search text   Advanced Search

This Article Abstract Full Text (PDF) P3Rs: Submit a response Alert me when this article is cited Alert me when P3Rs are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Spear, B. A. Articles by Taveras, E. M. Search for Related Content PubMed PubMed Citation Articles by Spear, B. A. Articles by Taveras, E. M. Related Collections Nutrition & Metabolism

Recommendations for Treatment of Child and Adolescent Overweight and Obesity

^a Department of Pediatrics, School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama
^b Department of Pediatrics, Saint Louis University, St Louis, Missouri
^c Georgia Diabetes Coalition, Atlanta, Georgia
^d Obesity Program, Division of Endocrinology, Children's Hospital Boston, Harvard Medical School, Boston, Massachusetts
^e Departments of Pediatrics and Psychiatry and Behavioral Sciences, University of Washington–Child Health Institute, Seattle, Washington
^f Department of Pediatrics, James H. Quillen College of Medicine, East Tennessee State University, Johnson City, Tennessee
^g Obesity Prevention Program, Department of Ambulatory Care and Prevention, Harvard Pilgrim Health Care and Harvard Medical School, Boston, Massachusetts
^h Division of General Pediatrics, Children's Hospital Boston, Boston, Massachusetts

	ABSTRACT

TOP ABSTRACT NUTRITIONAL TREATMENT FOOD BEHAVIORS DIETARY INTERVENTIONS PHYSICAL ACTIVITY TELEVISION VIEWING AND MEDIA... BEHAVIORAL APPROACHES OTHER INTERVENTIONS RECOMMENDATIONS REFERENCES

 
In this article, we review evidence about the treatment of obesity that may have applications in primary care, community, and tertiary care settings. We examine current information about eating behaviors, physical activity behaviors, and sedentary behaviors that may affect weight in children and adolescents. We also review studies of multidisciplinary behavior-based obesity treatment programs and information about more aggressive forms of treatment. The writing group has drawn from the available evidence to propose a comprehensive 4-step or staged-care approach for weight management that includes the following stages: (1) Prevention Plus; (2) structured weight management; (3) comprehensive multidisciplinary intervention; and (4) tertiary care intervention. We suggest that providers encourage healthy behaviors while using techniques to motivate patients and families, and interventions should be tailored to the individual child and family. Although more intense treatment stages will generally occur outside the typical office setting, offices can implement less intense intervention strategies. We not ony address specific patient behavior goals but also encourage practices to modify office systems to streamline office-based care and to prepare to coordinate with professionals and programs outside the office for more intensive interventions.

Key Words: obesity • treatment

Abbreviations: GI—glycemic index • PSMF—protein-sparing modified fast • CDC—Centers for Disease Control and Prevention • FDA—Food and Drug Administration • CE—consistent evidence • ME—mixed evidence

Treatment for children who are overweight or obese seems easy, that is, just counsel children and their families to eat less and to exercise more. In practice, however, treatment of childhood obesity is time-consuming, frustrating, difficult, and expensive. In fact, choosing the most effective methods for treating overweight and obesity in children is complex at best. This is especially true for primary care providers, who have limited resources to offer interventions within their offices or programs and few providers to whom they can refer patients.

The need for evidence-based treatment recommendations is a critical health care issue, because obese children and adolescents are at risk for developing many of the comorbidities seen in obese adults. Studies demonstrated that fasting serum glucose, insulin, and triglyceride levels and the prevalence of impaired glucose tolerance and systolic hypertension increase significantly as children become obese (BMI of 95th percentile).¹ Even children and adolescents who are overweight (BMI of 85th to 94th percentile) are at risk for comorbidities. Therefore, interventions using dietary modifications, increased physical activity, and behavioral therapy may be beneficial for overweight children and adolescents, with more-aggressive intervention directed toward obese children and adolescents.²

Health care professionals, however, may find it difficult to determine which interventions will be most efficacious for their patients. To date, no clinical trials have determined whether specific dietary modifications alone (ie, without behavioral interventions and increased physical activity) are effective in reducing childhood overweight and obesity rates. Comprehensive interventions that include behavioral therapy along with changes in nutrition and physical activity are the most closely studied and seem to be the most successful approaches to improving long-term weight and health status.^3,4 However, the clinical trials testing these interventions often are limited in their ability to determine the relative efficacy of individual strategies. Ultimately, children and adolescents (and adults, for that matter) become overweight or obese because of an imbalance between energy intake and expenditure. Dietary patterns, television viewing and other sedentary activities, and an overall lack of physical activity play key roles in creating this imbalance and therefore represent opportunities for intervention.

This report reviews evidence about the treatment of obesity that may have application in the primary care setting. It examines current information about eating behaviors, physical activity behaviors, and sedentary behaviors that may affect weight gain. Many of the studies are correlational, rather than interventional. Also examined are studies of multidisciplinary, behavior-based, obesity treatment programs and information about more-aggressive forms of treatment, such as bariatric surgery. Reviews are followed by evidence-based treatment recommendations.

Studies of obesity treatment in the primary care setting have not been conducted. To provide guidance on obesity treatment to providers, the treatment writing group has drawn from the available evidence to propose a comprehensive approach (as yet untested) that is reasonable, feasible, and flexible. This report suggests that providers encourage healthy behaviors, use techniques to motivate patients and families, establish office systems that support monitoring and care of these children, and implement a staged approach to intervention that is tailored to the individual child and family.

	NUTRITIONAL TREATMENT

TOP ABSTRACT NUTRITIONAL TREATMENT FOOD BEHAVIORS DIETARY INTERVENTIONS PHYSICAL ACTIVITY TELEVISION VIEWING AND MEDIA... BEHAVIORAL APPROACHES OTHER INTERVENTIONS RECOMMENDATIONS REFERENCES

 
Data Limitations
Virtually no clinical trials examining the effects of any specific dietary prescription on body weight or adiposity in children control for the effects of potentially confounding factors, such as treatment intensity, behavioral intervention strategies, and physical activity. Although comprehensive approaches aiming to modify diet, physical activity, family behavior, and the social and physical environment are undoubtedly needed, studies involving multiple modalities cannot assess the efficacy of any specific component (eg, diet). In the absence of data on the relative efficacy of various dietary prescriptions in the treatment of obesity in children, it is sometimes necessary to make inferences from the childhood obesity prevention and adult treatment literature.

