Exercise or increasing physical activity is one of the cornerstones of pediatric obesity treatment, along with dietary and behavior change (11). Exercise can increase energy expenditure and create a negative energy balance, facilitating weight loss. In addition, exercise increases fitness, and exercise may have independent effects on many of the diseases often associated with obesity. This paper provides an evidence-based review (37) of controlled clinical research examining the utility of exercise and physical activity in the treatment of childhood and adolescent obesity with body composition, fitness, and coronary risk factors as outcomes. Recommendations for future research are presented.
Articles in which overweight or obese children and adolescents were placed on exercise programs for the purpose of weight loss were identified using computerized literature searches in several data bases, including Medline and PsychInfo, from January 1966 through November 1998. Thirteen studies met the following criteria for inclusion: 1) obese children or adolescents were provided either different types of exercise programs or an exercise program that was compared with a no-exercise control condition, 2) subjects were either randomly assigned to groups or matched on demographic and anthropometric variables, and 3) the exercise program was at least 2 months in duration. Table 1 presents subjects’ age, sample size, group assignment, and dietary and exercise components. Results for the most common body composition changes during treatment and follow-up are shown in Table 2, with changes in measures of fitness shown in Table 3. Treatment was operationalized as intervention continuing as long as subjects were seen at least once per month. Where available, baseline values, end of treatment changes, and end of follow-up changes have been provided. The preferred test of between group differences is the significant differences in rate of change between groups over time, but if this information was not available, within group differences from baseline to treatment and follow-up are presented.
Methodological Characteristics of the Studies
Random assignment or stratified random assignment was used in 12 of the 13 studies. The remaining study matched subjects to groups based on age, weight, height, and percent overweight (40). The studies were divided into three groups. The first group compared exercise versus no exercise controls, without a diet. The second group compared diet plus exercise versus diet alone controls. The third group compared different types of exercise programs.
There is considerable variability between studies in the range of subject ages, including subjects 7–13 yr of age (3,16,18–21,30,38), “prepubertal” participants (32), adolescents (22), and children and adolescents combined in the same samples (40,41). One study used males only (4), another female subjects (22), with the remaining studies using male and female subjects.
Definitions of obesity included: 20% above ideal weight for height (4,16,18–21,40), two standard deviations above weight for height standards (4), a body mass index (BMI) greater than the 85th percentile for sex and age (22), body weight that exceeded the 95th percentile for age (32), tricep skinfold greater than the 85th percentile for gender, age, and ethnicity (38), and the combination of elevated weight for height and skinfold measures (3,41).
Characteristics of exercise prescription.
Nine studies used aerobic exercise, with frequency varying from three to four times per week (3,20,41), five times per week (22,30,38), or daily exercise (40). The school-based studies provided two (4) and five (22) extra physical education classes per week emphasizing intensive exercise. Two studies contrasted aerobic versus lifestyle programs (18,20), one aerobic versus physical education classes (22), and one study targeting reductions in sedentary behavior (16). Seven studies incorporated caloric restrictions of various types (3,18–21,40,41). The content of the nutrition education classes used by Hills and Parker (32) was not described.
Evaluation of studies.
Because of the small number of randomized, controlled studies evaluating exercise interventions for obesity in pediatric samples, the effects of exercise were generally established by qualitative comparison of results. Quantitative meta-analytic methods were used to evaluate the six studies that compared the short-term effects of diet plus exercise versus diet alone. Effect sizes were established using the formula Meandiet + exercise − Meandiet/Standard Deviationdiet.
The results will be discussed according to the type of experimental design.
Exercise Versus No Exercise Controls
In two studies Gutin et al. (30,38) found significant reductions in body fat and increases in fitness compared with controls, whereas no significant effects of exercise on body composition or fitness were found by Blomquist et al. (4).
