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 Barlow, S. E. Search for Related Content PubMed PubMed Citation Articles by Barlow, S. E. Related Collections Nutrition & Metabolism

Expert Committee Recommendations Regarding the Prevention, Assessment, and Treatment of Child and Adolescent Overweight and Obesity: Summary Report

Division of Pediatric Gastroenterology, Nutrition, and Hepatology, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas

	ABSTRACT

TOP ABSTRACT EPIDEMIOLOGIC FEATURES DEFINITIONS AND TERMINOLOGY OVERVIEW OF PROVIDER OFFICE... PREVENTION ASSESSMENT TREATMENT APPENDIX: EXPERT COMMITTEE... REFERENCES

 
To revise 1998 recommendations on childhood obesity, an Expert Committee, comprised of representatives from 15 professional organizations, appointed experienced scientists and clinicians to 3 writing groups to review the literature and recommend approaches to prevention, assessment, and treatment. Because effective strategies remain poorly defined, the writing groups used both available evidence and expert opinion to develop the recommendations. Primary care providers should universally assess children for obesity risk to improve early identification of elevated BMI, medical risks, and unhealthy eating and physical activity habits. Providers can provide obesity prevention messages for most children and suggest weight control interventions for those with excess weight. The writing groups also recommend changing office systems so that they support efforts to address the problem. BMI should be calculated and plotted at least annually, and the classification should be integrated with other information such as growth pattern, familial obesity, and medical risks to assess the child';s obesity risk. For prevention, the recommendations include both specific eating and physical activity behaviors, which are likely to promote maintenance of healthy weight, but also the use of patient-centered counseling techniques such as motivational interviewing, which helps families identify their own motivation for making change. For assessment, the recommendations include methods to screen for current medical conditions and for future risks, and methods to assess diet and physical activity behaviors. For treatment, the recommendations propose 4 stages of obesity care; the first is brief counseling that can be delivered in a health care office, and subsequent stages require more time and resources. The appropriateness of higher stages is influenced by a patient's age and degree of excess weight. These recommendations recognize the importance of social and environmental change to reduce the obesity epidemic but also identify ways healthcare providers and health care systems can be part of broader efforts.

Key Words: obesity • prevention • assessment • treatment • clinical practice pattern • chronic care model • office management • motivational interviewing • overweight • patient education • nutrition assessment

Abbreviations: AST—aspartate aminotransferase • ALT—alanine aminotransferase • CDC—Centers for Disease Control and Prevention • NAFLD—nonalcoholic fatty liver disease • USDA—US Department of Agriculture • CE—consistent evidence • ME—mixed evidence

In 1997, when the Department of Health and Human Services Health Resources and Service Administration convened the first expert committee to develop recommendations on the evaluation and treatment of child and youth obesity, few studies of this problem had been conducted to provide evidence for the recommendations. Since then, increasing scientific attention has resulted in an expanded body of literature on the causes, comorbidities, and treatment of this problem. The condition remains frustrating and difficult to treat but, with more-current scientific information available, in 2005 the American Medical Association, in collaboration with the Health Resources and Service Administration and the Centers for Disease Control and Prevention (CDC), convened a new expert committee that was charged with providing revised recommendations. These new recommendations use current, evidence-based data, as well as clinical experience when evidence does not exist, to provide updated practical guidance to practitioners (see Appendix for the complete recommendations).

Representatives from 15 national health care organizations formed the expert committee. The steering committee, composed of representatives from the American Medical Association, the Health Resources and Service Administration, and the CDC, invited these member organizations because they serve children at high risk of obesity, they represent experts in obesity-related conditions, or they represent experts in aspects of obesity treatment. The representatives from the 15 member organizations submitted nominations for the experts who would compose the 3 writing groups and work on the following 3 areas of focus: prevention, assessment, and treatment of childhood overweight and obesity. Special care was taken both to ensure that a broad range of disciplines, including medicine, nutrition, nursing, psychology, and epidemiology, was represented and to capture the interests of diverse cultural groups. The experts in these groups reviewed the scientific information that forms the basis of the expert committee recommendations. Their work is referred to throughout this report according to the area of review (prevention, assessment, or treatment), and their reports accompany this article.^1–3

Each multidisciplinary writing group reviewed the current literature to develop the recommendations. Because the science continues to lag behind the obesity epidemic, many gaps in evidence-based recommendations remain. With few exceptions, randomized, controlled, intervention trials have not been performed to prove or to disprove the effect of a particular behavior on weight control in obese children. The available studies often examine associations between health behaviors and weight or between health behaviors and energy balance. Even less evidence exists about the process of addressing obesity in a primary care setting. The purpose of the expert committee was to offer practical guidance to clinicians by providing recommendations in all areas of obesity care, including those that lack the best possible evidence. When evidence of an effect on obesity was not available, the writing groups considered the literature, clinical experience, the likelihood of other health benefits, the possible harm, and the feasibility of implementing a particular strategy before including it. Although a thorough evidence-based review was beyond the scope of this project, the writing groups provided a broad rating of the evidence, so that readers can appreciate the limitations of these recommendations and watch for new studies that will refine them. The rating categories were as follows:

1. recommends with consistent evidence (CE), that is, multiple studies generally show a consistent association between the recommended behavior and either obesity risk or energy balance;
   
2. recommends with mixed evidence (ME), that is, some studies demonstrated evidence for weight or energy balance benefit but others did not show significant associations, or studies were few in number or small in sample size;
   
3. suggests, that is, studies have not examined the association of the recommendation with weight or energy balance, or studies are few, small in number, and/or without clear findings; however, the expert committee thinks that these recommendations could support the achievement of healthy weight and, if future studies disprove such an effect, then these recommendations are likely to have other benefits and are unlikely to cause harm.
   

