Child Obesity Can Be Reduced With Vigorous Activity Rather Than Restriction of Energy Intake

Abstract

The global prevalence of juvenile obesity has increased to high levels, leading many governments and foundations to investigate what types of interventions may be able to prevent this problem. Such scientific investigations are designed to identify the key factors that need to be incorporated into public health efforts in schools and communities. Unfortunately, the results of the research on preventive intervention trials have been mixed, with many well-conceived trials failing to show clearcut favorable impact on the indices of adiposity used as outcomes (e.g., refs. 1,2,3). When so much effort and money is spent in trials that fail to provide evidence of success, the whole enterprise of obesity prevention is brought into question. Thus, it is important to examine carefully the assumptions on which the experimental trials are based. The typical paradigm used to view pediatric obesity is based on the commonsense knowledge derived from adult experience to the effect that reducing energy intake results in weight loss and that adding some physical activity (PA) to the lifestyle adds to the energy deficit and the resulting weight loss. A corollary of this paradigm is that youths who become fatter than their age mates ingest more energy than their age mates. However, recent evidence and theory suggest that this paradigm might be misleading as applied to pediatric obesity prevention. The necessity to view the problem of pediatric obesity prevention in a different way is illustrated by the results of a study recently reported by our group (4). This cross-sectional study, which investigated the relations among diet, PA, and body composition, is noteworthy for several reasons: (i) it had a relatively large sample size (661 adolescents); (ii) it employed high-quality measurements of percent body fat (%BF) with dual-energy X-ray absorptiometry (DXA) and visceral adipose tissue (VAT) with magnetic resonance imaging (MRI); and (iii) free-living diet and PA were measured with 4–7 separate 24-h recalls over a 2–3 month period. Special care was taken to train the subjects in the diet and PA recalls, using a multiple-pass approach so that accurate and complete data could be obtained, in light of the evidence of underreporting of energy intake in youths (5). Our main hypotheses, which were based on the energy balance paradigm, were that %BF would be highest in youths who ingested the most energy, especially in the form of fat, and who did the least moderate and vigorous PA. However, to our surprise we found that %BF was related to low levels of energy intake and low levels of vigorous, but not moderate, PA; i.e., those youths who did the most vigorous PA and ingested the most energy were the leanest. When we grouped the subjects into categories of vigorous PA, those who did no vigorous PA had a %BF of 28.6 and ingested 7296 kJ/day, and those who did at least 1 h/day of vigorous PA had a %BF of 19.4 and ingested 9217 kJ/day (Figure 1). When VAT was used as the outcome measure, energy intake was the sole and negative predictor of VAT; i.e., the youths who ingested the most energy had the least VAT. Other descriptive studies, both cross-sectional and longitudinal, have also found that youths who are relatively active tend to ingest more energy and accumulate less fat than inactive youths (6,7). Percent body fat for 661 adolescents in relation to hours per day of vigorous physical activity (PA), adjusted for age, race, and sex (P < 0.01 for group differences). After further adjustment for energy intake (EI), the influence of vigorous PA was still significant (P = 0.024). The mean energy intake (kJ/day) of each group is shown above the bars. Adapted from data in ref. 4. It is noteworthy that our study showed lower levels of %BF to be associated with greater amounts of vigorous PA, but not with moderate PA. Moderate PA includes activities such as walking, whereas vigorous PA includes sports, games, and dance activities that probably place more of a mechanical load on the musculoskeletal tissues of the body. We obtained similar results when we measured moderate and vigorous PA with 5 days of accelerometry in 421 youths (8); others who have examined PA intensity have also found that vigorous PA, to a greater degree than moderate PA, is linked with lower BF (9,10). These results suggest that growing youths need to be in positive energy balance in order to provide the nutrients needed for optimal growth. Some of the ingested nutrients will be partitioned into lean tissue (i.e., bone and muscle) and some into fat tissue; the balance of the nutrient partitioning will determine the relative fatness of the individual. Recent evidence emanating from studies of bone development have suggested that this balance is largely determined by the mechanical stimulation to the tissues, with larger amounts of stimulation leading stem cells to differentiate into bone and muscle rather than fat (11,12). Because vigorous PA provides mechanical stimulation to the tissues, it might be useful to consider pediatric obesity prevention as dependent on adequate doses of vigorous PA, rather than restriction of energy intake. This paradigm also emphasizes that we should use some index of body composition rather than a weight index such as BMI to evaluate the success of prevention programs. Vigorous PA