Diet Versus Diet and Exercise Versus No Intervention
Three studies demonstrated significantly greater changes in body composition for the diet plus exercise group compared with the diet only group at post-treatment (21,32,40), while the other three studies did not find significant differences between groups (3,19,41). The mean effect size for the six studies, based on one study using weight as the dependent variable (32), three using percent overweight (19,21,40), and two using percent body fat (3,41), was 0.45. Two studies included follow-up assessments, with one showing significant additive effects of exercise at 12 months (21), a second showing no significant differences between groups at 10 yr (17,19). Two studies showed diet and exercise to be more effective than diet alone in increasing fitness (Table 3) as measured by physical work capacity (21) and submaximal oxygen consumption (41), with one study finding no differences between groups (3). In addition to fitness effects, two randomized, controlled studies examined the effects of exercise on coronary risk factors. In these studies diet plus exercise provided greater beneficial effects on high density lipoprotein (HDL) (3) and systolic blood pressure (41) compared with diet alone and control.
Types of Exercise Programs
Epstein et al. (18,20) conducted two studies that compared the effects of lifestyle activity versus traditional programmed aerobic exercise. In the first comparison (18) lifestyle activity showed larger reductions in percent overweight at 6 and 17 months than aerobic exercise. No differences in fitness were observed at 6 months, as the aerobic group showed larger initial changes, but the lifestyle program showed better maintenance. Epstein et al. (20)compared lifestyle and aerobic exercise programs with a low intensity calisthenic group to control for the nonspecific effects of exercise. At 6- and 12-month assessments all groups showed similar reductions in percent overweight and weight. The lifestyle group maintained the decreases in percent overweight at 2 yr, while the aerobic and calisthenic exercise groups showed increases in percent overweight. At 5 yr the lifestyle exercise group had lower percent overweight than the calisthenic group, while at 10 yr both lifestyle and aerobic groups had lower percent overweight and weight than the calisthenic group (17). Programmed exercise demonstrated larger improvements in fitness than lifestyle activity at 12 months.
Based on the observation that sedentary behaviors are related to the development of childhood obesity (9,28), Epstein et al. (16) compared groups that were differentially reinforced for increased physical activity and/or reduction in sedentary behaviors. Reducing sedentary behaviors proved more successful than increasing physical activity at both the end of treatment and 1-yr follow-up. All groups showed equivalent increases in fitness.
Finally, Ewart et al. (22) compared the effects of programmed aerobic exercise or physical education classes on BMI of adolescent girls at high risk for hypertension. No differences between exercise groups were found at the end of the 18-wk intervention. The aerobic exercise intervention showed significant improvements in fitness and systolic blood pressure compared with the physical education group, while both groups showed significant reductions in diastolic blood pressure.
This review highlights the limited number of randomized controlled studies investigating the efficacy of exercise in the treatment of pediatric obesity. The small number of controlled outcome studies in combination with the use of different dependent variables across studies limits the strength of conclusions that can be drawn. The only area in which there are sufficient studies to make quantitative data-based conclusions for body composition changes is in the comparison of diet versus diet plus exercise. These results indicate diet plus exercise improve short-term obesity treatment by almost one-half standard deviation above and beyond the contribution made by diet alone (Evidence Category B). Exercise reliably increases fitness effects whether compared with diet alone or no exercise controls (Evidence Category B). There are too few studies comparing exercise versus no exercise, or comparing different exercise programs to make firm conclusions. In addition, there are very limited data on long-term effects of exercise, and any conclusions would be premature.
Future Research Directions
A very important aspect of exercise intervention research is adherence. Failure to adhere to the exercise program limits short- and long-term effectiveness of the intervention and makes accurate evaluation impossible. Adherence may be affected by characteristics of the exercise program, such as intensity and duration, but also may relate to the structure of the exercise program. It is tempting to use on-site exercise interventions to facilitate measurement of adherence and better control intensity and duration of the exercise. However, research with adults has shown superior long-term effects for home-based, rather than exercise site-based, programs (33,39). If this effect generalizes to pediatric programs, it becomes extremely important to balance experimental control over the characteristics of the exercise program versus superior long-term effectiveness of programs that maximize choice.
Improving adherence to physical activity in obese pediatric samples may be very challenging, as obesity predicts poor adherence to exercise programs (10) and environmental contingencies need to be stronger for obese than nonobese children to choose active behaviors over sedentary ones (13,14). One way to improve adherence to physical activity programs may be to build activity into the child or adolescent’s lifestyle, which has shown to maximize short- and long-term adherence in obese children (18,20). Compared with aerobic exercise, lifestyle activity programs maximize choice and perceived control over exercise behavior, both of which have been shown to be important determinants of exercise adherence (48).