The report provides qualitative ratings of evidence for the recommended lifestyle behaviors. The summary report recommends assessment of the lifestyle behaviors that are targets for change but does not rate evidence for the assessment process; the literature in this area, cited in the assessment report,² is sparse and has limited applicability to an office setting. The writing groups also addressed the implementation of clinical care for obesity. At the level of the family, the writing groups suggested strategies to encourage and to support a patient or family that chooses to change eating or physical activity behaviors. At the level of the provider office, the committee suggested ways in which the office system can change to track overweight and obese children and to support family management of this chronic condition. The scarcity of studies about the process of obesity treatment precluded an evidence review. The recommendations represent a consensus based on the best available information. Ongoing research will eventually provide the best possible evidence for childhood obesity care, and future recommendations will reflect new knowledge. In the meantime, clinicians, who routinely make clinical decisions in the absence of the best possible evidence, will find updated guidance for this pervasive condition.

The writing groups presented their recommendations to the expert committee for discussion and revision in May 2006. Once consensus was reached, the committee members then presented the recommendations to their member organizations for endorsement (see "Acknowledgments" for expert committee and writing group participants).

	EPIDEMIOLOGIC FEATURES

TOP ABSTRACT EPIDEMIOLOGIC FEATURES DEFINITIONS AND TERMINOLOGY OVERVIEW OF PROVIDER OFFICE... PREVENTION ASSESSMENT TREATMENT APPENDIX: EXPERT COMMITTEE... REFERENCES

 
Childhood Obesity Epidemic
The rapid increase in the prevalence of childhood obesity has alarmed public health agencies, health care clinicians, health care researchers, and the general public. On the basis of measured heights and weights from nationally representative samples of US children assessed approximately every 5 years, obesity prevalence has increased from 5% in 1963 to 1970 to 17% in 2003 to 2004.⁴ Clinicians are faced with addressing this problem with a steadily increasing number of patients.

Obesity and overweight are defined on the basis of age- and gender-specific BMI normative values that were established when the distribution of BMI values was constant.⁵ The increase in obesity prevalence is therefore measured against a stable cutoff point, the 95th percentile BMI for gender and age.

Demographic Features
The obesity epidemic has disproportionately affected some racial/ethnic groups. In 2003–2004, the prevalence rates were particularly high among black girls (24%) and among Mexican American boys (22%).⁴ Rates have also increased among Native American and Asian American youths.^6,7 Overall, poverty has been associated with greater obesity prevalence among adolescents; however, subgroups have differed. In 1 report, for example, obesity prevalence among younger black male adolescents was higher in nonpoor families than in poor families but prevalence among older black male adolescents was higher in poor families.⁸ Higher family socioeconomic status is associated with lower obesity prevalence among white girls but not among black girls.⁷

Causes
Both genes and environment contribute to obesity risk. Twin studies have clearly demonstrated a genetic risk,⁹ and the discovery of leptin, ghrelin, adiponectin, and other hormones that influence appetite, satiety, and fat distribution provides insight into metabolic mechanisms for physiologic risk.¹⁰ With multiple substances and multiple gene sites associated with obesity, the system is complex, redundant, and likely not amenable to a simple pharmaceutical intervention. However, genes are not destiny. Just as behavior and environment strongly influence a person's risk of developing skin cancer, behavior and environment influence the development of obesity in genetically at-risk people. At a population level, the increase in prevalence is too rapid to be explained by a genetic shift; rather, it must result from changes in eating and physical activity behaviors that have shifted the balance of energy intake and energy expenditure.

The influence of specific behavior changes on energy balance is difficult to determine. Many cross-sectional studies and some longitudinal studies have examined the relationships between specific behaviors (for instance, intake of sugar-sweetened beverages or participation in daily physical education classes) and obesity. Interventional studies that examine prospectively the impact of a behavior on weight or BMI are rare. Each of the writing groups reviewed the literature for evidence of the influence of behaviors on either energy balance or BMI. The review found evidence for only a few behaviors. One important limitation of these studies is measurement validity. For assessment of energy intake under normal, free-living circumstance, subjects must report the food they consume, through either recall or a food diary. These methods are inaccurate and subject to underreporting.^11,12 Measuring physical activity is somewhat less problematic, with improved accelerometers and the capacity to measure accurately the total energy expenditure through labeled-water techniques. Probably a bigger challenge in this scientific area is the large number of possible eating and activity behaviors that may contribute to energy imbalance. If greater sugar-sweetened beverage intake, larger portion sizes at all meals and snacks, more-frequent snacks, more ready-to-eat foods, more restaurant eating, more television viewing, fewer physical education classes, less walking to and from school, less outside play at home, more escalators, elevators, and automatic doors, and so forth, all coexist, then the impact of any one of those behaviors on obesity prevalence may be unmeasurable.