Food Groups and Childhood Overweight
Fruits and Vegetables
Eight studies evaluating the relationship between fruit and/or vegetable intake and body weight were reviewed; none was longitudinal. A nationally representative study found an association between lower intake of fruits and overweight in both boys and girls and an association between lower intake of vegetables and overweight in boys only.⁵

Evidence from case-control studies that evaluated the intake of fruits and adiposity was mixed. Two studies found an inverse association with adiposity,^6,7 and 3 found no association.^8–10 All of the studies that evaluated the intake of vegetables found no relationship with adiposity.^5–12 The single study that evaluated the intake of fruits and vegetables combined found an inverse relationship with adiposity.¹² The studies that found a significant relationship with fruit or vegetable intake tended to have larger sample sizes than did those that found no relationship.^6,12

School-based interventions have increased fruit and vegetable consumption, but the effect of these dietary changes on weight or weight loss has not been evaluated. School-based studies frequently combine increased fruit and vegetable intake with decreased fat intake, which makes it difficult to comment on the association between fruit and vegetable intake and weight. It should be noted, however, that in none of the studies reviewed was increased fruit and vegetable intake related to increased adiposity. The evidence was more compelling for fruits alone or for fruits and vegetables combined than for vegetables alone, possibly because different fruits and vegetables have differing effects on children's weight. Some of the most commonly consumed vegetables are relatively high in energy because of the way they are prepared. For example, more than one third of the total vegetable intake in the United States consists of iceberg lettuce, frozen potatoes (usually French fries), and potato chips. On balance, the evidence indicates that greater fruit and vegetable intake may provide modest protection against increased adiposity.¹³ Research indicates that children are least likely to consume adequate amounts of foods from the fruit and vegetable groups, compared with other food groups.¹⁴

Fruit Juice
Intake of 100% fruit juice does not seem to be related to childhood obesity unless it is consumed in large quantities. Of the 10 articles reviewed, 3 found a positive association between consumption of large amounts of 100% juice (>12 fl oz/day) and increased incidence of overweight and 1 found a positive association with BMI of >95th percentile.¹⁵ However, none of the longitudinal studies^16–18 or the nationally representative studies^19,20 reported any relationship between 100% fruit juice consumption and BMI.

In a small case-control study of 7- to 10-year-old children, obese children consumed greater amounts of 100% fruit juice than did nonobese control subjects.¹⁰ One limitation of that study, however, beside small sample size, was the fact that the beverages reported as fruit juice on a food frequency questionnaire might have included artificially flavored drinks containing little or no fruit juice. This does not seem to have been a weakness of other studies.

In a cross-sectional study of 2- to 5-year-old children, those (n = 19) who consumed >12 oz of 100% fruit juice per day were at increased risk of short stature and overweight.²¹ Additional analysis of the same study population found that only apple juice was significantly related to BMI. Welsh et al¹⁵ found that fruit juice consumption among children 2 to 3 years of age with a BMI of 95th percentile was associated with continued obesity 1 year later. There was no significant difference in children with BMI of <95th percentile 1 year later. In that study, fruit juice was defined as vitamin C-containing juice (orange juice or juice with vitamin C added).

Skinner et al¹⁸ monitored children longitudinally from 24 to 72 months of age and found no relationship between 100% juice intake and anthropometric measurements. However, one criticism of that study was that only 3 children consumed 12 oz of juice per day over time and only 1 of those reported 12 oz/day at all 7 dietary interviews. When fruit juice consumption was examined as a continuous variable, there still was no significant association between intake and BMI. In fact, children with a higher intake of fruit juice were more likely to have a lower Ponderal index (an indicator of weight status analogous to BMI but calculated as weight divided by height to the third power).

The 1994 Continuing Survey of Food Intakes by Individuals data on preschool-aged children who reported intake of >12 oz of 100% juice daily found no relationship between fruit juice consumption and BMI.²⁰ Similarly, a study of preschool-aged children enrolled in the Supplemental Nutrition Program for Women, Infants, and Children program, 79% of whom reportedly consumed >12 oz of fruit juice daily, found no relationship between 100% fruit juice consumption and BMI.²² A study of preschool-aged children in Germany found no association between excessive consumption of fruit juice and BMI.¹⁶ Similar data for adolescents are lacking, but data suggest that fruit juice consumption declines as children mature.

The American Academy of Pediatrics recently recommended that fruit juice consumption be limited to 4 to 5 oz/day for children 1 to 6 years of age and 8 to 12 oz/day for children 7 to 18 years of age.²³ Those recommendations, however, were based on considerations of nutrient and gastrointestinal problems. More research was deemed necessary before overweight could be considered a consequence of excess fruit juice consumption. The US Department of Agriculture has stressed the important contribution to nutrient intake of 100% fruit juices and advises that, when consumed in quantities consistent with the Dietary Guidelines for Americans, fruit juice is advantageous for healthy children.¹³

Sweetened Beverages
Intake of soft drinks and sweetened fruit drinks has increased dramatically among US children, particularly among adolescents, in recent decades. According to a national survey, soft drinks were the sixth leading food source of energy among children, constituting >50% of total beverage consumption and representing the primary source of energy intake for US adolescents.²⁴ Although there is no clear evidence that consumption of sugar per se affects food intake and weight gain, there is evidence to suggest that "liquid sweets," or energy consumed as a liquid, may be less well regulated by the body than energy consumed in a solid form. Furthermore, several studies suggest that consumption of soft drinks and other sweetened beverages is related to increased energy intake.

Of the 19 studies reviewed, 6 were longitudinal studies, 3 were nationally representative, cross-sectional studies, and 10 were case-control studies or other cross-sectional studies.^{8,10,12,19,25–39} Although the evidence is mixed, the larger, more strongly designed, and higher-quality studies substantiated the idea that sweetened beverage intake is related to overweight among children. Of the 6 longitudinal studies, 3 found intake of soda or total sweetened beverages to be associated positively with at least 1 measure of adiposity, whereas 3 found no significant associations. A large, nationally representative study by Troiano et al³⁹ that measured height and weight directly found a positive association between energy from soda and overweight. Two smaller national studies by Forshee and Storey,¹⁹ which found no such association, were based on reported heights and weights.

In a nationally representative sample of 2- to 19-year-old youths,³⁹ soft drink intake was greater among overweight youths than among nonoverweight youths in all age groups. Furthermore, the Growing Up Today Study,²⁶ a 1.5-year longitudinal study of children 9 to 14 years of age, found that high levels of consumption of sweetened beverages at baseline were associated with increased BMI.

In a recently published, randomized, controlled trial conducted among 103 high school students who regularly consumed sugar-sweetened beverages, students were assigned to either an experimental group that received home deliveries of noncaloric beverages or a control group that received no intervention. After 6 months, responses to the intervention were associated inversely with baseline BMI values. Among the heaviest one third of the cohort, BMI was significantly lower in the experimental group, compared with the control group (–0.75 ± 0.34 kg/m²).⁴⁰

Consuming excessive quantities of low-nutrient, energy-dense foods such as sugar-sweetened beverages is a risk factor for obesity. Reducing intake of sugared beverages may be one of the easiest and most-effective ways to reduce ingested energy levels.⁴¹

Dairy Foods and Calcium
As early as 1984, it was reported that dietary calcium intake was related inversely to BMI in adults. Only recently have additional research reports been published relating low dietary calcium intake to human adiposity.

Of the 7 studies reviewed that assessed dietary calcium intake, 4 found no associations^7,10,42,43 and 3 found inverse associations^12,44,45 between calcium intakes and various measures of adiposity. In a cross-sectional study of primarily white youths, intake of calcium, after controlling for dietary energy and intake of dairy foods, was lower among overweight than nonoverweight 9- to 14-year-old youths.¹² No studies found a positive association between calcium intake and adiposity.

Although energy intake was controlled for in most of those analyses, such epidemiologic findings may be misleading, because dairy products reportedly are avoided by individuals concerned about their weight. However, prospective studies of preschool-aged children confirmed that greater longitudinal intake of calcium was associated with lower body fat.^44,45

The data suggest a potential role for calcium and dairy foods in the development of overweight and the potential for preventing weight gain by improving the dairy food intake of youths, indicating that a low intake of calcium may be associated with increased adiposity. However, the relative importance of calcium and dairy foods, in comparison with each other and in comparison with other factors involved in the development of overweight, remains to be established.

