Physical activity and preference for selected macronutrients


Medicine & Science in Sports & Exercise:
Roundtable Consensus Statement

TREMBLAY, A., and V. DRAPEAU. Physical activity and preference for selected macronutrients. Med. Sci. Sports Exerc., Vol. 31, No. 11, Suppl., pp. S584–S589, 1999.

Objective: The impact of physical exercise on macronutrient preferences was examined with a perspective to improve preventive and therapeutic strategies of obesity.

Design: The literature was reviewed pertaining to the acute effects of physical activity and the short-term and chronic effects of exercise training on macronutrient preferences.

Results: The presently available literature does not permit to establish a consensus regarding the impact of physical activity, be it acute or long-term, on macronutrient selection. However, one observation stands out and that is the fact that dietary fat intake needs to be controlled in order for exercise to produce a negative energy and fat balance.

Conclusion: Because active individuals do not systematically choose foods that are low in fat content, it is important to provide nutritional guidelines in a context where physical activity aims at reducing or better controlling body weight.

Author Information

Division of Kinesiology, PEPS, Laval University, Ste-Foy, Québec, G1K 7P4, CANADA

Address for correspondence: Angelo Tremblay, Ph.D., Division of Kinesiology, PEPS, Laval University, Ste-Foy, Québec, G1K 7P4, Canada; E-mail:

Roundtable held February 4–7, 1999, Indianapolis, IN.

Article Outline

The lack of physical activity in an industrialized world is a factor that likely explains a significant portion of the increased risk to develop obesity among sedentary individuals. Theoretically, the corollary of this statement implies that weight-reducing programs based on the regular practice of aerobic physical activities should favor a substantial body weight and fat loss. However, clinical experience and experimental data reveal that the use of exercise alone to induce long-term negative energy balance does not systematically produce the expected outcome. This phenomenon may be explained by the fact that the energy cost of additional exercise is sometimes not sufficient to produce a significant change in daily energy balance. The failure of regular activity participation to induce weight loss in some obese individuals might be also partly attributable to a compensation in postexercise energy expenditure and intake. Specifically, this means that a decrease in nonexercise physical activity participation and an increase in postexercise energy intake can compensate substantially or even totally for the surplus of energy expended during physical activities.

In a recent series of studies, diet composition has been shown to exert a significant influence on postexercise energy intake. In the first study to document this issue, the intake of a diet conforming to dietary guidelines for fat intake prevented the compensation in postexercise energy intake to overcome the energy cost of exercise (23). Conversely, a high-fat diet was associated with a large overfeeding that was sufficient to totally compensate for the impact of exercise on energy expenditure. In subsequent studies, the physical activity-low fat diet combination was also found to permit a substantial acute negative energy balance (5,14). Taken together, these observations suggest that the activity-low fat diet combination can induce an acute daily energy deficit that can be as large as 1000 kcal·d1 (4.2 MJ·d1), but they also draw attention on the possibility that activity practice may result in an overall positive energy balance if diet has a high fat content.

The effect of physical activity and diet composition on energy balance emphasizes the importance to make appropriate food choices if an active individual wishes to lose or maintain body weight. This rationale also raises the question as to whether exercise per se can alter food choices and/or macronutrient selection. The main aim of this paper is to document this issue with a perspective to improve preventive and therapeutic strategies of obesity.

Back to Top | Article Outline


Evidence statement.

There is a need toward the development of standardized measurements of macronutrient preferences (evidence category B).

The concept of macronutrient preference is not systematically considered by nutrition agencies. In practical terms, a macronutrient preference likely reflects a preferential taste for foods having a high content of a specific macronutrient. There is no well-accepted specific measurement for this phenotype, but it can be estimated with different indirect measurements. One of them is the habitual macronutrient content of the diet, be it assessed by dietary record, recall, or history. According to this procedure, a high percentage of ingested calories from a given macronutrient is assumed to reflect an increased preference for this macronutrient.