Scientists continue to study obesity but, given its complex causes, years or decades may pass before the most effective intervention or prevention strategies are identified. The recommendations presented here are evidence based where evidence is available; where evidence is not available or is incomplete, the expert committee has combined data with clinical judgment, including selected interventions when such interventions are reasonable and are unlikely to cause harm. An example is the recommendation to increase fiber intake. Although studies have not demonstrated that increased fiber intake leads to improved weight, foods that are high in fiber have lower energy density and could displace other foods, resulting in overall reduced energy intake. This diet change, even if unproven, has other nutritional benefits and is unlikely to cause harm. As discussed above, this summary report includes a general assessment of the quality of evidence for each behavior. The prevention, assessment, and treatment reports provide detailed descriptions of studies for each topic.^1–3

	DEFINITIONS AND TERMINOLOGY

TOP ABSTRACT EPIDEMIOLOGIC FEATURES DEFINITIONS AND TERMINOLOGY OVERVIEW OF PROVIDER OFFICE... PREVENTION ASSESSMENT TREATMENT APPENDIX: EXPERT COMMITTEE... REFERENCES

 
Measurement of Body Fat
High levels of body fat are associated with increasing health risks. However, no single body fat value, whether measured as fat mass or as percentage of body weight, clearly distinguishes health from disease or risk of disease. Even if body fat level could be measured easily, other factors, such as fat distribution, genetics, and fitness, contribute to the health assessment.

BMI, a measure of body weight adjusted for height, is a useful tool to assess body fat. BMI is defined as weight (in kilograms) divided by the square of height (in meters). BMI levels correlate with body fat^13,14 and also correlate with concurrent health risks, especially cardiovascular risk factors.¹⁵ High BMI predicts future adiposity, as well as future morbidity and death.¹⁶ The sensitivity of BMI of >85th percentile for identifying the fattest children is good,¹⁷ and, in contrast to more-precise measures of body fat (such as dual-energy x-ray absorptiometry), health care providers can assess weight and height routinely. Although BMI does not measure body fat directly and therefore may lead to imprecise assessment of adiposity, it is feasible and has acceptable clinical validity if used thoughtfully. Another practical benefit of BMI use for children is the continuity with recommended assessments of adult body weight.

For children, the distribution of BMI changes with age, just as weight and height distributions change. As a result, although absolute BMI is appropriate to define body weight in adults, percentiles specific for age and gender define underweight, healthy weight, overweight, and obesity in children.

The validity of BMI depends in part on the cutoff points used. Like body fat levels, BMI and BMI percentiles are continuous, and any cutoff point will be imperfect in distinguishing those with health risks from those without. When a high cutoff point is selected, patients with "normal" BMI despite high body fat levels will be misclassified as healthy. When the cutoff point is low, patients with high BMI despite normal body fat levels (for example, muscular athletes) will be misclassified as unhealthy. The cutoff point selection must balance overdiagnosis and underdiagnosis. Because body fat levels and health risks are continuous, clinicians should rely on BMI as a useful tool that triggers concern and assessment, but they should recognize that other clinical information influences the need for intervention.

Pediatric Cutoff Points and Terminology: Same Cutoff Points, New Terms
The use of 2 cutoff points, namely, BMI of 95th percentile and 85th percentile, captures varying risk levels and minimizes both overdiagnosis and underdiagnosis. When BMI is <85th percentile, body fat levels are likely to pose little risk. When BMI is 95th percentile, body fat levels are likely to be high. BMI of 85th to 94th percentile indicates health risks that vary depending on body composition, BMI trajectory, family history, and other factors. These cutoff points are unchanged from the 1998 expert committee recommendations¹⁸ and CDC⁵ and Institute of Medicine¹⁹ recommendations.

The expert committee recommends different terminology. The committee suggests that, when BMI is 95th percentile, the term "obesity" should replace "overweight" and, when BMI is 85th to 94th percentile, "overweight" should replace "at risk of overweight." The compelling reasons for this revision are clinical. The term obesity denotes excess body fat more accurately and reflects the associated serious health risks more clearly than does the term overweight, which is not recognized as a clinical term for high adiposity. Overweight denotes high weight from high lean body mass as well as from high body fat levels and so is appropriate for the 85th to 94th percentile category, which includes children with excess body fat as well as children with high lean body mass and minimal health risks. These terms provide continuity with adult definitions and avoid the vagueness of "at risk of overweight," which has been confusing to patients and health care providers. Because the recommended cutoff points have not changed, these terms will not affect the prevalence rates of the BMI categories.

Exceptions to the use of 85th and 95th percentile BMI values as cutoff points occur for older and younger children. For older adolescents, BMI of 95th percentile is higher than BMI of 30 kg/m², the adult obesity cutoff point. The committee therefore recommends that obesity in youths be defined as BMI of 95th percentile or BMI of 30 kg/m², whichever is lower. For children <2 years of age, BMI normative values are not available. Weight-for-height values above the 95th percentile in this age group can be categorized as overweight.

Stigmatization associated with the term obesity has been one reason for the use of the term overweight. The negative connotation of obesity results from pervasive social prejudice and deserves attention.^20–22 However, the committee recommends that clinicians address this concern through supportive demeanor and language in the clinical encounter. The terminology and cutoff points for both adults and children have been debated, but several groups have weighed the advantages and disadvantages and made similar recommendations (Table 1).

Much legitimate concern exists about stigmatization of overweight and obese children.^21,23 Public concern followed decisions to assess BMI in schools, because of the potential harm of labeling a child with a condition that is a target of prejudice.²⁴ Health care visits are generally a good place to identify excess weight, because the setting frames the condition as a health problem and because the visit is private. Therefore, clinicians must take responsibility for identification but must approach the subject sensitively, to minimize embarrassment or harm to self-esteem. Consistent with the 1998 recommendations,¹⁸ the expert committee urges clinicians to be supportive, empathetic, and nonjudgmental. A careful choice of words will convey an empathetic attitude. Adult patients have identified "fatness," "excess fat," and "obesity" as derogatory terms,²⁵ and obese adolescents prefer the term "overweight."²⁶ Younger children and their families may respond similarly, and clinicians should discuss the problem with individual families by using more-neutral terms, such as "weight," "excess weight," and "BMI." Therefore, the expert committee recommends the use of the clinical terms overweight and obesity for documentation and risk assessment but the use of different terms in the clinician's office, to avoid an inference of judgment or repugnance.