Dietary Fiber
Many governmental and scientific health agencies recommend that adults consume at least 20 to 25 g of fiber per day. Because children require less total energy, an "age + 5" rule for dietary fiber intake has been recommended.⁴⁶ This means, for example, that a 5-year-old child should consume at least 10 g of fiber per day and fiber intake should approach adult levels (20–25 g per day) by 15 years of age.

Unfortunately, persons of all ages in the United States eat far fewer than the recommended number of servings of whole-grain products, vegetables, and fruits.⁴⁷ In 1994 to 1996, only 3% of individuals 2 years of age consumed 3 daily servings of vegetables (with at least one third being dark green or orange vegetables), whereas only 7% consumed 6 daily servings of grains (with 3 being whole grains).⁴⁸ Currently, dietary fiber intake throughout childhood and adolescence averages 12 g/day or 5 g/1000 kcal (4200 kJ), a level of intake that has not changed in the past 30 years.⁴⁶ Because total carbohydrate content has increased considerably during this period, most of this increase seems to be in the form of fiber-poor refined grains, starchy vegetables, and sugar-sweetened beverages. It is worth investigating whether this apparent increase in consumption of fiber-poor foods is causally related to the observed increase in childhood obesity prevalence.

Dietary fiber may be related to body weight regulation through plausible physiologic mechanisms that have considerable support in the scientific literature. A large number of short-term studies suggest that high-fiber foods induce greater satiety. Epidemiologic studies generally support a role for fiber in body weight regulation among free-living individuals consuming self-selected diets, although conclusive intervention studies that address this are lacking.⁴⁹ Therefore, there is considerable reason to conclude that fiber-rich diets containing nonstarchy vegetables, fruits, whole grains, legumes, and nuts may be effective in the prevention and treatment of obesity in children. Such diets may have additional benefits, independent of changes in adiposity, in the prevention of cardiovascular disease and type 2 diabetes mellitus.^50,51

Macronutrient Alterations
Carbohydrates and Fat
Several adult studies have shown that significant weight loss can be achieved over 3 to 6 months with energy-restricted or ad libitum dietary prescriptions varying widely in macronutrient composition.^52–59 However, follow-up rates have been disappointing. Weight loss at follow-up times of 12 to 18 months rarely exceeds 5% of baseline weight.^{52–54,60–67} Although ad libitum, very- low-carbohydrate diets seem to be more efficacious than energy-restricted, low-fat diets over the short term,^52,54,55,63 Foster et al⁵⁴ found no significant group difference in mean body weight at 12 months. A study by Stern et al,⁶³ which included patients with type 2 diabetes, had similar results. With regard to pediatric data from a short-term study, Sondike et al⁶⁸ reported greater weight loss (–9.9 kg, compared with –4.1 kg) for adolescents who were instructed to follow an ad libitum, very-low-carbohydrate diet, compared with an ad libitum, low-fat diet, for 12 weeks. Findings from that study must be interpreted cautiously, however, in light of data on adults indicating poor compliance and weight regain over the long term on an "Atkins-type" diet.^54,63 In addition, there is widespread concern about the safety of severe carbohydrate restriction, especially for children.^69,70 Although very-low-carbohydrate diets may have some beneficial effects on risk factors for cardiovascular disease and type 2 diabetes,^54,63 the overall effects of this approach on other disease processes and on growth and development are unknown.

Very-low-fat diets have been shown to promote weight loss in several studies with adults.^71–74 However, those studies were not included in our systematic review for 1 of the following reasons: the design was not a randomized control trial,^72,73 body weight was not a primary outcome,⁷⁴ the intensity of intervention varied (ie, very-low-fat diets combined with other intensive lifestyle changes were compared with usual care),⁷⁴ or long-term follow-up data were not included.⁷¹

Protein
Evidence of long-term effectiveness (>1 year after treatment) of a high-protein, low-carbohydrate diet (also known as a protein-sparing modified fast [PSMF]) is extremely limited. There are 2 obvious reasons for this lack of evidence. First, relatively few studies of programs that use this type of intervention have been conducted.^75–77 Second, the studies that do exist suffer from substantial methodologic limitations. For example, all studies were from the same treatment program and all analyzed only 1 component of a multicomponent intervention that included diet and physical activity. A major concern with the use of a PSMF diet to treat childhood overweight is that the very low energy intake may compromise children's growth.

The PSMF is not a diet to be used for long-term treatment of overweight. Rather, the purpose of using a PSMF diet is to bring about rapid weight loss during the initial phase of treatment while minimizing the negative effects of a very-low-energy diet. In studies using PSMF, patients were on the diet for a relatively short initial treatment period and then were placed on a reduced-energy, balanced-macronutrient, maintenance diet. The goal was for the children to maintain the significant weight loss achieved during the "acute" treatment phase. All studies reported a statistically significant decrease in measures of overweight at the end of the initial treatment period. However, only 1 study actually compared the outcomes with a PSMF diet versus a balanced-macronutrient diet.⁷⁵ That study found that subjects on the PSMF diet lost significantly (P < .001) more weight from baseline to after treatment (BMI decrease: 5.2 ± 1.3 kg/m²) than did the children on the comparison, balanced-macronutrient diet (BMI decrease: 2.4 ± 1.4 kg/m²). Although the investigators found significant weight loss after the initial treatment, the same degree of weight loss was not maintained at 1 year. In contrast, other researchers at the same facility^76,77 studying longer-term outcomes found that children were generally able to maintain the weight loss after the initial treatment. Average BMI values were significantly lower than baseline values (P < .0001) both immediately and 1 year after treatment. In summary, these studies demonstrated that children initially treated with a PSMF diet were able to maintain some weight loss at 1 year. However, the researchers did not provide a true diet comparison, because the PSMF diet contained 200 kcal (840 kJ) less per day than did the balanced-macronutrient control diet and potential differences in other aspects of the multicomponent program were not accounted for.

Alternative Approaches
Dietary interventions based on energy density (ie, energy per mass of food) also have been considered as an approach to weight management. A series of short-term feeding studies, summarized by Rolls,⁷⁸ suggest that decreasing energy density decreases energy intake independent of macronutrient ratio, possibly because of effects on satiety. Diets of low energy density, which are typically rich in vegetables, fruits, legumes, and minimally processed grain products, allow individuals to consume satisfying portions of food while reducing their energy intake. Other studies included in the review indicate that the volume of food consumed exerts a stronger effect than energy content. Decreasing the energy density but maintaining or increasing the volume of core foods in a weight management program may help decrease energy intake. In a preliminary report of ad libitum diets in obese women, greater weight loss was achieved at 6 months by reducing energy density, with emphasis on increasing consumption of water-rich foods and decreasing consumption of high-fat foods, than by reducing fat intake alone⁷⁹; however, weight loss did not differ between dietary intervention groups at 12 months.