can reduce fat mass at the same time that it increases fat-free mass (FFM), thereby improving body composition without necessarily reducing weight (13). There are also other reasons to use fatness rather than weight to assess the health of individuals. In adults, higher levels of fat mass are associated with greater mortality, whereas higher levels of FFM are associated with greater longevity (14). In youths, %BF tends to explain more of the variance in cardiometabolic risk factors than does BMI. For example, in a cohort of 464 adolescents %BF (as measured with DXA) explained 25.5% of the variance in fasting insulin, whereas BMI explained 17.7% of the variance, after adjusting for age, race, and sex (15). A related factor to consider is that cardiometabolic risk in youths is related to lower levels of PA and aerobic fitness as well as to elevated body fatness (16,17,18); i.e., it is the youth who is both obese and unfit/inactive who is at greatest risk. Thus, it may be sensible for future studies to focus on this unfit-kid (rather than merely obese-kid) syndrome. Because vigorous PA stimulates the development of bone and muscle, the best scenario for a youth wishing to develop a healthy body composition may be to engage in a relatively large amount of vigorous PA while ingesting sufficient energy and nutrients to support the tissue-building process. For example, there is evidence that PA and calcium intake interact in their relation to bone density in youths; i.e., high levels of PA are beneficial to bone density only if the youths also ingest sufficient amounts of calcium (19). Drawing causal conclusions from nonexperimental studies must be considered tentative until confirmed by experimental trials. For example, vigorous PA may influence %BF but it is also possible that %BF influences how much vigorous PA the youths can do. Indeed, it is likely that the relationship is reciprocal. To guide large-scale preventive interventions, we need to know what happens when we randomize youths to different groups, with PA as the main variable distinguishing the groups, and follow them over time. In studies where the youths were chosen to be obese at baseline, the projects can be construed as obesity treatment or secondary prevention, rather than primary prevention. Behavioral interventions focusing on both diet and PA have helped obese youths to reduce their relative weight (20). Thus, the energy balance paradigm and restriction of energy intake are appropriate to help obese youths reduce relative weight. However, many of these studies did not measure body composition, so it is not clear how much of the change in relative weight was related to changes in fat mass and how much due to changes in FFM. Moreover, the behavioral techniques did not control the PA programs, so it is unclear how much of the effect on weight was due to changes in diet and how much due to changes in PA. We have conducted a series of investigations to determine the effect of controlled vigorous PA, without restriction of energy intake, on body composition. In obese youths, we have found favorable effects on %BF, VAT, bone density, aerobic fitness, and some cardiometabolic risk factors, often accompanied by increases in dietary energy intake (21,22,23,24). It appears that an aerobic exercise prescription of 155–180 min/week at moderate-to-high intensity is effective in improving the body composition and fitness of overweight youths (25). To determine the value of PA to prevent fat accretion, it is necessary to study broader samples of youths who are not necessarily obese at baseline. Projects which have assessed the effectiveness of PA doses that have worked well for obese youths have generally found them to be ineffective for nonobese youths (e.g., refs. 26,27). Such results might tempt some health professionals to conclude that PA alone is insufficient to prevent obesity in nonobese youths. However, another possibility is that the PA dose needed to be effective in nonobese youths may be greater than the dose that is effective in obese youths. Support for this idea is provided by a project that exposed nonobese youths to 5 week of vigorous PA for 120–150 min/day during a summer camp, and found reductions in BF, as measured with magnetic resonance imaging, without weight change (28). Of course, 120–150 min/day would be difficult to manage within a typical day, but it does show what is possible, even in a relatively short intervention period of 5 week. Our research group implemented a preventive intervention with 8–11-year-old black girls who varied over the spectrum of fatness and fitness. We chose black girls as the target group because they are a subgroup that is at high risk for obesity; this strategy also eliminated any confounding effects of race and sex. We employed a PA dose considerably greater than that typically used in previous studies of youths; i.e., 80 min/day of mostly vigorous PA, offered 5 days/week for a 10-month period. The idea was to assure that the girls in the intervention group received a substantial dose of PA and to use a sample size that was sufficient to detect meaningful changes in the outcome variables (n = 201). The PA sessions took place after school in the same schools the girls attended during the school day; thus, the intervention could be incorporated into the regular schedules of the girls. We found favorable effects for %BF, BMI, VAT, bone density, and