Recognition of the need for a third cutoff point to define severe obesity in childhood obesity seems to be evolving. An adolescent weighing 180 pounds and another weighing 250 pounds are in the same BMI category (>95th percentile) but face markedly different social and medical effects. New data indicate that extreme obesity in children is increasing in prevalence, and these children are at high risk for multiple cardiovascular disease risk factors.²⁷ A definition of severe childhood obesity would help identify these children so that their particular risks and treatment needs can be established. The expert committee proposes recognition of the 99th percentile BMI, which is BMI of 30 to 32 kg/m² for youths 10 to 12 years of age and 34 kg/m² for youths 14 to 16 years of age. The marked increase in risk factor prevalence at this percentile provides clinical justification for this additional cutoff point. Although much additional study with larger and more-diverse samples is needed to characterize the medical and social risks of this category, the committee recommends that clinicians recognize this BMI cutoff point and ensure that best efforts are made to provide treatment to these youths and their families. Because the 97th percentile is the highest curve available on the growth charts, Table 3 provides 99th percentile cutoff points according to age and gender.

	OVERVIEW OF PROVIDER OFFICE PROCESS

TOP ABSTRACT EPIDEMIOLOGIC FEATURES DEFINITIONS AND TERMINOLOGY OVERVIEW OF PROVIDER OFFICE... PREVENTION ASSESSMENT TREATMENT APPENDIX: EXPERT COMMITTEE... REFERENCES

View larger version (30K): [in this window] [in a new window]   FIGURE 1 Universal assessment of obesity risk and steps to prevention and treatment. DM indicates diabetes mellitus.

Chronic Care Model
The expert committee recommendations are comprehensive and ambitious. Health care-centered efforts alone cannot effect change, but they can complement and potentially enhance evolving public health efforts, such as school wellness policies, parks and recreation programs, and shifts in child-targeted food advertisements. In addition, health care provider offices and health care systems will need to change, in many cases, to implement these recommendations. These recommendations can serve as guides that will improve as new information becomes available.

The traditional office visit model works best for acute problems, such as otitis media or joint injury; the clinician assesses the single problem, orders additional testing as needed, and presents a treatment plan (generally short-term) to the patient. However, the complexity of chronic problems, such as diabetes mellitus or obesity, and their requirement for patient education about self-management often overwhelm both the patient and the clinician during an office visit. The chronic care model²⁸ envisions a new structure that integrates community resources, health care, and patient self-management to provide more-comprehensive and more-useful care. This paradigm envisions offices linked to community resources, such as exercise programs; support for self-management, which requires educating patients and families about assessment and monitoring; an expanded practice team that supports patient self-management and monitors adherence to evidence-based care pathways; and clinical information systems that can remind the team of routine tests and treatments and can monitor the practice's adherence to goals.^28,29 Changes in office procedure require deliberate planning and evaluation, and the rapid-cycle quality improvement method may be a useful approach for continuous quality improvement.³⁰ In this model, practices plan a change and a method to measure that change, implement the change, and then examine the measure of change. The plan is modified depending on how well the goals are met. The cycle is repeated until the practice is satisfied with the change. For example, a practice could plan to include the BMI category next to the vital signs on the patient's chart for well-child visits. After 3 months, the practice would measure the percentage of charts that included the BMI category. If only 60% of charts included the BMI category, then staff members would discuss barriers and propose a new plan, such as readily available BMI calculators or the need for the office secretary to add growth charts with BMI curves to records of established patients with weight-for-height curves. With the new routine in place, progress toward the goal would again be measured.

The chronic care model has obvious applications in childhood obesity, and several large health maintenance organizations have initiated some of these approaches. Kaiser Permanente has trained staff members to use motivational interviewing, and Wellpoint has distributed parental toolkits in primary care clinics, to help educate office personnel about children's development and appropriate nutrition and physical activity. Several programs have linked to community efforts, such as community-based exercise programs. The care model for obesity recognizes the importance of changes in the school, worksite, and community. Figure 2 shows how the environment and the medical system support the patient and the family in their management of the condition.

View larger version (23K): [in this window] [in a new window]   FIGURE 2 Obesity care model.

Obesity Care and Cultural Values of Patients
Beliefs about what is an attractive weight or a healthy weight, what foods are desirable or appropriate for parents to provide children, how families should share meals, the importance or enjoyment of physical activity, and the authority parents have over children at different ages, as well as many other attitudes that affect lifestyle habits, are influenced by cultural values and beliefs. Some studies have examined differences between identified racial, ethnic, or cultural groups, such as the observation that black girls are more satisfied with heavier bodies than are white girls.³² Low-income mothers may recognize obesity as a problem, not on the basis of growth curves but when they perceive that high weight restricts their child's tolerance for physical activity.³³ A study of low-income minority parents of preschool-aged children showed that Hispanic parents had indulgent feeding styles more often than did low-income black parents.³⁴ Population studies indicate that levels of vigorous physical activity differ according to age and racial group.^35,36 However, studies in these areas are incomplete. Barriers to behavior change may be related to community circumstances, such as lack of safe recreation areas, rather than values and preferences. Clinicians should inform themselves about the values or circumstances that may be common in the population they serve, especially if that population differs from their own. However, a clinician's knowledge of an individual family's personal values and circumstances, which are not dictated by the family's ethnic, racial, or economic group, may be most helpful in tailoring recommendations.