The glycemic index (GI) has been proposed to affect body weight regulation and risk for obesity-associated complications.⁸⁰ The GI is defined as the area under the glucose dose-response curve after consumption of 50 g of available carbohydrate from a test food, divided by the area under the curve after consumption of 50 g of available carbohydrate from a control food (either white bread or glucose). Short-term feeding studies indicated that hunger and cumulative food intake were greater 3 to 5 hours after a high-GI versus low-GI meal, controlled for macronutrient and energy contents.⁸¹ However, not all observational studies found a direct association between GI and weight gain. Translational studies found that pair-fed rodents consuming nutrient-controlled, high-GI diets had 70% to 90% greater adiposity than did those consuming low-GI diets.⁸² Few long-term clinical trials evaluating low-GI diets in children have been conducted. After controlling for potentially confounding factors, 1 nonrandomized study found that children attending an obesity treatment clinic and assigned to a low-GI diet lost more weight than did those assigned to a low-fat diet.⁸³ A small-scale, randomized, controlled trial found that adolescents lost more weight on a diet with low glycemic load (mean GI x carbohydrate amount) than on a low-fat diet.⁵⁷ Studies comparing the effects of high-GI versus low-GI diets on body weight in adults have produced conflicting results; some showed that low-GI diets led to weight loss,⁸⁴ whereas others showed no difference in weight.⁸⁵

Summary of Macronutrient Alterations
Data on optimal dietary approaches for weight management in children are lacking, and long-term studies of available interventions in adults have not demonstrated efficacy. Therefore, research into the development and testing of novel dietary approaches to obesity prevention and treatment is warranted. An emerging body of literature suggests that a focus on the macronutrient ratio is too simplistic and the quality of dietary carbohydrates and fats is an important consideration. The evidence for children and adolescents does not support any specific macronutrient or dietary strategy at this time.

	FOOD BEHAVIORS

TOP ABSTRACT NUTRITIONAL TREATMENT FOOD BEHAVIORS DIETARY INTERVENTIONS PHYSICAL ACTIVITY TELEVISION VIEWING AND MEDIA... BEHAVIORAL APPROACHES OTHER INTERVENTIONS RECOMMENDATIONS REFERENCES

 
Breakfast Skipping
Evidence supports observations that obese children are more likely to skip breakfast or to eat smaller breakfasts than leaner children. The evidence seems to suggest that breakfast skipping may be a risk factor for increased adiposity, particularly among older children or adolescents. However, the strength of the evidence is limited because what constitutes a breakfast has not been defined consistently.¹³

Fifteen studies examining the link between breakfast skipping and adiposity were reviewed. Two studies were longitudinal studies,^38,86 2 were nationally representative, cross-sectional studies,^31,87 and 11 were other types of cross-sectional investigations.^9,35,88–96 Both longitudinal studies^38,86 found that, for girls, breakfast skipping was related to weight gain among those who had normal weight at baseline but was related to weight loss among those who were overweight at baseline. For boys, no relationship was found with breakfast skipping except for weight loss among those who were overweight at baseline in 1 of the 2 studies. The 2 nationally representative studies^31,87 did not find an association between breakfast skipping and reported BMI in younger children, but Siega-Riz et al,⁸⁷ who studied food intake patterns for adolescents, did find a positive association.

Of the remaining 11 studies, 5 found a positive association between breakfast skipping and a measure of adiposity,^{88,90–92,96} indicating that breakfast skippers were more likely to have a weight higher than normal. Four studies found no relationship between breakfast skipping and a measure of adiposity,^35,89,93,95 and 2 studies reported a negative relationship between breakfast skipping and a measure of adiposity, indicating that breakfast skipping was associated with lower measures of adiposity.^9,86

Population-based surveys have revealed that many children, particularly adolescents, skip breakfast and other meals but consume more food later in the day, and this pattern has increased in recent years. Overweight children and adolescents have been shown to be more likely to skip breakfast and to consume a few large meals each day than their leaner counterparts, who are more likely to consume smaller, more-frequent meals. Overweight children have also been reported to eat smaller breakfasts and larger dinners, in comparison with nonoverweight children. It has been suggested that eating breakfast reduces fat intake and limits snacking over the remainder of the day.⁸⁸

Snacking
In a review of the literature, the American Dietetic Association¹³ found that snacking frequency or snack food intake might not be associated with adiposity in children. The majority of the studies reviewed found no association between snacking and adiposity.^{35,36,88,96–99} Francis et al¹⁰⁰ found no relationship between snacking and changes in BMI among girls with nonoverweight parents. Among girls with overweight parents, only fat intake from energy-dense snacks was associated with increased BMI over the 4-year study. However, mixed results were reported among the 7 case-control and other cross-sectional studies that examined the amount of snack food consumed in relation to adiposity. Two found a positive relationship,^35,98 whereas 5 found no relationship.^{8,10,88,97,98} Comparisons of the findings from those studies are limited because there was no clear definition of what constituted a snack or snack food. The best evidence suggests that snacking frequency is not associated with adiposity in children; however, studies that examined total snack food intake produced more-mixed results.

According to national surveys, although the average size of snacks and the energy per snack remained relatively constant, the frequency of self-defined snacking increased from 1977 to 1996 among children in all age groups between 2 and 18 years. Reportedly, between one fourth and one third of the energy intake of adolescents is derived from snacks.³⁹ Furthermore, snacks tend to have higher energy density and fat content than meals, and frequent snacking has been associated with high intakes of fat, sugar, and energy. The primary snacks selected by teens include potato chips, ice cream, candy, cookies, breakfast cereal, popcorn, crackers, soup, cake, and carbonated beverages.

Eating Out
Evidence shows that consuming food away from home, particularly at fast food establishments, may be associated with adiposity, especially among adolescents. A total of 12 observational studies were reviewed, including 2 longitudinal studies with children and 1 longitudinal study with adults,^38,101,102 2 nationally representative, cross-sectional studies,^31,103 and 7 other cross-sectional studies.^{89,92,104–107} Study sample sizes ranged from just over 50¹⁰⁴ to >60000.¹⁰¹ The majority of studies focused on older children and adolescents. In a longitudinal study of girls, Thompson et al¹⁰¹ reported a positive association between eating at fast food establishments and BMI z scores for elementary school-aged girls but no association with eating at coffee shops or other types of restaurants. Taveras et al¹⁰³ found, in a study of >14000 girls and boys, that greater consumption of fried foods eaten away from home was evident for heavier adolescents and that increasing consumption of fried foods eaten away from home over time led to an increase in BMI. In addition, the frequency of eating fried foods away from home was associated with greater intakes of total energy, sugar-sweetened beverages, and trans fats, as well as less consumption of low-fat dairy foods and fruits and vegetables. The other longitudinal study,³⁸ which was conducted in Japan, found no relationship between eating out in general and BMI among preschool-aged children. A nationally representative study by Lin et al³¹ found no association between food eaten away from home and reported BMI. Pereira et al¹⁰² found in the Coronary Artery Risk Development in Young Adults (CARDIA) study that consumption of fast foods was associated directly with body weight and insulin resistance over 15 years among young black and white adults. Findings from the other studies were mixed, ranging from positive associations to inverse relationships. Because both the largest longitudinal study and the largest cross-sectional study took place outside the United States (in Japan and Iran, respectively), their findings are not directly applicable to the US fast food environment. However, the limited evidence currently available suggests that frequent patronage of fast food restaurants may be a risk factor for overweight/obesity in children¹³ and fast food ingestion year after year may accumulate into larger weight gains that can be clinically significant.¹⁰³