Another way to increase adherence would be to increase the reinforcing value of being active or reduce access to sedentary behaviors that are powerful reinforcers for obese children and compete with being active. Identifying methods of enhancing the reinforcing value of exercise would be of great importance as it would provide putative benefits to adherence, weight control, and fitness, thus enhancing the overall effect of exercise as a means of treating and preventing obesity. Children with sedentary lifestyles commonly choose sedentary behaviors when given the choice between being sedentary or active, and research suggests limiting access to sedentary behaviors can result in increased physical activity (12–14). In addition, sedentary behaviors can be used to reinforce obese children for being more physically active (43).
One effect of combining diet and exercise is to increase negative energy balance and theoretically improve weight loss over diet alone. In addition, diet may influence adherence or performance to physical activity and thus weight loss. For example, basic animal research suggests that caloric reductions lead to spontaneous increases in activity, although these increases may only be limited to nonobese animals (46). There are suggestive data that controlling for calories, a greater carbohydrate versus fat intake will lead to significant differences in energy expenditure (2). In addition, intake of foods with a low glycemic index may improve performance and lower perception of workloads, making exercise more pleasant (8). It would be potentially very important to determine whether variations in total calories or the macro- or micro-nutrient composition of the diet influenced activity levels in children, and then diets could be used to enhance physical activity in obese children.
There may be individual differences that influence the effects of exercise on pediatric obesity. Research suggests that age is related to activity level, the types of activities engaged in, and equally important, types of activities enjoyed (44). Interventions may need to be different across developmental stages. An example of one type of exercise that may depend on age is resistance training. Recent data suggests that resistance training can prevent the gain of additional adiposity in adolescents (50), but safety can be an issue when using resistance training in prepubescent children (1).
A variety of other individual difference variables may influence physical activity in the treatment of obesity, including gender, degree of obesity, and comorbidities. Boys and girls are socialized differently regarding exercise, have different exercise skills, may prefer different exercise activities (44), and girls show greater increases in adiposity during puberty than boys (36). Programs may need to be tailored to gender differences. In addition, obesity and low initial levels of fitness may attenuate the positive consequences associated with being active, reducing the motivation to maintain an exercise program. Likewise, a variety of comorbid conditions may influence exercise programs. For example, there is an increased prevalence of obesity in children and adolescents with asthma (26,34), and the asthma may interfere with participation in physical activity, which would make it more difficult for obese children with asthma to lose and maintain weight loss.
Ethnic and minority subjects.
Few studies have included children or adolescents from ethnic or minority backgrounds in exercise intervention studies, despite data indicating these persons tend to be less physically active (24,52) and are more likely to be obese than nonminority children (27,42). There may be important characteristics of ethnic or minority groups that can influence exercise patterns. For example, the environments in which minority children live may be more dangerous than environments in which nonminority families live, reducing access of minority children to outdoor activities. In addition, exercise may be less valued and socially accepted in certain minority environments compared with people with nonminority status. Research with children from ethnic and minority backgrounds is needed to determine whether the effects of exercise on pediatric obesity generalize to these groups, and the potential barriers to implementing exercise programs need to be identified.
Physical and psychological comorbidity.
Exercise may be particularly useful for treating the comorbid conditions often associated with obesity. Obesity is often associated with cardiovascular risk factors, and exercise improves the treatment effect for obese children with elevations in lipids (3) and blood pressure (41). These improvements in comorbid conditions may be mediated by exercise influenced weight loss, improvements in fitness produced by exercise, or a combination of weight loss and fitness change.
Obese children and adolescents seeking treatment often experience psycho-social comorbidity (6). Exercise can reduce psychopathology in adults (7,47), as well as enhance mood in normal weight adults (23) and normal weight children and adolescents (5), but there is no controlled research examining these factors in obese children and adolescents. Improvements in weight as well as fitness changes may be related to improvement in psychological comorbidity in obese children and adolescents.
Characteristics of exercise programs.