	PREVENTION

TOP ABSTRACT EPIDEMIOLOGIC FEATURES DEFINITIONS AND TERMINOLOGY OVERVIEW OF PROVIDER OFFICE... PREVENTION ASSESSMENT TREATMENT APPENDIX: EXPERT COMMITTEE... REFERENCES

 
Importance
Given the difficulty of behavior-based weight loss and subsequent weight maintenance and the expense and potential harm of medication and surgery, obesity prevention should be a public health focus. Efforts must begin early in life, because obesity in childhood, especially among older children and those with more-severe obesity, is likely to persist into adulthood.³⁷ Therefore, childhood represents an important opportunity to establish healthy eating and activity behaviors that can protect children against future obesity. Pediatric providers are accustomed to addressing health behaviors, such as car seat use, tobacco avoidance, and avoidance of risky sexual behavior, and they provide guidance on nutrition in early childhood routinely. In addition, they know the family's medical history and social and behavioral interactions. They are well positioned to guide families in the areas of eating and activity.

The targets of obesity prevention should be all children, starting at birth. Lifestyle behaviors to prevent obesity, rather than intervention to improve weight, should be aimed at children with healthy BMIs (5th–84th percentile) and some children with BMIs in the overweight category (85th–94th percentile), depending on their growth pattern and risk factors. Clinicians should be aware of the increased risk of obesity for children with obese parents and those whose mothers had diabetes mellitus during the child's gestation. Indeed, young children with 1 or 2 obese parents are at high risk of obesity in young adulthood, even if their current weight is normal.

Target Behaviors
The expert committee recommends that clinicians advise patients and their families to adopt and to maintain the following specific eating, physical activity, and sedentary behaviors. These healthy habits may help prevent excessive weight gain and also are unlikely to cause harm, on the basis of current knowledge. The level of evidence is indicated, and the prevention report provides references.¹

Evidence supports the following:

1. limiting consumption of sugar-sweetened beverages (CE);
   
2. encouraging consumption of diets with recommended quantities of fruits and vegetables; the current recommendations from the US Department of Agriculture (USDA) (www.mypyramid.gov) are for 9 servings per day, with serving sizes varying with age (ME);
   
3. limiting television and other screen time (the American Academy of Pediatrics³⁸ recommends no television viewing before 2 years of age and thereafter no more than 2 hours of television viewing per day), by allowing a maximum of 2 hours of screen time per day (CE) and removing televisions and other screens from children's primary sleeping area (CE) (although a relationship between obesity and screen time other than television viewing, such as computer games, has not been established, limitation of all screen time may promote more calorie expenditure);
   
4. eating breakfast daily (CE);
   
5. limiting eating out at restaurants, particularly fast food restaurants (CE) (frequent patronage of fast food restaurants may be a risk factor for obesity in children, and families should also limit meals at other kinds of restaurants that serve large portions of energy-dense foods);
   
6. encouraging family meals in which parents and children eat together (CE) (family meals are associated with a higher-quality diet and with lower obesity prevalence, as well as with other psychosocial benefits); and
   
7. limiting portion size (CE) (the USDA provides recommendations about portions, which may differ from serving sizes on nutrition labels, and a product package may contain >1 serving size).
   

The prevention writing group also suggests, on the basis of analysis of available data and expertise, the following behaviors:

1. eating a diet rich in calcium (the USDA provides recommendations about serving size and daily number of dairy product servings);
   
2. eating a diet high in fiber;
   
3. eating a diet with balanced macronutrients (energy from fat, carbohydrates, and protein in proportions for age, as recommended by Dietary Reference Intakes)³⁹;
   
4. encouraging exclusive breastfeeding to 6 months of age and maintenance of breastfeeding after introduction of solid food to 12 months of age and beyond, consistent with American Academy of Pediatrics recommendations^40,41;
   
5. promoting moderate to vigorous physical activity for at least 60 minutes each day⁴²; and
   
6. limiting consumption of energy-dense foods.
   

Implementation
The complexity of obesity prevention lies less in the identification of target health behaviors and much more in the process of influencing families to change behaviors when habits, culture, and environment promote less physical activity and more energy intake. Handing families a list of recommended eating and activity habits as if it were an antibiotic prescription fits into traditional medical training and the structure of the office visit, but such an approach is rarely effective. The prevention writing group has provided suggestions on how to interact with families to promote target behaviors and how to create office systems that support the clinician's ongoing commitment to obesity prevention. The appendix of the prevention report presents an example of office visit structure, interaction between provider and patient, and specific language to illustrate this approach.¹

The Role of the Family
Several studies of obesity treatment in children have demonstrated the importance of parents’ participation in weight control programs.^43–45 The commitment of parents and other caregivers to helping the child develop healthy habits to prevent obesity is likely to be very important. Parents can serve as role models, authority figures, and behavioralists to mold their children's eating and activity habits. Clinicians can influence children's habits indirectly by teaching and motivating parents to use their authority effectively. For very young children, clinicians should focus the discussion on parenting behavior. The greater independence of adolescents means that clinicians should discuss health behaviors directly with them, although clinicians should encourage parents to make the home environment as healthy as possible.

Parenting actions that support the target behaviors differ with the age of the child, and clinicians can provide appropriate material to assist parents. For instance, clinicians can discuss or provide information about encouraging free safe movement for infants, appropriate food portions for toddlers, limited stroller use for preschoolers, and easy breakfast alternatives for teenagers. The prevention report contains a list of age-specific parenting actions.¹

Clinicians function as counselors in obesity prevention and obesity care. Briefly presented below are some counseling techniques that can be used to encourage parents and patients to improve healthy behaviors. Short courses at local or national meetings can give clinicians greater opportunities to learn counseling techniques, which are generally not taught as part of the usual health care education.