	DIETARY INTERVENTIONS

TOP ABSTRACT NUTRITIONAL TREATMENT FOOD BEHAVIORS DIETARY INTERVENTIONS PHYSICAL ACTIVITY TELEVISION VIEWING AND MEDIA... BEHAVIORAL APPROACHES OTHER INTERVENTIONS RECOMMENDATIONS REFERENCES

 
Use of Balanced-Macronutrient/Low-Energy Diets
As stated previously, limited research exists for evaluating dietary treatment programs in isolation. However, a few dietary components have been evaluated. Although the outcomes are mixed, evidence does suggest that, in both the short term and the long term, a reduced-energy diet (less energy than required to maintain weight but not less than 1200 kcal [5040 kJ]/day) may be an effective part of a multicomponent weight management program in children 6 to 12 years of age.^108–113 Use of a reduced-energy diet (not less than 1200 kcal [5040 kJ]/day) in the acute treatment phase for adolescent overweight is generally effective for short-term improvement in weight status; without continuing interventions, however, weight is regained.¹³

Six studies that used a reduced-energy diet (not less than 1200 kcal [5040 kJ]) for 6- to 12-year-old youths were reviewed. The studies indicated that the majority of treatment groups decreased in 1 measure of adiposity.^108–112 Only 2 studies reported an increase in weight at posttreatment or follow-up assessments.^112,113

Six studies used an energy-deficit dietary treatment for adolescents. Five focused exclusively on adolescents,^{62,68,114–116} whereas the sixth provided treatment for 11- to 16-year-old youths.¹¹⁷ Five of the 6 studies reported a decrease in 1 measure of adiposity. Saelens et al¹¹⁴ reported a statistical difference in posttreatment weight status among teens who received a behaviorally based treatment, compared with a single-session, energy-deficit and activity approach, but differences diminished at the 3-month follow-up assessment. Only 2 studies reported follow-up periods of 1 year.^62,117 In both of those studies, follow-up weight status was not higher than baseline. Generalizing the results of these studies is difficult because of differences in the treatment environment, duration, and intervention strategies. Treatment settings were outpatient clinics or boarding schools, whereas interventions ranged from computer-based programs with additional nutrition and activity counseling to health center-based, multicomponent programs. In addition, the length of the programs ranged from 3 weeks to 9 months.¹³

Traffic Light Diet
Much of our current understanding of individual/family treatment of pediatric overweight comes from 4 long-term, family-based studies conducted by Epstein et al.^118–123 The studies by Epstein et al^118–123 targeted children 6 to 12 years of age. The traffic light diet (sometimes called the stoplight diet) was developed by Epstein et al^118–123 for use in research on overweight. Perhaps because of the groundbreaking nature of their research, the traffic light diet has become broadly recognized and in some cases copied. The traffic light diet is part of a larger core package of interventions that generally includes family components, physical activity, and interactions with a behavioral therapist. The core intervention program was used in all studies, whereas other variables were manipulated. This presents a problem in trying to isolate the independent effects of the specific dietary intervention on weight loss.

The goal of the traffic light diet was to provide the most nutrition with the lowest energy intake. Daily energy intakes ranged from 900 to 1200 kcal (3780–5040 kJ), with later studies increasing intake to 1500 kcal (6300 kJ)/day.¹²² Food groups were divided into 3 categories, namely, green, yellow, and red. Low-energy, high-nutrient foods (eg, most fruits and vegetables) are considered "green" and may be eaten often. Moderate-energy foods (eg, most grains) are considered "yellow" and may be eaten in moderation, whereas high-energy, low-nutrient foods are considered "red" and should be eaten sparingly. Families were instructed to stay within a prescribed energy range and to reduce "red" food servings to less than a prescribed value for the week (eg, <4 times per week). In addition to the basic diet, and depending on the arm of the intervention study, participants might have been given self-monitoring training and support, praise and modeling, therapist contact, and/or behavioral contracting, in which children were given rewards for meeting dietary and activity goals. Once children/families met their weight goals, counseling was provided to ensure consumption of a balanced diet that would maintain a healthy weight.

The intervention and research program by Epstein et al^118–123 demonstrated modest sustained weight loss in children 5 years and even 10 years after the intervention.¹²³ However, not all of the behavioral interventions provided sustained weight loss.¹²³ It remains unclear what part the diet itself played in these overall results. Because the research by Epstein et al^118–123 focused primarily on white, middle-class, intact families with younger children (6–12 years of age), it is also unclear how well results may be generalized beyond this population.¹³

Food Guide Pyramid
The Food Guide Pyramid was designed as a general guide for diet and exercise and not as a weight loss tool. Although it may be used as a component of a comprehensive childhood weight management program, the evidence does not indicate that, by itself, the Food Guide Pyramid is an effective weight loss tool. Only 1 study was identified that used the pyramid as a weight loss tool. Saelens et al¹¹⁴ found that adolescents who used the Food Guide Pyramid as part of their weight management program gained weight over the course of treatment and at follow-up evaluations. This was compared with adolescents in the control group who ate a balanced, lower-energy diet, whose weight either stabilized or decreased slightly.

	PHYSICAL ACTIVITY

TOP ABSTRACT NUTRITIONAL TREATMENT FOOD BEHAVIORS DIETARY INTERVENTIONS PHYSICAL ACTIVITY TELEVISION VIEWING AND MEDIA... BEHAVIORAL APPROACHES OTHER INTERVENTIONS RECOMMENDATIONS REFERENCES

 
Importance of Physical Activity
Although obesity has a complex development, involving environmental, physiologic, and genetic factors, the basic cause of this condition is an imbalance between energy intake and energy expenditure. Physical activity is the only modifiable component of the energy expenditure portion of the energy balance equation. Consequently, increasing physical activity has the potential to improve weight loss and maintenance. Studies indicate that an increase in sedentary activities, particularly television viewing, and an overall decrease in physical activity are contributing to an increased incidence of overweight and obesity in children and adolescents.^124–127 Strategies to increase physical activity should include increases in structured and nonstructured physical activity and reductions in the amount of time spent in sedentary activities. Schools have a unique combination of factors, including facilities, fitness instructors, and contact with large numbers of young people for many hours each day during much of the year, that make them a good environment in which to study physical activity interventions for weight management and to implement proven approaches.¹²⁸

Role of Physical Activity in Weight Management
Noting that accurate measurement of physical activity is complex and that comparisons between studies are difficult because of differences in designs and methods, some researchers have questioned whether it is possible to demonstrate an effect of physical activity in reducing obesity.¹²⁹ Because it is easier to reduce energy intake by 500 to 1000 kcal (2100–4200 kJ)/day than to increase energy expenditure by a similar amount, physical activity has less impact on weight loss than does dietary intervention.