A second strategy to assess macronutrient preference consists of using a buffet test meal that offers a variety of foods of different macronutrient composition. Such a test is generally performed in the laboratory under conditions mimicking as closely as possible free living conditions. In this context, it is postulated that a preference for a specific macronutrient will be reflected by a preferential selection of foods having a high content of this macronutrient.

Macronutrient preference can also be evaluated by the search of a macronutrient mix providing the greatest appreciation by a subject. An example of this experimental strategy was used by Drewnowski and Greenwood (6), who tested different carbohydrate-lipid solutions to determine whether human obesity is associated with a specific macronutrient preference. Their results revealed a greater preference for lipid in obese and postobese subjects than in their lean counterparts.

Up to now, there is no consensus on how to specifically measure macronutrient preference, and this phenotype is thus measured with different strategies that do not necessarily reflect the same phenomenon across available studies. Therefore, if macronutrient preference is proven to be a key issue in the study of energy balance, it will be relevant to pursue investigations in this area to establish an operational definition of this concept.

Back to Top | Article Outline


Evidence statement.

The acute effect of physical activity on macronutrient preference is not clearly established (evidence category B).

Prolonged vigorous physical activities are known to induce glycogen depletion, which may represent an effect promoting a specific preference for carbohydrate to facilitate glycogen store replenishment (14). This hypothesis is partly documented by studies in which the acute effect of exercise on macronutrient preference has been tested. As shown in Table 1, animal experimentation does not systematically support this idea.

This table also summarizes human studies that have specifically focused on postexercise macronutrient preference. In their first study pertaining to this issue, Verger et al. (24) found an increased preference for foods with high carbohydrate content after a prolonged exercise. However, this finding could not be reproduced in a subsequent study (25). Indeed, by using a comparable exercise test, they observed an increase in relative protein intake. In a more recent study, Westerterp-Platenga et al. (26) noted an increased preference for carbohydrate and a decreased preference for fat after exercise. In addition, these authors criticized the experimental approach of Verger et al. for the reduced availability of foods containing predominantly one macronutrient in their buffet-type meal. Accordingly, Horio and Kawamura also found a preference for sucrose after exercise in university students (9).

In our laboratory, we examined the impact of exercise sessions of different intensity but of similar energy cost on the composition of postexercise ad libitum intake from a buffet-type meal (10). The results showed no effect of exercise on postexercise food quotient (FQ). At best, we found a nonsignificant positive correlation between exercise RQ and postexercise FQ.

In summary, the available literature pertaining to the acute effect of physical activity on macronutrient preference does not permit to identify a clear effect of activity. In addition, available data also do not allow to establish a relationship between the composition of the substrate mix oxidized during exercise and spontaneous food selection in the postexercise state.

Back to Top | Article Outline


Evidence statement.

There is no clear effect of short-term exercise-training on macronutrient preference (evidence category B).

A summary of the main studies documenting the impact of short-term exercise-training on macronutrient preference in animals and humans is presented in Table 2. A diversity of effects is reported in animals, ranging from an enhancing effect of training on spontaneous carbohydrate and protein intake or either the opposite or just a significant decrease in fat intake. Interestingly, ad libitum energy intake was found to fluctuate according to changes in macronutrient intake. An accentuation of the preference for carbohydrate and/or a decrease in spontaneous fat intake were associated with no compensation or a decrease in energy intake.

Table 2 also shows that data collected in humans are less conclusive. Indeed, the majority of available data tend to show that short-term training has no effect on both macronutrient selection and energy intake. This table also presents recent data demonstrating that short-term training can increase preference for fat in female subjects only (1). In summary, as for data related to the acute effect of aerobic activities, the currently available literature does not permit to identify a systematic effect of the regular practice of physical activity over several weeks on macronutrient selection.

Back to Top | Article Outline


Evidence statement.

There is no specific effect of chronic training on macronutrient preference. However, chronic training increases daily energy intake in trained individuals (evidence category B).

