: the estimated average increase in intake in the lower-exercise-dose group of the longer trial.[16]

123.6 kcal/day: the estimated average increase in the higher-exercise-dose group of that same trial.[16]