Patient-Centered Communication
The theories that follow assume that a clinician's instruction to change a behavior will be effective only if the parent or family recognizes a potential problem and wants to address or to prevent that problem. Therefore, part of the clinician's task is to help motivate families. Counseling techniques are presented here as ways to encourage obesity prevention behaviors, but the same approach has applications in obesity treatment behaviors.

The stages of change theory describes several cognitive stages that precede actual behavior change.⁴⁶ According to this theory, a person may initially be unaware of a problem, then move to awareness of the problem but have no plans to address it, then plan a new behavior, and finally actually begin the new behavior. A clinician can help patients and families move along these stages, rather than prescribing a new behavior to those who are not ready. Recent work indicates that parents of overweight and obese children are often unaware of the condition.⁴⁷

The technique of motivational interviewing, which also takes into account patients’ readiness to change, uses nonjudgmental questions and reflective listening to uncover the beliefs and values of a parent or patient. By eliciting the concerns of patients, the clinician can evoke motivation, rather than try to impose it, and then help patients formulate a plan that is consistent with their own values. This approach avoids the defensiveness created by a more-directive style.

1. nondirective questions about the parent's or patient's attitude should be used ("Your child's BMI is above the 95th percentile. What concerns, if any, do you have about her weight?") The clinician's next steps depend on the parent's response. This approach differs from a directive style, in which the clinician informs the family of the seriousness of the condition ("Your child's BMI is very high, and it is important that your child gets control of her weight before it becomes a bigger problem.");
   
2. reflective listening, in which the clinician summarizes the parent's comments without judging them, should be used ("If I heard you correctly, you are concerned about how much television your child is watching, but you know your child is safe and happy watching television when he is home alone.") Reflections help build rapport and allow the patient to understand and to resolve ambivalence;
   
3. values and current health practices should be compared; if a parent values her child being healthy and a good student, then the clinician can help the parent examine how activities other than television could improve the child's health and academic performance;
   
4. importance/confidence rulers should be used ("On a scale from 0 to 10, with 10 being the highest, how important is it to you to change your child's television viewing?" "On a scale from 0 to 10, with 10 being the highest, how confident are you that you could decrease his television viewing to 2 hours a day?") The number the parent gives leads the clinician to ask, "What would it take to get you to a higher number?" The clinician should thus help the parent think of solutions to the problem.
   

Table 4 presents an example of an interchange during an office visit that focuses on obesity prevention and incorporates motivational interviewing techniques.

The Role of the Provider's Office
The provider's office system can enhance or undermine the clinician's efforts to address obesity prevention consistently. The expert committee endorses the following office practices.

1. routine documentation of BMI. Although clinicians visually recognize obesity in many children without seeing the plotted BMI values, they may overlook excess body fat in children in the overweight (85^th–94th percentile BMI) category and miss an opportunity to guide the family toward healthier behaviors. To document BMI consistently and accurately, offices need reliable scales for infants and children, recumbent infant length boards, and wall-mounted stadiometers. This equipment needs regular calibration. Staff members must know how to measure weight and height accurately, how to calculate BMI, and how to plot the measures on the growth curves;
   
2. establishment of procedures to deliver obesity prevention messages to all children. When the patient's individual risk of obesity is low, these messages can promote appropriate general health or wellness, rather than weight control. One example from collaborative efforts in Maine and Massachusetts is the 5 2 1 0 message, which reminds families to eat 5 fruits and vegetables, spend no more than 2 hours on screen time, include 1 hour of physical activity or active play, and consume little or no sugar-sweetened beverages. Clinicians remind families of these goals at all health supervision visits and have posters in the office and handouts that reinforce these recommendations. Although the specific content of such messages may vary until research establishes the best approach, simple memorable guidelines, presented early and repeated regularly, can be delivered efficiently in the office and are likely to be effective teaching tools;
   
3. establishment of procedures to address children who are overweight (85th–94th percentile BMI) and obese (95th percentile BMI). For instance, when a child is overweight, a practice may plan to review the family history, child's blood pressure, child's cholesterol level, and BMI percentile over time and then assess health risks on the basis of that information. Offices should flag charts of overweight and obese children, so that all providers at all visits are aware and can monitor growth, risk factors, and social/emotional issues;
   
4. involvement and training of interdisciplinary teams, including nurses, physicians, and administrative staff members, regarding their respective responsibilities and skills;
   
5. chart audits to establish baseline practices, to help set goals for practice improvement, and then to measure the improvement over time. Offices can use the techniques for continuous quality improvement from the rapid-cycle improvement method described above.
   

The Roles of the School and the Community
These recommendations focus on the office visit and the opportunity to influence the family routine and home environment, but the child's school and community environments can either support or impede obesity prevention behaviors. Clinicians can support school and community programs that help prevent obesity through local, state, or national advocacy, and they can encourage patients’ families to voice their preference to their schools through parent-teacher organizations or school board meetings or directly to principals, teachers, and after-care program directors. The Institute of Medicine report on obesity prevention provides a model for school policies.¹⁹ It recommends adequate physical education and recess periods and the establishment of nutritional standards for all foods served at school, including foods from vending machines and other competitive foods.¹⁹ To improve the community environment, providers can advocate for the establishment and maintenance of safe parks and recreation centers, and they can urge local grocery stores to offer healthy, low-cost food that is consistent with the most common cultures of the community members.