In adults, increasing physical activity did not result in significant weight loss over a 6-month period. Most weight loss occurred as a result of decreased energy intake. Sustained physical activity did reduce the risk of weight regain and also decreased cardiovascular and diabetes risk factors, independent of the reductions in these risks that are associated with weight loss.¹³⁰

Several but not all studies have demonstrated that increased physical activity is associated with decreased BMI in children and adolescents.^131–133 The largest of those studies¹³³ examined the association between changes in BMI over 1 year and same-year changes in self-reported recreational physical activity and in recreational inactivity (television, videotapes, and video games) among 11887 boys and girls 10 to 15 years of age. After correction for growth- and development-related changes in BMI, an increase in physical activity was associated with decreasing relative BMI for girls (–0.06 kg/m² per 1-hour increase in daily activity; 95% confidence interval [CI]: –0.11 to –0.01 kg/m² per 1-hour increase) and for overweight boys (–0.22 kg/m² per 1-hour increase; CI: –0.33 to –0.10 kg/m² per 1-hour increase). Conversely, higher levels of inactivity were correlated with increased BMI in girls (+0.05 kg/m² per 1-hour increase in television, videotapes, and video games; CI: +0.02 to +0.08 kg/m² per 1-hour increase). One study found a relationship between inactivity in overweight preschool-aged boys but not girls.¹³² The third study, involving 47 boys and girls 5 to 10.5 years of age, measured total energy expenditure directly by using the double-labeled water technique and calculated basal metabolic rate by using the Schonfield equation. It used these measurements to calculate physical activity levels, as follows: physical activity level = total energy expenditure/basal metabolic rate. Body fat and BMI were used to estimate body composition. Body fat and BMI were found to be significantly inversely correlated with physical activity levels.¹³¹

Studies that use weight loss as the only criterion with which to assess the value of increased physical activity may miss other important benefits this confers. In a meta-analysis, P. McGovern, PhD (unpublished data, 2006) found that physical activity decreased fat mass but not BMI. Other studies indicated that exercise also improved cardiovascular risk factors.^134,135

Risks of Physical Inactivity
Physical activity may play a role in preventing weight gain and other health problems. Physical inactivity has been shown to be a risk factor for obesity and insulin resistance in school-aged children.¹³⁶ Inactive children may be at increased risk of developing health problems later in life. Several studies suggest that sedentary children are more likely than active children to become sedentary adults and to have increased risks of obesity, diabetes, hypertension, dyslipidemias, and cardiovascular diseases.^{129,135,137,138} A sedentary lifestyle is also associated with increased risks of several cancers common in adults.¹³⁹

Structured Versus Nonstructured Physical Activity
There is some debate in the literature regarding whether structured or unstructured activities should be promoted as a means to increase physical activity. The position of the American Academy of Pediatrics on physical fitness and activity in schools advocates increases in both forms of activity.¹⁴⁰ It states that the development of a physically active lifestyle should be a goal for all children. Opportunities to be physically active should include team, individual, noncompetitive, and lifetime sports, as well as recreational activities. The opportunity to be active on a regular basis, as well as the enjoyment and competence gained from activity, may increase the likelihood that a physically active lifestyle will be adopted.¹⁴⁰

Beyond the school setting, increasing physical activity, even unstructured physical activity, seems to be beneficial.¹³³ It is thought that increasing the frequency or intensity of physical activity can reduce sedentary activities, particularly television viewing. This, in turn, can reduce excess energy balance effectively.¹⁴¹ The goal is not to eliminate television watching; data suggest that children and adolescents can engage in both television viewing and physical activity as long as sedentary behavior is not at the expense of physical activity.^142–145

Amount of Physical Activity
Since 2000, the US Department of Agriculture has recommended that children and adolescents participate in 60 minutes of moderate-intensity physical activity most days of the week, preferably daily.¹⁴⁶ This position was reaffirmed in the 2005 Dietary Guidelines for Americans¹⁴⁷ and is supported by the American Academy of Pediatrics¹⁴⁰ and the Centers for Disease Control and Prevention (CDC).¹⁴⁸ The American Academy of Pediatrics recommends that 30 minutes of this activity occur during the school day.¹⁴⁰ Very obese children may need to start with shorter periods of activity and gradually increase the time spent being active. The CDC suggests that parents can help children meet this activity goal by serving as role models, incorporating enjoyable physical activity into family life, monitoring the time their children spend watching television, playing video games, and using the computer, and intervening if too much time is spent in sedentary pursuits.¹⁴⁸

Barriers to Physical Activity
Barriers to physical activity for the pediatric population include lack of opportunities for activity during the school day and environmental factors, such as lack of access to facilities in which to be active and urban environments designed for vehicular transportation that limit activity outside of school.^149–156 In the past decade, schools have been urged to spearhead improvements in childhood wellness through changes in the food and activity programs they offer. The amount of time spent on physical education, however, has decreased in the past 15 years. Between 1991 and 2003, the percentage of high school students enrolled in daily physical education classes decreased from 41.6% to 28.4%. Only 8% of elementary schools, 6.4% of middle/junior high schools, and 5.8% of senior high schools provided daily physical education or allocated the recommended amount of time per week (150 minutes for elementary and 225 minutes for junior and senior high schools), according to a 2000 study.¹⁴⁹

More consideration needs to be given to the types of activities performed during physical education class, because time spent in class does not correlate with activity. Data for 37000 students collected by the CDC as part of the annual Youth Risk Behavior Surveillance Survey found that high school students were active for only 16 of the 50 minutes in an average gym class. Spending more time in physical education classes did not help. When states required an extra year of physical education classes for high school students, which is 200 more minutes of physical education per week, male students reported, on average, another 7.6 minutes per week spent exercising or playing sports in gym class. Female students reported, on average, an extra 8.1 minutes per week spent exercising in class.¹⁵⁰

Increasing the intensity of activity during gym class can improve fitness and reduce body fat measurably. Fifty overweight (BMI of 95th percentile) children in middle school were assigned randomly to lifestyle-focused, fitness-oriented, gym classes or standard gym classes for 9 months. The children were evaluated for fasting insulin and glucose levels and body composition and assessed with maximal oxygen consumption treadmill testing at the beginning and at the end of the school year. Overweight children who participated in the fitness-oriented gym classes for 9 months showed significant improvements in body composition, fitness, and insulin levels.¹⁵¹ These studies^149–151 indicate that public health policies should focus on revising school curricula to include adequate time for and intensity of physical activity.

Safety concerns, such as heavy traffic and high crime rate, lack of equipment, lack of space, and urban development that favors vehicular transportation are barriers to activity outside of school.^152–156 The World Health Organization has identified transport-related physical activity as an important intervention with which to address the global obesity epidemic, as well as environmental issues such as traffic congestion and its associated pollutants.¹⁵² In the United States, a decrease in transport activity parallels the increase in pediatric obesity. Walking or biking to and from school can help students meet their physical activity needs. However, heavy traffic, lack of bicycle lanes, unmarked intersections, and other obstacles have reduced the number of children who transport themselves to school today, compared with previous generations.

