Can Diet and Exercise Really Change Metabolism?

, University of Alberta

Abstract and Introduction


Resting metabolic rate (RMR; kcal/min) accounts for approximately 60% to 75% of total daily energy expenditure. The energy required to digest and absorb food, also called the thermic effect of feeding, provides an additional 5% to 10%, and the energy expended in physical activity (thermic effect of activity) comprises approximately 20% to 30% of daily energy output. The major determinants of RMR include such factors as body surface area, age, gender, thyroid hormones and catecholamines, genetics, body temperature, and in women, the phase of the menstrual cycle. Most factors are beyond the control of the individual, yet it is believed that voluntary behaviors, specifically dietary intake and exercise, can influence resting metabolism--but for how long and by how much is not well-defined. In general, it is widely believed that dieting decreases metabolic rate and exercise increases it. However, present research suggests that any change may be short-term. Also, the relationship of diet, exercise, and metabolism is not a simple one. Very restrictive diets can result in a transient decrease in metabolic rate. Whether this decrease is sustained has not been clearly shown. Many believe that a permanently reduced metabolism is the reason that diets do not work and that weight regain after weight loss seems inevitable. Current research efforts are focused on ways to maintain or increase metabolic rate through diet and exercise.


Although current research efforts are focused on ways to maintain or increase metabolism through diet and exercise, it may be that a person's metabolic rate is relatively fixed and that only extreme changes in diet and/or exercise have a significant influence on resting energy expenditure. Evidence suggests that total daily energy expenditure is the more important value for consideration when trying to alter energy balance.

What Is Metabolic Rate?

An individual's metabolism is commonly measured at rest after a 12-hour fast so as to control for effects of digestion and physical activity. Thus, it is actually the resting metabolic rate (RMR; kcal/min) or resting energy expenditure (REE; kcal/day) that is reported in research studies investigating the effects of different interventions on metabolism.

The RMR represents the energy required to maintain vital body functions including respiration, heart rate, and blood pressure. More specifically, it is estimated that the energy consumed at rest is distributed among kidneys (7%), heart (10%), skeletal muscle (18%), brain (19%), liver (27%), and the remainder to other sources (19%).[1] Resting metabolic rate accounts for approximately 60% to 75% of total daily energy expenditure.[2] The energy required to digest and absorb food, also called the thermic effect of feeding (TEF), provides an additional 5% to 10%, and the energy expended in physical activity (thermic effect of activity, or TEA) comprises approximately 20% to 30% of daily energy output (Fig. 1).

Figure 1. Total energy expenditure (TEE) is the total number of kilocalories expended per day (kcal/d). About 60% to 75% is resting energy expenditure (REE), 20% to 30% thermic effect of activity (TEA), and 5% to 10% thermic effect of food (TEF).

What Factors Influence Metabolic Rate?

The major determinants of REE and RMR appear in prediction equations of energy expenditure and include such factors as body surface area (a calculation based on height and weight), age, and gender (Fig. 2). The greater the body surface area, the greater the RMR. Resting metabolic rate gradually declines with age, although a large component of this decline may be due to a decrease in muscle mass. For example, males have a higher RMR than females, primarily because of greater muscle mass. Fat-free mass (FFM) is a better predictor than surface area of oxygen consumption in young men.[3] Together, FFM, fat mass (FM), age, and gender account for approximately 80% of variance in RMR.[4]

Figure 2. Comparing two 30-year-old women of the same height and medium frame, the obese woman has a lower resting energy expenditure (REE) per kg of body weight than the average-weight woman. But the REEs are about the same when the average-weight woman's higher proportion of fat-free mass (FFM) is included in the calculation.