The majority of exercise programs have focused on aerobic exercise. However, there has been very little research on the best way to implement aerobic exercise programs to maximize weight loss and adherence. The best schedule for increasing intensity or duration of aerobic activity needs to be identified. Research in young, normal-weight men indicates that isocaloric low intensity, long duration exercise results in greater total fat oxidation than moderate intensity exercise (49). These results underscore the importance of investigating the effects of intensity of exercise on weight loss in obese children. Several other important issues need to be empirically studied. For example, can resistance training be used safely in a pediatric population to maximize the development of lean body mass and increase total energy expenditure? Is it better to focus on one activity or is cross training better? Can the combination of aerobic and resistance training enhance the effects of aerobic exercise?
The structure of the program may be important for developing active lifestyles in treating obesity. Data from several trials incorporating moderate to intense aerobic exercise with children in the United States (29,30) and Japan (31,45) suggest that school-based exercise interventions may be a promising approach to treating childhood and adolescent obesity. It is important to investigate further the efficacy of school-based exercise programs since they provide an opportunity to develop healthy, active lifestyles in large numbers of children and adolescents (51). In addition, use of the family to structure and support activity programs may be useful for long-term change, as parent activity level is a strong predictor of child activity (25,35), and inclusion of parents in the treatment may improve long-term outcomes of obesity treatment (15).
There are limited controlled data on physical activity in pediatric obesity treatment. The initial research is encouraging, but more research is needed to maximize the effects of physical activity on obese children. Investigators need to take into account subject characteristics, characteristics of the exercise programs, and the outcomes to be achieved to develop the most effective exercise programs. The development of an active lifestyle in obese children has the potential for multiple benefits on obesity, comorbid physical and psychological problems, and acquisition of an active lifestyle that may accrue lifelong health benefits.
This paper was supported in part by grants HD RO1 23713, HD RO1 25997 and HD RO1 20829 awarded to L. H. Epstein.
1. American Academy of Pediatrics. Strength training, weight and power lifting, and body building by children
. Pediatrics 86: 801–803, 1999.
2. Bandini, L. G., D. A. Schoeller, and W. H. Dietz. Metabolic differences in response to a high-fat vs a high-carbohydrate diet. Obes. Res. 2: 348–354, 1994.
3. Becque, M. D., V. L. Katch, A. P. Rocchini, C. R. Marks, and C. Moorehead. Coronary risk incidence of obese adolescents
: reduction by exercise
plus diet intervention. Pediatrics 81: 605–612, 1988.
4. Blomquist, B., M. Borjeson, Y. Larsson, B. Persson, and G. Sterky. The effect of physical activity
on the body measurements and work capacity of overweight boys. Acta Paediatr. Scand. 54: 566–572, 1965.
5. Boyd, K. R. and D. W. Hrycaiko. The effect of a physical activity
intervention package on the self-esteem of pre-adolescent and adolescent females. Adolescence 32: 693–708, 1997.
6. Braet, C., I. Mervielde, and W. Vandereycken. Psychological aspects of childhood obesity
: a controlled study in a clinical and nonclinical sample. J. Pediatr. Psychol. 22: 59–71, 1997.
7. Brown, D. R. Exercise
, fitness, and mental health. In: Exercise
, Fitness, and Health, C. Bouchard, R. J. Shephard, T. Stephens, J. R. Sutton, and B. D. McPherson (Eds.). Champaign, IL: Human Kinetics, 1990, pp. 607–633.
8. Demarco, H. M., K. P. Sucher, C. J. Cisar, and G. E. Butterfield. Pre-exercise
carbohydrate meals: application of glycemic index. Med. Sci. Sports Exerc. 31: 164–170, 1999.
9. Dietz, W. H. and S. L. Gortmaker. Do we fatten our children
at the television set? Obesity
and television viewing in children
. Pediatrics 75: 807–812, 1985.
10. Dishman, R. K. Determinants of participation in physical activity
. In: Exercise
, Fitness and Health, C. Bouchard, R. J. Shepard, T. Stephens, J. R. Sutton, and B. D. McPherson (Eds.). Champaign, IL.: Human Kinetics, 1990, pp. 75–101.
11. Epstein, L. H. Exercise
in the treatment of childhood obesity
. Int. J. Obes. 19: S117-S121, 1995.
12. Epstein, L. H., B. E. Saelens, M. D. Myers, and D. Vito. Effects of decreasing sedentary behaviors on activity
choice in obese children
. Health Psychol. 16: 107–113, 1997.