Currently, only one third of students who live within 1 mile of school walk or bike there and <3% of students who live within 2 miles of school walk or bike there. Initiatives such as the government-sponsored, community-implemented, Safe Routes to School program may help reverse this trend. A pilot study conducted in Marin County, California, found that the number of children walking to school increased from 14% in 2000 to 23% in 2002 and the number riding bicycles more than doubled, from 7% to 15.5%, in the same period. Similar results have been observed in other communities around the country. These experiences highlight the need for parents and other interested community members to take an active role in reducing barriers to physical activity.¹⁵³

Lack of access to safe exercise environments is of particular concern in low-socioeconomic status and minority communities, because this may account for racial and economic disparities in health, obesity, and physical activity rates. Burdette and Whitaker¹⁵⁴ found an inverse association between neighborhood safety and television viewing among preschool-aged children. Parents who rated their neighborhoods as unsafe were more likely to report that their preschool-aged children watched >2 hours of television daily. No association was found between television viewing and obesity in these young children; however, early television viewing may establish a pattern of sedentary activity that leads to obesity in later childhood. An observational study examined the associations between community physical activity-related settings (eg, sports areas, public pools and beaches, parks and green spaces, and bicycle paths) and race, ethnicity, and socioeconomic status in 409 communities throughout the United States.¹⁵⁵ The researchers reported that higher median household incomes and lower poverty rates were associated with increasing levels of available physical activity-related facilities and settings. Communities with greater proportions of ethnic minorities had fewer physical activity-related settings. Using data from the first wave of the National Longitudinal Study of Adolescent Health (N = 20745), Gordon-Larsen et al¹⁵⁶ demonstrated a direct relationship between decreased access to physical activity facilities and overweight. They found that communities with low socioeconomic status and large minority populations had reduced access to recreational facilities. These factors were associated with decreased physical activity levels and increased incidence of overweight. These associations suggest that lack of opportunities for physical activity may contribute to the disproportionately greater incidence of obesity in ethnic minority groups and groups with low socioeconomic status.¹⁵⁶

Other researchers have found that schools with large minority populations are less likely to have programs that support healthy eating and physical activity.^157–159 A survey of 3600 households with children 9 to 13 years of age that was conducted by the CDC in 2002 found that non-Hispanic black parents and Hispanic parents cited concerns about transportation, lack of local facilities, and expense as barriers to their children participating in physical activity and organized sports outside of school more often than did non-Hispanic white parents.¹⁶⁰ There are many communities and neighborhoods in which inadequate school wellness programs, lack of access to facilities for physical activity outside of the classroom, and lack of discretionary income may contribute to the high obesity rates seen among economically disadvantaged individuals.

Higher economic status does not guarantee that people will live in neighborhoods that encourage more activity. Built environments in suburban communities often are not conducive to walking, biking, and other physical activities. Neighborhood comparison and correlational studies with physical activity transport outcomes suggest that residents from communities with higher density, greater connectivity, and more land use mixture have higher rates of walking/cycling for utilitarian purposes than do low-density, poorly connected, and single-land use neighborhoods. Environmental variables seem to add to the variance accounted for beyond sociodemographic predictors of walking/cycling for transport.¹⁶¹

In what is thought to be the first study to examine the link between obesity in rural communities and environmental factors, Boehmer et al,¹⁶² from the St Louis University School of Public Health, found that residents of rural communities who felt isolated from recreational facilities, stores, churches, and schools were more likely to be obese than were those who thought they were closer to such facilities. Closeness counted; people who thought that safe walking and/or biking routes were within 10-minute walking distance of their homes were more likely to be active. Approximately 25% of the population in states in the South and Midwest live in rural environments.¹⁶² To ensure that people of all income levels have opportunities to be physically active as part of their daily routines, community leaders and environmental planners need to address safety and access issues.

Psychosocial barriers such as perceptions of class rank and self-esteem, and their impact on physical activity, have been studied but are not well understood. Physical activity self-efficacy (confidence in one's ability to participate in exercise) has been widely studied as a potential psychosocial correlate of increased levels of physical activity. However, this association is not clear for children and adolescents.¹⁶³

Reducing Sedentary Activities
As a first step toward addressing neighborhood safety barriers to activity, the American Academy of Pediatrics recommends that activities that can be performed indoors, such as exercising to videotapes, using hula hoops, and dancing to popular music, should be encouraged.¹⁶⁴ A complementary strategy for promoting physical activity among children and adolescents is to decrease their inactivity by decreasing the time spent in sedentary activities such as television viewing, leisure time use of the computer, and video game playing. Staying active while watching television by stretching, performing calisthenics, or using exercise equipment can also reduce the time spent in sedentary pursuits. Television viewing may have a negative effect on both sides of the energy balance equation. It may displace active play and physical activity time and it is associated with increased food and energy intake, as an accompaniment to television viewing and as a result of food advertising.¹⁶⁵

Summary
Addressing childhood obesity requires a comprehensive holistic approach. Although the evidence is limited, increased physical activity alone has not improved children's weight status substantially. Promotion of routine physical activity in children from preschool age on may help prevent the development of overweight and obesity and has other benefits, including reductions in cardiovascular disease risk factors. Particular consideration should be given to methods of increasing activity in adolescents. Studies suggested that time, cost, availability, and convenience were key factors that influenced what adolescents ate and whether they were physically active.^166,167 Students reported that social support from friends and family members, as well as teachers and adults who modeled healthy behaviors, enhanced their likelihood of eating healthy foods and being physically active.^168–170 Finally, the American Academy of Pediatrics¹⁴⁰ recommends that (1) all children meet the goal of 60 minutes of moderate activity per day; (2) schools be provided with the necessary resources to incorporate 30 minutes of moderate to intense activity into each student's daily schedule; (3) clinicians instruct parents on techniques for increasing activity in the home environment, including reducing time spent in sedentary activities; and (4) health care providers become involved in the community to address access and safety issues.

	TELEVISION VIEWING AND MEDIA USAGE

TOP ABSTRACT NUTRITIONAL TREATMENT FOOD BEHAVIORS DIETARY INTERVENTIONS PHYSICAL ACTIVITY TELEVISION VIEWING AND MEDIA... BEHAVIORAL APPROACHES OTHER INTERVENTIONS RECOMMENDATIONS REFERENCES

 
Television Viewing and Obesity
Investigators have examined many aspects of diet and physical activity, but some of the strongest evidence of a behavioral risk for overweight in children points to the impact of television viewing. Epidemiologic and experimental evidence from the past decade supports decreased television viewing as a primary preventive intervention for the reduction of overweight and other chronic disease risks. Many cross-sectional^{136,171–177} and longitudinal^{133,173,178–180} observational studies in the United States document the effect of television viewing on overweight. These studies are reinforced by others in at least 12 other countries.^{173,181–190} The observational studies have been corroborated by randomized, controlled trials designed to reduce levels of both television viewing and overweight. In a randomized, controlled trial, Gortmaker et al¹⁹¹ showed that reductions in television viewing were associated with decreased obesity. Among girls, each 1-hour reduction in television viewing predicted reduced obesity prevalence. Guillaume et al¹⁹⁰ found a significant relationship with BMI and systolic blood pressure for television viewing in boys. In a randomized, controlled, school-based trial, Robinson¹⁹² showed that intervention groups had statistically significant decreases in BMI with reductions in television viewing and eating meals in front of the television. That author also found reductions in waist circumference and waist/hip ratios. In a randomized trial, Epstein et al¹⁹³ found that, at 1-year follow-up assessments, children who were counseled regarding decreasing sedentary activities versus increasing physical activity or a combination of the 2 had a greater decrease in the percentage of overweight than did children from the other 2 groups. In fact, children in the sedentary activity-reduction group increased their liking for high-intensity activity and reported lower energy intake than did children in the exercise group.