Other determinants of resting metabolism include thyroid hormones and catecholamines,[5,6] genetics,[7] body temperature,[8] and in women, the phase of the menstrual cycle.[9,10] An increased release of thyroid hormones or catecholamines increases RMR. Conversely, a reduction in energy intake specifically decreases triiodothyronine (T3) levels[11] and norepinephrine release,[12] which can subsequently decrease metabolic rate. An individual's RMR is, to a certain extent, genetically determined, and it can also increase in response to extremes in body temperature. In addition, it has been estimated that during the luteal phase of the menstrual cycle, there may be up to a 5% increase in RMR.[9] Although most factors are beyond the control of the individual, it is believed that voluntary behaviors, specifically dietary intake and exercise, can also influence resting metabolism.[6]

Although there is a significant amount of information in both the scientific literature and in the public media regarding metabolic rate, this information is often contradictory. Whether metabolic rate can be altered and the extent to which it can be altered are not well-defined. In general, it is widely believed that dieting decreases metabolic rate and exercise increases it.

Common Weight-Loss Practices

Weight-loss practices of overweight adults were assessed in a report based on data from the Behavioral Risk Factor Surveillance System in the US (1985-1988).[13] Information from 21 states was available for 112,108 individuals who were considered to be overweight based on a body mass index (BMI) of at least 27.3 for women or at least 27.8 for men. The 4 possible weight control practice categories were:

  • increasing physical activity

  • eating fewer calories

  • increasing physical activity plus eating fewer calories

  • not trying to lose weight

For those who were trying to lose weight, the most common weight practices were increasing physical activity plus taking in fewer calories (33.8% of women, 19.9% of men) and just taking in fewer calories (32.3% of women, 24.1% of men). Increasing physical activity as a sole weight-loss practice was reported by 3.4% of women and 4% of men.[13]

Since dieting and exercise continue to be the predominant strategies used by individuals who are concerned about weight loss and weight maintenance, the remainder of this report focuses on how dieting, exercise, or dieting plus exercise influence resting metabolic rate.

Dieting and Metabolic Rate

Does dieting (energy restriction) decrease metabolic rate? Restrictive dieting can cause RMR to drop, at least temporarily. There is debate as to whether this decline is directly related to the loss of protein or FFM , or if there is an adaptive mechanism for energy conservation that lowers RMR below that which can be explained by losses in lean body tissue. As body protein is degraded, increased nitrogen is lost from the body. Tissue nitrogen loss during energy restriction depends on the degree and duration of the restriction, the composition of the diet, and how the nitrogen losses are measured.[14] Methodologic considerations, such as the amount of time between RMR measurements and the accuracy of the techniques used to assess body composition, also have an impact on the magnitude of the change in RMR observed.

In one report, 24 obese, postmenopausal women consumed a low-calorie diet for 3.5 months[15] and lost a significant amount of weight (12.7 [+/-0.45] kg). Approximately 77% of the weight loss was from fat and 23% from FFM. Their REE decreased over the study period from a mean value of 1310 (+/-26) kcal/d to 1161 (+/-23) kcal/d. Despite this 11.4% drop, the mean REE of the food-restricted obese women in the study was similar to that of a comparison group of 24 nonobese, postmenopausal women (REE =1163 [+/-28] kcal/d). Thus, although REE of the previously obese subjects decreased after weight reduction, one might also consider that it "normalized" to that of individuals of similar body composition who had not lost weight. The TEF of the obese group after consuming a test meal was not statistically different after dieting (51 [+/-4] kcal) than before dieting (52 [+/-4] kcal); although the TEF was lower than that of the control subjects (56 [+/-3] kcal), the difference was not statistically significant.

Based on a meta-analysis of 22 comparable studies on dieting and metabolic rate, Thompson and coworkers[16] determined that RMR decreases significantly when an individual participates in a diet program and consumes an energy level of approximately 941kcal/d or less. These studies primarily included female subjects who were between the ages of 31 and 45 years, and the caloric intakes varied between studies.

History of dieting and metabolic rate. We attempted to determine whether a history of dieting results in a sustained reduction in metabolic rate in overweight women.[17] A total of 52 sedentary women who had a long history of dieting (age of first diet 22 years of age [+/-9 years]), 40 (+/-7) years of age with a BMI of 28.2 (+/-2.2), participated in the study. The REE was measured by indirect calorimetry; body composition was determined by skinfold measurements. At the time of the study, the women had abandoned dieting as a weight-loss method and their weights were stable. Their measured REE was 1429 (+/-147) kcal/d, which was not significantly different from the REE of 1418 (+/-120) kcal/d predicted using the equation reported by Mifflin and colleagues.[18] These women definitely had a long history of dieting, weight cycling, and restrained eating. Whether any specific episode of dieting affected their metabolic rate is not known. At the time of this cross-sectional study, however, the REE of these women was at a level predictable for their height, weight, and age, regardless of dieting history.