Theoretically, the long-term exposure to physical activity represents the condition that is expected to exert the greatest impact on macronutrient selection. Indeed, the feeding behavior of the regular exerciser is not only affected by the exercise stimulus but also by changes in body fatness that may influence macronutrient preference.

Table 3 summarizes human studies that have addressed this issue. Two of these studies were longitudinal interventions. In these two studies, neither a swimming program (11) or a rowing training program (4) had significant effect on habitual diet composition.

Table 3 also presents data obtained in cross-sectional studies comparing habitual macronutrient intake in trained individuals and sedentary controls. Again, no systematic pattern of food selection can be detected from this set of data despite the fact that these individuals are frequently exposed to dietary information promoting an increase in carbohydrate intake. It is, however, of interest to note that in each case, reported daily energy intake was increased in trained individuals.

Back to Top | Article Outline


Evidence statement.

The practice of physical activity is not associated with a preference for specific macronutrients in active individuals. Consequently, fat intake needs to be controlled in active individuals to favor a negative energy balance (evidence category B).

The analysis of studies documenting habitual macronutrient intake in relation to the practice of physical activities does not permit to clearly identify particularities in the feeding behavior of active individuals. Indeed, available data suggest that the exposure to an exercise stimulus may favor a preferential consumption of foods having a high content of either carbohydrate, fat, protein, or alcohol. In other words, physical activity seems to have the potential to induce about any change in macronutrient intake, depending on circumstances that are not well characterized and understood. In fact, the study of macronutrient preferences will remain a difficult issue of investigation as long as the following points will not be clarified: 1) the concept of macronutrient preference deserves a better characterization and should be measured using standardized methodology, and 2) physical activity-related factors and underlying biological mechanisms influencing macronutrient preferences should be characterized. In addition, the study of macronutrient preference under free living conditions will remain difficult because of factors such as the large availability of foods containing a high amount of more than one macronutrient, the supplementation of many foods with artificial compounds that mimic the taste or texture of some macronutrients, and health campaigns toward healthier food habits.

This analysis of relevant literature emphasizes at least one implication regarding the prevention and treatment of obesity that pertains to the relevance of reducing dietary fat intake in the context of a physical activity program to favor a spontaneous energy deficit. In that respect, the literature described above suggests that the ability of physical activity per se to spontaneously alter macronutrient selection in the obese is uncertain. This thus implies that dietary advice is likely necessary to guide obese individuals toward good food selection instead of only relying on the impact of physical activity to induce such changes.