Influences on Diet, Physical Activity, and Chronic Disease Risks
Television viewing is likely to influence overweight by replacing more vigorous activities, as well as affecting diet.^144,178,194 Foods are heavily advertised in children's television programming,¹⁰⁵ and television viewing is associated with children's between-meal snacking.¹⁹⁵ A randomized trial indicated that increased television viewing resulted in increased energy intake and decreased energy expenditure.¹⁹⁶

Other studies documented similar effects of television viewing on overweight among preschool-aged children.¹⁹⁷ Reducing excess television viewing among youths is a national health objective for 2010.¹⁹⁸ Since 1986, the American Academy of Pediatrics has recommended limiting television viewing to no more than 2 hours/day for children 2 years of age.¹⁹⁹ The American Academy of Pediatrics has broadened this guideline by recommending no television viewing for children <2 years of age and suggesting that total entertainment media time be limited to no more than 1 to 2 hours/day for children >2 years of age.²⁰⁰ Other studies demonstrated that having a television in the room where a child sleeps is a major predictor of television viewing^{177,180,201,202} and that, once in the room, televisions often are not removed.²⁰³ These data point to the utility of early intervention strategies to limit television viewing.

An important fact about television viewing is that it correlates only minimally with measures of moderate and vigorous physical activity²⁰³ and therefore is an independent risk factor for overweight. Similar findings on the impact of television viewing on overweight, independent of the effects of moderate and vigorous physical activity, have been reported in prospective studies of adults^204,205 and in studies of television viewing, physical activity, and diabetes incidence among men and women.^206,207

Several studies also indicated that television viewing has substantial effects on other risk factors for chronic disease, including smoking,²⁰⁸ reduced fruit and vegetable consumption,²⁰⁹ increased aggression,²⁰⁰ and less time spent reading and doing school homework.²¹⁰ During the developmental period in which television viewing becomes an entrenched habit (ie, the preschool/early primary school years),²¹¹ children also are developing physical activity skills and learning to read. Other potential benefits of reduced television viewing in this age group may be increased physical activity and reading.

Socioeconomic Status, Ethnicity/Race, Television Viewing, and Overweight
The prevalence of childhood and adolescent overweight in the United States has grown most rapidly among black and Hispanic youths, and health disparities have widened in the past decade.²¹² The treatment writing group strongly supports the Healthy People 2010 goal of eliminating gender-, race/ethnicity-, and socioeconomic status-associated disparities in health status, risks, and use of preventive services. Groups with lower socioeconomic status and racial/ethnic minority groups generally are at greater risk of morbidity and death resulting from chronic diseases, including cardiovascular disease, stroke, and diabetes mellitus.^213–216 Therefore, reducing television viewing among young ethnic minority children in the United States has the potential to reduce excess chronic disease among youths, as well as to reduce adult rates of morbidity and death resulting from chronic illnesses.¹³⁶ Several studies have noted substantially higher levels of television viewing among ethnic minority children, particularly black children, and among boys, compared with girls.^{175,177,217,218} A number of studies also reported stronger associations between television viewing and overweight among girls, compared with boys,^{126,191,210,219,220} including a randomized trial that found that the strongest effects of reduced television viewing were in black girls.²¹⁹ These differences according to gender and ethnicity²²¹ indicate the need to focus on cultural diversity²²² in developing interventions, as well as increasing awareness that efforts to reduce television viewing have the potential to reduce ethnic and gender disparities in overweight.

Other Media Usage
In the past 5 years, media use by children has increased significantly. However, limited research is available on uses of "screen time" other than television, such as computers, video games, DVDs, and instant messaging. In a recent study of parents of children 0 to 6 years of age, Vandewater et al²²³ found that, on a typical day, 75% of children watched television and 32% watched videotapes/DVDs, for 1 hour and 20 minutes, respectively, on average. New media also are making inroads with young children; 27% of 5- to 6-year-old children used a computer (for 50 minutes, on average) on a typical day. Many young children (one fifth of 0- to 2-year-old children and more than one third of 3- to 6-year-old children) also have a television in their bedrooms. The most common reason given was that this frees up other televisions in the house so that other family members can watch their own shows (54%). The majority of children 3 to 6 years of age fell within the American Academy of Pediatrics guidelines, but 70% of 0- to 2-year-old children did not.

Another study of older children and adolescents²²⁴ found that approximately one half (53%) of all 8- to 18-year-old youths said that their parents gave them no rules about television watching. Nearly one half (46%) said that they did have rules regarding screen time but only 20% said that the rules were enforced most of the time. Most importantly, youths with rules that were enforced reported 2 hours less of media exposure per day than did those in homes without this supervision. Despite the concerns parents express about the impact of media on their children, this study did not find much evidence of major parental efforts to curb or to monitor viewing habits.

Summary
Epidemiologic and experimental evidence from the past decade supports decreased television viewing as a primary preventive intervention for the reduction of overweight and other chronic disease risks. Screen time for children >2 years of age should be limited to no more than 1 to 2 hours/day. Television viewing is not recommended for children <2 years of age. Parents need to take an active role in setting total screen time limits and monitoring their children's viewing habits. Health care professionals should encourage parents not to put a television in the room where their child sleeps and to remove the television if it is already there.

	BEHAVIORAL APPROACHES

TOP ABSTRACT NUTRITIONAL TREATMENT FOOD BEHAVIORS DIETARY INTERVENTIONS PHYSICAL ACTIVITY TELEVISION VIEWING AND MEDIA... BEHAVIORAL APPROACHES OTHER INTERVENTIONS RECOMMENDATIONS REFERENCES

 
Techniques
Behavioral therapy for pediatric obesity uses a number of techniques that modify and control children's food and activity environments in ways that bring about weight loss. These techniques include removing unhealthy foods from the home, monitoring behavior by asking children or parents to keep track of the foods consumed, setting goals for energy consumption and physical activity, and rewarding children's and sometimes parents' successful changes in diet and physical activity. Additional behavioral approaches include training in problem solving and other parenting skills. These techniques have been described in detail elsewhere.²²⁵

Subjects, Settings, and Delivery Formats
Most published trials of behavioral interventions have taken place in specialty treatment centers staffed by physicians, nutritionists, exercise therapists, and/or psychologists. The programs studied were conducted or designed by a multidisciplinary team of providers, including a psychologist, and included children and adolescents 5 to 17 years of age. These programs generally included behavioral interventions in conjunction with changes in diet and physical activity, delivered at least in part in a group setting. Comparative data that identify the optimal frequency of visits do not exist. However, most outpatient-based interventions included 8 to 16 initial weekly group sessions lasting 45 to 90 minutes, followed by visits of decreasing frequency for a total duration of 4 to 12 months.

In one trial, group-only treatment was as effective in producing weight loss and was more cost-effective than combined group and individual family sessions.²²⁶ Two inpatient programs based in Belgium and an 8-week summer camp program in Massachusetts showed efficacy in producing weight loss and improved psychological well-being.^227–229 Another trial, conducted in Germany in an inpatient treatment program with 9- to 19-year-old obese adolescents, compared self-management of weight and muscle relaxation training as additions to a structured exercise and diet program and found no added benefit beyond inpatient effects.²³⁰

Few studies of pediatric obesity treatment have been conducted as part of primary care. One trial of overweight adolescents included a single session with a primary care provider, followed by either telephone- and mail-based behavioral intervention or no additional treatment. There was some evidence of better efficacy among the behaviorally treated adolescents, although absolute efficacy was less than with more-intensive clinic-based and inpatient interventions.¹¹⁴ Approaches using Internet-based treatment offer some