Dieting Plus Exercise and Metabolic Rate: Impact and Issues

It has been hypothesized that the addition of exercise to a diet program will accelerate fat loss, preserve fat-free weight, and prevent or decelerate the decline in resting metabolic rate more effectively than will diet restriction alone.[19] Poehlman and coworkers[19] have extensively reviewed numerous studies that examined the combined effects of diet and exercise on body composition and resting metabolic rate. They found that the optimal combination of diet and exercise remains elusive.

More recently, Thompson and colleagues[16] in their meta-analysis of 22 studies also investigated the effect of diet plus exercise. The duration of the studies ranged from 1-16 weeks. Thompson's group found that when researchers added moderate-intensity aerobic exercise to a low-calorie diet, some of the decrease in RMR was prevented. However, the RMR did not return to baseline values during the study period. The mechanism for this effect has been attributed, in part, to the preservation of FFM or at least to decreased loss of FFM when exercise is added to a dieting protocol. The effect appears to be highly dependent on the type, duration, and intensity of the exercise prescription.

Donnelly and coworkers[20] also investigated the effect of low-calorie diets and exercise on body composition but found quite variable results. Retention of FFM was observed in some, but not all, studies when exercise was added to a diet protocol. Comparisons between studies were difficult due to different methodologies used, poor documentation of the exercise intervention, and perhaps provision of an exercise stimulus too low in intensity. They concluded that if diets are properly prescribed and if the subjects are well monitored, the loss of FFM should be limited to 20% to 30% of total weight loss, which is considered reasonable. That target for conservation of FFM is generally accepted by researchers who believed that when weight is lost under conditions of moderate exercise and dieting, the composition of that weight loss is approximately 75% fat and 25% lean tissue.

Examining whether different exercises produce different metabolic effects. Whenever the subject of exercise in relation to diet and metabolism is raised, a key issue is the impact of different types of exercise. In one study of 3 months' duration, 69 obese women were assigned to 1 of 4 treatment groups:

  • diet only (control)

  • diet plus endurance exercise

  • diet plus weight training

  • diet plus endurance exercise and weight training[21]

Participants consumed a liquid diet of 522kcal/d. Changes in body weight, percentage body fat, FM, and FFM did not differ between groups. Declines in RMR were approximately 7% to 12% of baseline, with no differences among groups. The main effects of exercise were related to a strength index, which was calculated as the sum of 1-repetition maximum values (kg) for a bench press, lateral pull-down, knee extension, and knee-flexion exercise. The strength index decreased 6% in the control and endurance exercise groups, and it increased in both the weight-training (10%) and the endurance-exercise plus-weight-training (3%) groups.[21] Neither aerobic endurance exercise nor weight training appeared to prevent the diet-induced decreases in RMR.

Expressing RMR: Absolute versus relative values. Since changes in FFM are often a result of diet and exercise, another key issue when examining changes in metabolism related to diet and exercise is whether to include FFM as a correction factor. In fact, research results can be quite different between the absolute RMR and the relative RMR corrected by FFM measurements. One outpatient weight-loss program that provided nutrition education, behavior modification, exercise, and psychological counseling followed some participants for more than 3 years.[22] Test/retest metabolic measurements were available for 35 subjects (28 females, 7 males) at 15 (+/-5.3) months (range, 9-42 months). The subjects had no history of medication or medical problems that might affect metabolic rate. Their mean age was 40 (+/-12) years old, and body weights were relatively stable at the time of retesting. The subjects had lost 19% of their body weight, of which approximately 72% was from fat. Absolute RMR (kcal/day) decreased by 11%; however, relative RMR (kcal/kg FFM) was not different from baseline, and RMR presented as kcal/kg body weight significantly increased. The investigators discussed the issues surrounding appropriate units used for reporting RMR. One argument against using FFM as a correction factor is that the proportion of FFM relative to total mass is different at retest; a direct comparison of relative RMR before and after weight loss is therefore not valid.