	ASSESSMENT

TOP ABSTRACT EPIDEMIOLOGIC FEATURES DEFINITIONS AND TERMINOLOGY OVERVIEW OF PROVIDER OFFICE... PREVENTION ASSESSMENT TREATMENT APPENDIX: EXPERT COMMITTEE... REFERENCES

 
Risks
When a child's BMI is above the 85th percentile, the clinician should assess medical and behavioral risks before initiating any intervention. Medical risks include risk of future or persistent obesity, risk of future obesity-related medical conditions, and identification of current obesity-related medical conditions. Behavioral risks include current eating habits, physical activity, and sedentary behaviors that promote energy imbalance. These evaluations must precede behavior-based treatment.

Medical Assessment
Responsibility
Screening children for obesity-related medical problems falls squarely in the purview of health care providers, especially primary care providers. Providers are responsible for considering any current obesity-associated medical conditions, such as hyperlipidemia, risks of future conditions associated with obesity and ameliorated by weight control, and rare conditions that cause obesity, such as primary Cushing syndrome or Prader-Willi syndrome. Because weight control alone may not treat many conditions adequately, diagnosis must be followed by appropriate treatment.

Body Fat Assessment
The BMI percentile, although imperfect, is the recommended screen for body fat in routine office practice. Offices should use the 2000 CDC BMI charts, rather than the International Obesity Task Force standards, because the CDC charts provide the full array of percentile levels (which makes them more appropriate for assessment of individual children), whereas the International Obesity Task Force charts provide only overweight and obesity categories.^5,48

Skinfold thickness measurements are not recommended. Although these measurements provide information about body fat and risks of medical conditions,⁴⁹ they are not feasible in routine clinical care, because they are difficult to perform accurately without careful training and experience and reference data are not readily available.

Similarly, waist circumference measurements are not recommended currently. Waist circumference measurements can provide indirect information about visceral adiposity, which tracks with cardiovascular and metabolic risk factors, and are more easily performed than skinfold thickness measurements,^50–52 but reference values for children that identify risk over and above the risk from BMI category are not available. In the future, cutoff points that provide additional information and can influence evaluation or treatment may make waist circumference measurement a useful clinical tool.

BMI percentile categories guide assessment of medical risk; 5th to 85th percentile is healthy weight, 85th to 94th percentile is overweight, and 95th percentile is obese, with >99th percentile being an emerging category that indicates a high likelihood of immediate medical problems. Because no objective assessment to distinguish high body fat from high lean body mass is clinically practical, clinicians must also consider the family history of obesity and medical problems, the child's past BMI pattern, and the child's current medical conditions and current health behaviors as they decide whether to recommend intervention.

Parental Obesity
Parental obesity is a strong risk factor for a child's obesity persisting into adulthood, especially for young children.³⁷ Genetic vulnerability plays an important role in the development of obesity. Although it is currently not possible to test for specific genotypes or to adapt therapy on the basis of genetic information, knowledge of strong familial risks for obesity, especially parental obesity, should lead to greater efforts to establish or to improve healthy behaviors.

Family Medical History
Several obesity-related medical conditions are familial. Family history predicts type 2 diabetes mellitus or insulin resistance, and the prevalence of childhood diabetes is especially high among several ethnic and racial backgrounds common in the United States, including Hispanic, black, and North American Indian.⁵³ Cardiovascular disease and cardiovascular disease risk factors (hyperlipidemia and hypertension) are also more common when family history is positive.⁵⁴ Offices should review and regularly update the family history regarding first- and second-degree relatives.

Evaluation of Weight-Related Problems
Screening
Obesity-related medical conditions affect almost every organ system in the body. A review of systems and a physical examination represent an inexpensive way to screen for many of these conditions, although some conditions are without symptoms or signs. Summarized below are important weight-related medical conditions, with their common symptoms and appropriate screening tests. Tables 5 and 6 present a review of systems and physical examination findings in the order typically followed in an office visit.

In obesity hypoventilation syndrome, the weight of fat on the chest and abdomen impairs ventilation; these patients are severely obese. Symptoms are similar to those of obstructive sleep apnea, and diagnosis is made through polysomnography, which demonstrates elevated carbon dioxide levels. These patients may have elevated hemoglobin and hematocrit levels. They require continuous positive airway pressure therapy until substantial weight loss relieves the condition.

Respiratory Problems
Asthma may occur more frequently among obese children.⁵⁸ Shortness of breath and exercise intolerance may be symptoms of asthma, rather than signs of poor physical conditioning. Diagnosis is made in the usual way. Obese patients with asthma may need guidance about asthma management during physical activity or outdoor play, to minimize the limitations on exercise.

Gastrointestinal Problems
Nonalcoholic fatty liver disease (NAFLD) is a condition of increasing concern because of the increasing prevalence of obesity and diabetes, which are important risk factors. The term NAFLD includes simple steatosis, steatohepatitis, fibrosis, and cirrhosis resulting from fatty liver. Knowledge of prevalence, natural history, and effective management is incomplete, although studies are ongoing. NAFLD generally causes no symptoms, although some patients have right upper quadrant abdominal pain or tenderness or mild hepatomegaly. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels, which are usually elevated, are reasonably good screens. Ultrasonography and other imaging methods can demonstrate changes consistent with nonalcoholic steatohepatitis but cannot indicate the degree of inflammation or fibrosis. Liver biopsy is the standard method for diagnosis. Weight loss leads to improved liver test results and histologic features, and studies of medications are ongoing.⁵⁹ When and how often to perform ALT and AST testing have not been determined; pending evidence-based recommendations, the expert committee suggests biannual screening starting at 10 years of age for children with BMI of 95th percentile and those with BMI of 85th to 94th percentile who have other risk factors. This schedule coincides with diabetes screening recommendations.⁶⁰ ALT or AST results 2 times normal levels should prompt consultation with a pediatric hepatologist.