Back to Top | Article Outline


1. Ambler, C., A. Eliakim, J. A. Brasel, L. W. N. G. Burke, and D. M. Cooper. Fitness and the effect of exercise training on the dietary intake of healthy adolescents. Int. J. Obes. 22: 354–362, 1998.
2. Blair, S. N., N. M. Ellsworth, W. L. Haskell, et al. Comparison of nutrient intake in middle-aged men an women runners and controls. Med. Sci. Sports Exerc. 13: 310–315, 1981.
3. Butterworth, D. E., D. C. Nieman, R. Perkins, B. J. Warren, and R. G. Dotson. Exercise training and nutrient intake in elderly women. J. Am. Diet. Assoc. 93: 653–657, 1993.
4. de Wijn, J. F., J. Leusink, and G. B. Post. Diet, body composition and physical condition of champion rowers during periods of training and out of training. Biblio. Nutr. Diet. 27: 143–148, 1979.
5. Dionne, I., M. White, and A. Tremblay. Acute effects of exercise and low-fat diet on energy balance in heavy men. Int. J. Obes. 21: 413–416, 1997.
6. Drewnowski, A., and M. R. C. Greenwood. Cream and sugar: human preferences for high-fat foods. Physiol. Behav. 30: 629–633, 1983.
7. Even, P. C., N. Rieth, S. Roseau, and C. Larue-Achagiotis. Substrate oxidation during exercise in the rat cannot fully account for training-induced changes in macronutrients selection. Metabolism 47: 777–782, 1998.
8. Gerardo-Gettens, T., G. D. Miller, B. A. Horwitz, et al. Exercise decreases fat selection in female rats during weight cycling. Am. J. Physiol. 260: R518–R524, 1991.
9. Horio, T., and Y. Kawamura. Influence of physical exercise on human preferences for various taste solutions. Chemical Senses 23: 417–421, 1998.
10. Imbeault, P., S. Saint-Pierre, N. Almeras, and A. Tremblay. Acute effects of exercise on energy intake and feeding behaviour. Br. J. Nutr. 77: 511–521, 1997.
11. Katch, F. I., E. D. J. Michael, and E. M. Jones. Effects of physical training on the body composition and diet of females. Res. Q. 40: 99–104, 1969.
12. Keim, N. L., T. F. Barbieri, and A. Z. Belko. The effect of exercise on energy intake and body composition in overweight women. Int. J. Obes. 14: 335–346, 1990.
13. King, N. A., V. J. Burley, and J. E. Blundell. Exercise-induced suppression of appetite: effects on food intake and implications for energy balance. Eur. J. Clin. Nutr. 48: 715–724, 1994.
14. King, N. A., A. Tremblay, and J. E. Blundell. Effects of exercise on appetite control: implications for energy balance. Med. Sci. Sports Exerc. 29: 1076–1089, 1997.
15. Larue-Achagiotis, C., C. Martin, P. Verger, M. Chabert, and J. Louis-Sylvestre. Effects of acute treadmill exercise and delayed access to food on food selection in rats. Physiol. Behav. 53: 403–408, 1993.
16. Larue-Achagiotis, C., N. Rieth, and J. Louis-Sylvestre. Exercise training modifies nutrient self-selection in rats. Physiol. Behav. 56: 367–372, 1994.
17. Miller, G. D., A. G. Dimond, and J. S. Stern. Exercise reduces fat selection in female Sprague-Dawley rats. Med. Sci. Sports Exerc. 26: 1466–1472, 1994.
18. Oudot, F., C. Larue-Achagiotis, G. Anton, and P. Verger. Modification in dietary self-selection specifically attributable to voluntary wheel running and exercise training in the rat. Physiol. Behav. 59: 1123–1128, 1996.
19. Parizkova, J., and L. Stankova. Influence of physical activity on a treadmill on the metabolism of adipose tissue in rats. Br. J. Nutr. 18: 325–332, 1964.
20. Reggiani, E., S. Bertolini, G. Chiodini, et al. Effect of physical activity and diet on lipemic risk factors for atherosclerosis in women. Int. J. Med. 5: 183–186, 1984.
21. Smith, M. P., J. Mendez, M. Druckenmiller, and P. M. Kris-Etherton. Exercise intensity, dietary intake, and high-density lipoprotein cholesterol in young female competitive swimmers. Am. J. Clin. Nutr. 36: 251–255, 1982.
22. Thompson, P. D., B. Lazarus, E. Cullinane, et al. Exercise, diet, or physical characteristics as determinant of HDL-levels in endurance athletes. Atherosclerosis 46: 333–339, 1983.
23. Tremblay, A., N. Alméras, J. Boer, E. K. Kranenbarg, and J. P. Després. Diet composition and postexercise energy balance. Am. J. Clin. Nutr. 59: 975–979, 1994.
24. Verger, P., M. T. Lanteaume, and J. Louis-Sylvestre. Human intake and choice of foods at intervals after exercise. Appetite 18: 93–99, 1992.
25. Verger, P., M. T. Lanteaume, and J. Louis-Sylvestre. Free food choice after acute exercise in men. Appetite 22: 159–164, 1994.
26. Westerterp-Plantenga, M., C. R. Verwegen, M. J. Ijedema, N. E. Wijckmans, and W. H. Saris. Acute effects of exercise or sauna on appetite in obese and nonobese men. Physiol. Behav. 62: 1345–1354, 1997.


© 1999 Lippincott Williams & Wilkins, Inc.