In reporting on the effect of diet and exercise on body composition, Ballor and Poehlman[23] reviewed 46 human studies to determine how exercise training and gender influence the composition of diet-induced weight loss. All of the studies had a minimum of 4 weeks' duration, included diet restriction, and determined body composition by either skinfold calipers or underwater weighing techniques. Otherwise, they varied in protocol, and not all included an exercise component. Comparisons were made between diet-only groups and diet-plus-exercise groups, as well as between males and females. The main finding from this review was that absolute (kg) and relative (%) loss of FFM was significantly less in the diet-plus-exercise groups compared with the diet-only groups. This finding was consistent in both males and females. Of the weight that was lost, approximately 72% was from fat and 28% was from FFM in males; 76% was from fat and 24% was from FFM in females. The conservation of FFM occurred with exercise training that was fairly modest (generally 3 days per week of low- to moderate-intensity exercise).[23]

In summary, it appears that restrictive diets (anything less than 1000kcal/d) result in an acute decrease in metabolic rate. This value is only an estimate and varies with the individual. Persons with a history of intermittent dieting do not appear to have a permanent reduction in metabolic rate. The addition of exercise to a diet protocol does seem to preserve FFM and RMR to a certain degree, but not completely, throughout the course of a weight-loss program.

Can We Increase Our Metabolic Rate?

There is evidence to suggest that increased food intake can increase RMR. This has been demonstrated in particular in patients with anorexia nervosa participating in nutritional repletion programs. Increasing food intake alone can result in an increase in RMR greater than that which would be expected in association with the body mass gain.

It is also known that exercise increases metabolic rate during a bout of exercise and immediately after the exercise. However, whether there is a longer-term effect remains unresolved. Many of the studies in this area have investigated exercise effects during an energy restriction, but the results are mixed. Ballor[24] suggests that the more severe the dietary restriction, the smaller the effect of exercise training on metabolic rate and conservation of FFM. In animal studies, an interaction between the severity of dietary restriction and exercise training has been observed: Exercise has been found to elevate RMR during moderate dieting, but not during severe dietary restriction.[24] That is, a certain level of dietary intake seems to be needed for exercise to stimulate metabolic processes effectively.

In a study to determine the effect of exercise regimen on metabolic rate and body composition, women who had a history of chronic dieting were randomized into 2 exercise groups (endurance training and resistance training).[25] They were compared to a control group of women who did not exercise. The prescribed exercise program was of moderate intensity, designed to minimize potential injury and maximize compliance. One group participated in an outdoor walking program of progressively increasing distances. The other group participated in a weight-training fitness class of progressively increasing resistance exercises. The exercise was done 3-4 times per week for 3 months. At this level of intensity, there were no exercise-induced changes in FFM or REE, regardless of the type of exercise. In addition, there was no weight loss in either of the exercise groups, but the control group gained an average of 2.5kg over the same 3-month period. Positive effects resulting from the exercise included an increase in fitness level and maintenance of initial body weight for the duration of the study. This study suggests that exercise at a level beyond moderate intensity may be required to induce changes in FFM and REE.[25] However, the tradeoff is this: A sedentary overweight population may be less likely to adhere to a more vigorous exercise program.