13. Epstein, L. H., B. E. Saelens, and J. G. O’Brien. Effects of reinforcing increases in active behavior versus decreases in sedentary behavior for obese children
. Int. J. Behav. Med. 2: 41–50, 1995.
14. Epstein, L. H., J. A. Smith, L. S. Vara, and J. S. Rodefer. Behavioral economic analysis of activity
choice in obese children
. Health Psychol. 10: 311–316, 1991.
15. Epstein, L. H., A. Valoski, R. R. Wing, and J. McCurley. Ten-year follow-up of behavioral family-based treatment for obese children
. JAMA 264: 2519–2523, 1990.
16. Epstein, L. H., A. M. Valoski, L. S. Vara, et al. Effects of decreasing sedentary behavior and increasing activity
on weight change in obese children
. Health Psychol. 14: 109–115, 1995.
17. Epstein, L. H., A. M. Valoski, R. R. Wing, and J. McCurley. Ten-year outcomes of behavioral family-based treatment for childhood obesity
. Health Psychol. 13: 373–383, 1994.
18. Epstein, L. H., R. R. Wing, R. Koeske, D. J. Ossip, and S. Beck. A comparison of lifestyle change and programmed aerobic exercise
on weight and fitness changes in obese children
. Behav. Ther. 13: 651–665, 1982.
19. Epstein, L. H., R. R. Wing, R. Koeske, and A. Valoski. Effects of diet plus exercise
on weight change in parents and children
. J. Consult. Clin. Psychol. 52: 429–437, 1984.
20. Epstein, L. H., R. R. Wing, R. Koeske, and A. Valoski. A comparison of lifestyle exercise
, aerobic exercise
, and calisthenics on weight loss
in obese children
. Behav. Ther. 16: 345–356, 1985.
21. Epstein, L. H., R. R. Wing, B. C. Penner, and M. J. Kress. Effect of diet and controlled exercise
on weight loss
in obese children
. J. Pediatr. 107: 358–361, 1985.
22. Ewart, C. K., D. R. Young, and J. M. Hagberg. Effects of school-based aerobic exercise
on blood pressure in adolescent girls at risk for hypertension. Am. J. Public Health 88: 949–951, 1998.
23. Folkins, C. H. and W. E. Sime. Physical fitness training and mental health. Am. Psychol. 36: 373–389, 1981.
24. Fontvieille, A. M., A. Kriska, and E. Ravussin. Decreased physical activity
in Pima Indian compared with Caucasian children
. Int. J. Obes. 17: 445–452, 1993.
25. Freedson, P. S. and S. Evenson. Familial aggregation in physical activity
. Res. Q. Exerc. Sport 62: 384–389, 1991.
26. Gennuso, J., L. Epstein, R. Paluch, and F. Cerny. The relationship between asthma and obesity
in urban minority children
. Arch. Pediatr. Adolesc. Med. 152: 1197–2000, 1998.
27. Gortmaker, S. L., W. H. Dietz, A. M. Sobol, and C. A. Wehler. Increasing pediatric obesity
in the United States. Am. J. Dis. Child. 141: 535–540, 1987.
28. Gortmaker, S. L., A. Must, A. M. Sobol, K. Peterson, G. A. Colditz, and W. H. Dietz. Television watching as a cause of increasing obesity
in the United States, 1986–1990. Arch. Pediatr. Adolesc. Med. 150: 356–362, 1996.
29. Gutin, B. and N. Cucuzzo. Physical training improves body composition of black obese 7-to 11-year-old girls. Obes. Res. 3: 305–312, 1995.
30. Gutin, B., S. Owens, G. Slavens, S. Riggs, and F. Treiber. Effect of physical training on heart-period variability in obese children
. J. Pediatr. 130: 938–943, 1997.
31. Hayashi, T., M. Fujino, M. Shindo, T. Hiroki, and K. Arakawa. Echocardiographic and electrocardiographic measures in obese children
after an exercise
program. Int. J. Obes. 11: 465–472, 1987.
32. Hills, A. P. and A. W. Parker. Obesity
management via diet and exercise
intervention. Child Care Health Dev. 14: 409–416, 1988.
33. King, A. C., W. L. Haskell, C. B. Taylor, H. C. Kraemer, and R. F. Debusk. Group- vs home-based exercise
training in healthy older men and women: a community-based clinical trial. JAMA 266: 1535–1542, 1991.