aerobic fitness. Within the intervention group, those children who maintained the highest heart rates during the exercise sessions and attended most often exhibited the greatest decreases in %BF and the greatest increases in bone density. For example, Figure 2 in the Barbeau publication (29) shows that, within the intervention group, girls who attended <2 sessions/week showed slight increases in %BF, whereas those who attended >4 days/week had a mean reduction of ∼2.3% BF. Using these results as a foundation, we then implemented a 3-year intervention protocol (the MCG FitKid Program) on a larger basis, starting with third graders; nine schools were randomized to receive the after-school intervention and nine schools served as controls (30). Results of the first year of intervention showed beneficial effects on %BF, bone density, and aerobic fitness (31). Within the intervention subjects, more benefit was derived by the children who attended the most after-school sessions (32). We have recently completed a preliminary analysis of the 3-year results (33). We found favorable effects on %BF and fitness during the school years when the children were exposed to the intervention. However, during the summers following the first and second years of intervention, the favorable effects on %BF and aerobic fitness were lost. This shows the importance of maintaining exposure to vigorous PA during vacation periods. Over the 3-year period, the intervention group increased more than the controls in FFM and BMI. Thus, if we had used BMI as our index of fatness, we would have derived misleading results. Our results are consistent with a recent review that concluded that the main factor distinguishing effective from ineffective pediatric obesity prevention trials was the provision of moderate to vigorous aerobic PA (34). Of course, vigorous PA is only one potentially effective part of the many efforts needed to reduce the epidemic of pediatric obesity. Other potential behavioral targets include time spent watching TV, sugared drink and sweet snack ingestion, diet composition, and dietary patterns. Once we have a better understanding of the behaviors that are basic to development of a healthy body composition, we will then need to alter the environments surrounding our youths, making them less obesogenic and more fitogenic. What are the practical implications of the idea that vigorous PA is one key to prevention of pediatric obesity and associated health problems? Every community should provide fitogenic environments for their children; i.e., environments that encourage vigorous PA and discourage TV viewing and unhealthy snacks, especially sugar-sweetened drinks that do not provide any nutrients with the energy. These environments should include active physical education classes and recess periods during school hours as well as programs for the after-school hours, weekends, and vacation periods. Although moderate activities like walking may be valuable for especially unfit youths starting an exercise program, it seems that vigorous activities such as running, basketball, swimming, and dance, which provide greater mechanical stimulation to the tissues, are more likely to help a youth develop a healthy body composition. Of course, it is necessary for an unfit and obese child to build up gradually from lower doses of volume and intensity up to higher doses, in order to avoid injury or undue fatigue. It is also important that the activities be age appropriate. For example, younger children prefer games that involve stop and go movements rather than exercising on machines at constant loads, whereas some adolescents may do well with the adult model of exercising on machines or jogging at a constant pace. These ideas are consistent with the recommendations of a recent consensus report that youths should have at least 60 min/day of age-appropriate and moderate-vigorous PA for juvenile obesity prevention (35). In previous generations, when the prevalence of pediatric obesity was relatively low, it was common for youths to obtain substantial amounts of vigorous PA simply by playing outside. However, fears for safety have led many present-day parents to reject this degree of freedom for their children. Thus, communities need to step forward and provide safe and supervised fitogenic environments, especially in low-socioeconomic status neighborhoods where obesity is most prevalent. The main message of this article is that a paradigm shift may help to improve the effectiveness of preventive intervention trials aimed at helping youths to attain a healthy body composition. Eating nutrient-rich diets is certainly appropriate for youths because of the need to ingest the nutrients needed for development of muscles and bones. However, trying to limit their energy intake may run counter to the biological demands of growth. Thus, intervention trials may be more effective if we increase the emphasis on vigorous PA and reduce the degree to which we advise growing youths to restrict their energy intake. If this hypothesis is supported by future controlled trials, the implication will be that large-scale preventive efforts in schools and communities should pay more attention to fostering vigorous PA than to restricting energy intake. The research of my group at the Medical College of Georgia was supported mainly by the National Institutes of Health. The author declared no conflict of interest.