One study that showed a significant effect of exercise on RMR was done in an elderly population.[26] This study involved an exercise program for 18 elderly people (66 years old [+/-1.4 year]; 10 males, 8 females) who participated in cycling exercise for an 8-week period (3x/wk). For the first week they cycled at 60% of maximal oxygen consumption (VO2 max) until 150kcal was expended. By the eighth week, they cycled at 80% VO2 max until 300kcal was expended. There was no change in body weight, FM, or FFM during the study. VO2 max significantly increased from 1.9 (+/-0.1) L/min to 2.1 (+/-0.1) L/min (P <0.01). When measured on an outpatient basis, exercise training significantly increased RMR by 11% (P <0.01) (pre 1.17 [+/-0.031] kcal/min versus post 1.29 [+/-0.019] kcal/min). It was suggested that this increase is not just related to the exercise but to an increased energy flux. Increased food must have been consumed to compensate for the energy expenditure, as there was no weight change over the 8-week period. Both nutritional and sympathetic factors can mediate changes in metabolism in response to endurance exercise.[26]

Adequate food intake appears to be a key signal to the body to maintain energy-producing processes. In essence, it takes food to stimulate and sustain metabolism. For a depleted patient with anorexia nervosa, increased food intake can stimulate RMR. For an individual trying to lose weight, an adequate food intake can complement exercise, whereas a restricted food intake can antagonize an exercise program. The idea of maintaining a certain level of energy flux appears to be an important one. Increasing muscle mass and therefore FFM through weight training indirectly should increase RMR. The level of training required to do this has not been clearly shown, however, and it appears to be a very high intensity level.

Application: A Broader View

There is perhaps an unrealistic expectation regarding the potential for exercise in terms of weight loss, FFM gain, and increases in RMR. Zelasko[27] addressed this point in terms of weight loss, stating that the perception about the role that exercise can play in energy expenditure has become exaggerated over time. The big factor in the fat-burning phenomenon is time. Zelasko states that if total energy expenditure is the focus of the exercise, then higher-intensity exercise is better, because even though it uses a lower percentage of fat, it expends more fat energy and more total energy.[27]

The extent of the postexercise energy expenditure has also been questioned.[28] Excess postexercise oxygen consumption (EPOC) is the energy expended above resting values after a bout of exercise has been completed. It is highly dependent on exercise intensity and duration as well as individual and environmental factors. It is acknowledged that some extra calories are utilized after exercise. However, the energy expended from EPOC for the majority of people doing light or moderate activity is unlikely to have a significant effect on energy balance or weight loss.[28]

This does not mean that exercise is not important. In fact, it is one of the best predictors of long-term weight maintenance. However, it should be considered in the context of a 24-hour day. It is total energy expenditure that is a critical factor in energy balance. Given a fairly wide range of diet and exercise routines, RMR and TEF are likely to remain relatively fixed. It is the physical activity component that makes up the remainder of daily energy expenditure; this is the most variable component and the most amenable to significant change. Planned exercise is important, with the greatest emphasis on consistency, then duration, intensity, and frequency.[27] However, unplanned exercise (ie, strategies to be more active during the remainder of the day) is also a key factor. It is total daily energy expenditure that must be considered for successful long-term weight loss and/or weight maintenance.


Based on data from studies of diet and exercise and their effects on metabolism, the following conclusions can be drawn. Diets that restrict intake to less than 1000kcal/d should be avoided because such diets can result in a decrease in RMR and can antagonize the benefits of exercise, at least over the short term. A caloric intake that falls between REE and TEE--that is, a caloric restriction of approximately 15%-20%-- has been suggested for gradual weight loss. Exercise should be a component of any weight-loss program. Aerobic activity performed a minimum of 3x/wk with a gradual increase in intensity and frequency is suggested, with the further addition of a weight-training component. FFM and RMR will be preserved somewhat. Total daily energy expenditure should be the key consideration in strategies to alter energy balance.


  1. Passmore R, Draper MH: The chemical anatomy of the human body, in Thompson RH, Wooten ID, (eds): Biochemical Disorders in Human Disease, ed 3, London: Churchill, 1970, pp 1-14.

  2. Danforth E: Dietary induced thermogenesis: Control of energy expenditure. Life Sci 28:1821-1827, 1981.

  3. Miller AT, Blyth CS: Lean body mass as a metabolic reference standard. J Appl Physiol 5:311-316, 1953.

  4. Ravussin E, Harper I, Rising R, et al: Human obesity is associated with lower levels of physical activity: Result from doubly-labeled water and gas exchanges. FASEB J 5:554A, 1991.