34. Luder, E., T. A. Melnick, and M. Dimaio. Association of being overweight with greater asthma symptoms in inner city black and Hispanic children
. J. Pediatr. 132: 699–703, 1998.
35. Moore, L. L., D. A. Lombardi, M. J. White, J. L. Campbell, S. A. Oliveria, and R. C. Ellison. Influence of parents’ physical activity
levels on activity
levels of young children
. J. Pediatr. 118: 215–219, 1991.
36. Mueller, W. H. The changes with age of the anatomical distribution of fat. Social Sci. Med. 16: 191–196, 1982.
37. National Heart Lung Blood Institute Obesity
Education Initiative Expert Panel. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity
in Adults: The Evidence Report. Obes. Res. 6(Suppl. 2): S51-S209, 1998.
38. Owens, S., B. Gutin, J. Allison, et al. Effect of physical training on total and visceral fat in obese children
. Med. Sci. Sports 31: 143–148, 1999.
39. Perri, M. G., A. D. Martin, E. A. Leermakers, S. F. Sears, and M. Notelovitz. Effects of group- versus home-based exercise
in the treatment of obesity
. J. Consult. Clin. Psychol. 65: 278–285, 1997.
40. Reybrouck, T., J. Vinckx, G. Van Den Berghe, and M. Vanderschueren-Lodeweyckx. Exercise
therapy and hypocaloric diet in the treatment of obese children
. Acta Paediatr. Scand. 79: 84–89, 1990.
41. Rocchini, A. P., V. Katch, J. Anderson, et al. Blood pressure in obese adolescents
: effect of weight loss
. Pediatrics 82: 16–23, 1988.
42. Ryan, A. S., G. A. Martinez, R. N. Baumgartner, et al. Median skinfold thickness distribution and fat wave patterns in Mexican-American children
from the Hispanic Health and Nutrition Examination (HHANES 1982–1984). Am. J. Clin. Nutr. 51: 925S-935S, 1990.
43. Saelens, B. E. and L. H. Epstein. Behavioral engineering of activity
choice in obese children
. Int. J. Obes. 22: 275–277, 1998.
44. Sallis, J. F., B. G. Simons-Morton, E. J. Stone, et al. Determinants of physical activity
and interventions in youth. Med. Sci. Sports Exerc. 24: S248-S257, 1992.
45. Sasaki, J., M. Shindo, H. Tanaka, M. Ando, and K. Arakawa. A long-term aerobic exercise
program decreases the obesity
index and increases the HDL cholesterol concentration in obese children
. Int. J. Obes. 11: 339–345, 1987.
46. Sclafani, A. and A. Rendel. Food deprivation-induced activity
in dietary obese, dietary lean, and normal-weight rats. Behav. Biol. 24: 220–228, 1978.
47. Simons, A. D., C. R. McGowan, L. H. Epstein, D. J. Kupfer, and R. J. Robertson. Exercise
as a treatment for depression: an update. Clin. Psychol. Rev. 5: 553–568, 1985.
48. Thompson, C. E. and L. M. Wankel. The effects of perceived activity
choice upon frequency of exercise
behavior. J. Appl. Soc. Psychol. 10: 436–443, 1980.
49. Thompson, D. L., K. M. Townsend, R. Boughey, K. Patterson, and D. R. Bassett. Substrate use during and following moderate- and low-intensity exercise
: implications for weight control. Eur. J. Appl. Physiol. Occup. Physiol. 83: 1625–1627, 1998.
50. Treuth, M. S., G. R. Hunter, R. Figueroa-Colon, and M. I. Goran. Effects of strength training on intra-abdominal adipose tissue in obese prepubertal girls. Med. Sci. Sports Exerc. 30: 1738–1743, 1998.
51. Trudeau, F., L. Laurencelle, J. Tremblay, M. Rajic, and R. J. Shephard. Daily primary school physical education: effects on physical activity
during adult life. Med. Sci. Sports 31: 111–117, 1999.
52. Wolf, A. M., S. L. Gortmaker, L. Cheung, H. M. Gray, D. B. Herzog, and G. A. Colditz. Activity
, inactivity, and obesity
: racial, ethnic, and age differences among schoolgirls. Am. J. Public Health 83: 1625–1627, 1993.