  5. Bray GA, Atkinson RL: Factors affecting metabolic rate. Prog Food Nutr Sci 2:395-403, 1977.

  6. Danforth E: The role of thyroid hormones and insulin in the regulation of energy metabolism. Am J Clin Nutr 38:1006-1017, 1983.

  7. Bouchard C, Tremblay A, Nadeau A, et al: Genetic effect in resting and exercise metabolic rates. Metabolism 38:364-370, 1989.

  8. Poehlman E: A review: Exercise and its influence on resting metabolism in man. Med Sci Sports Exerc 21:515-525, 1989.

  9. Solomon SJ, Kurzer MS, Calloway DH: Menstrual cycle and metabolic rate in women. Am J Clin Nutr 36:611-616, 1982.

  10. Webb P: 24-hour energy expenditure and the menstrual cycle. Am J Clin Nutr 44:614-619, 1986.

  11. Loucks AB, Callister R: Induction and prevention of low T3 syndrome in exercising women. Am J Physiol 264:R924-R930, 1993.

  12. McCargar LJ, Clandinin MT, Fawcett DM, et al: Short term changes in energy intake and serum insulin, neutral amino acids and urinary catecholamine excretion in women. Am J Clin Nutr 47:932-941, 1988.

  13. Bennett EM: Weight-loss practices of overweight adults. Am J Clin Nutr 53 (6 supp):1519S-1521S, June 1991.

  14. Nettleton JA, Hegsted DM: Protein-energy interrelationships during dietary restriction: Effects on tissue nitrogen and protein turnover. Nutr Metab 18:31-40, 1975.

  15. Nelson KM, Weinsier RL, James LD, et al: Effect of weight reduction on resting energy expenditure, substrate utilization, and the thermic effect of food in moderately obese women. Am J Clin Nutr 55:924-933, 1992.

  16. Thompson JL, Manore MM, Thomas JR: Effects of diet and diet-plus-exercise programs on resting metabolic rate: A meta-analysis. Int J Sport Nutr 6:41-61, 1996.

  17. McCargar LJ, Sale J, Crawford SM: Chronic dieting does not result in a sustained reduction in resting metabolic rate in overweight women. J Am Diet Assoc, 1996. In press.

  18. Mifflin MD, St Jeor ST, Hill LA, et al: A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr 51:241-247, 1990.

  19. Poehlman ET, Melby CL, Goran MI: The impact of exercise and diet restriction on daily energy expenditure. Sports Med 11:78-101, 1991.

  20. Donnelly JE, Jakicic J, Gunderson S: Diet and body composition. Effect of very low calorie diets and exercise. Sports Med 12:237-249, 1991.

  21. Donnelly JE, Pronk NP, Jacobsen DJ, et al: Effects of a VLCD and physical-training regimens on body composition and resting metabolic rate in obese females. Am J Clin Nutr 54:56-61, 1991.

  22. Heshka S, Yang M-U, Wang J, et al: Weight loss and change in resting metabolic rate. Am J Clin Nutr 52:981-986, 1990.

  23. Ballor DL, Poehlman ET: Exercise training enhances fat-free mass preservation during diet-induced weight loss: A meta-analytical finding. Int J Obes 18:35-40, 1994.

  24. Ballor D: Exercise training elevates RMR during moderate but not severe dietary restriction in obese male rats. J Appl Physiol 70:2303-2310, 1991.

  25. Sale JE, McCargar LJ, Crawford SM, et al: Effects of exercise modality on metabolic rate and body composition. Clin J Sport Med 5:100-107, 1995.

  26. Poehlman ET, Danforth E Jr: Endurance training increases metabolic rate and norepinephrine appearance rate in older individuals. Am J Physiol 261:E233-239, 1991.

  27. Zelasko CJ: Exercise for weight loss: What are the facts? J Am Diet Assoc 95:1414-1417, 1995.

  28. Bahr R: Excess postexercise oxygen consumption--magnitude, mechanisms, and practical implications. Acta Physiol Scand Suppl 605:1-70, 1992.