Gallstones are more prevalent among overweight and obese children.⁶¹ In addition, rapid weight loss increases the risk of gallstones.⁶² Intermittent episodes of intense colicky pain in the right upper quadrant of the abdomen are classic symptoms, but milder pain and epigastric pain can occur. On physical examination, the right upper quadrant may be tender. Ultrasonography can identify gallstones and cholecystitis.

Several common pediatric gastrointestinal problems, including gastroesophageal reflux disease and constipation, are exacerbated by obesity.^63,64 Symptoms, signs, and management are the same as for children of normal weight, but clinicians should be aware of the increased likelihood of these conditions and should provide appropriate medical and behavioral treatment in addition to weight control.

Endocrine Disorders
Type 2 diabetes mellitus is one of the most serious complications of childhood obesity. As many as 45% of children with newly diagnosed diabetes mellitus have type 2 rather than type 1 disease.⁵³ Patients may not have symptoms such as polyuria and polydipsia; consequently, identification requires laboratory screening for children at risk. Risk factors are BMI of 85th percentile; family history of diabetes; black, Hispanic, or Native American background; and other related conditions, such as polycystic ovary syndrome, acanthosis nigricans, or cardiovascular risk factors. The American Diabetes Association currently recommends screening with a fasting glucose test when a child is overweight and has 2 additional risk factors. Screening should begin at puberty or 10 years of age and should be performed every 2 years. A fasting glucose level of 126 mg/dL or a casual glucose level of 200 mg/dL indicates diabetes and requires referral to a pediatric endocrinologist. Fasting glucose levels of 100 mg/dL are considered prediabetes, indicating future risk for diabetes.⁶⁰

Polycystic ovary syndrome occurs in 8% of young women 18 to 25 years of age, with prevalence depending on the definition used. Women with polycystic ovary syndrome are more likely to be obese.⁶⁵ Infrequent menses (<9 cycles per year) is the most important finding that should lead to additional evaluation. Physical examination findings that are common but not diagnostic for polycystic ovary syndrome are hirsutism, excessive acne, and acanthosis nigricans. Women with polycystic ovary syndrome often have insulin resistance or type 2 diabetes and may have metabolic syndrome. Reproductive hormone laboratory tests can diagnose the condition but generally require interpretation by a subspecialist, such as an endocrinologist, gynecologist, or adolescent physician (see the assessment report²); these specialists can initiate and monitor treatment to protect fertility.

Hypothyroidism is a frequent concern of parents, but this condition does not usually cause severe obesity. The prevalence is 1 case per 1000.⁶⁶ Symptoms include fatigue and decline in academic performance. Cessation of linear growth is an important sign, and a goiter may be present. Thyroid function tests are generally unnecessary when a child has normal linear growth velocity and no other symptoms of hypothyroidism.

Primary Cushing syndrome is extremely rare. The population incidence is probably 2 cases per 1 000 000 annually, with onset in adulthood being more common than onset in childhood.⁶⁷ Because the condition is treatable, clinicians should be aware of the physical examination findings, which include "moon facies" and "buffalo hump," although exogenous obesity can also lead to this distribution of adipose tissue. Primary Cushing syndrome generally leads to short stature and therefore is extremely unlikely in a tall obese child. The striae found in Cushing syndrome are violaceous in color and thus differ from the commonly seen striae resulting from rapid weight gain. If Cushing syndrome is suspected, then the child should be referred to an endocrinologist for appropriate testing.

Evaluation of puberty in obese children requires careful attention to physical examination findings and knowledge of the normal range of puberty onset. Adipose tissue in the breast area must be distinguished from true breast development, which is generally discernible through pigmented erectile areolae. A suprapubic fat pad, which can hide the penis and give it the appearance of a micropenis, must be manually reflected away. Obese children tend to begin puberty earlier than children of normal weight but, when onset is truly premature, these children require an endocrinologic evaluation just as do children of normal weight. Children at risk for endocrine disorders are white girls <7 years of age and black girls <6 years of age with breast tissue or pubic hair and boys <9 years of age with pubic hair or enlargement of the penis.

Nervous System Disorders
Pseudotumor cerebri is an extremely rare condition (incidence estimates for children are 1 case per 100 000 annually⁶⁸), but obesity is one of several risk factors⁶⁹ and, untreated, the condition can lead to vision loss. Patients describe severe headaches with photophobia. Patients may have double vision if they have impairment of cranial nerve VI. Optic disks are blurred. When suspected, this condition requires urgent referral to the neurology service.

Cardiovascular Risk Factors
Approximately 13% of overweight children have elevated systolic blood pressure, and 9% have elevated diastolic blood pressure.¹⁵ Blood pressure should be assessed at all health supervision visits, and offices should have large cuffs, including thigh cuffs, which allow accurate assessment of blood pressure for severely obese youths. The National Heart, Lung, and Blood Institute has updated tables defining elevated blood pressure levels according to age, gender, and height percentile, which offices should have available for easy reference.⁷⁰ Three or more readings above the 95th percentile for either systolic or diastolic blood pressure indicate hypertension. Information on the National Heart, Lung, and Blood Institute Web site (www.nhlbi.nih.gov/health/prof/heart/hbp/hbp_ped.htm) includes recommendations for evaluation, which may include ambulatory blood pressure monitoring to identify "white coat" hypertension or abnormal diurnal blood pressure patterns. Primary care providers can follow these detailed recommendations for evaluation and treatment or can refer patients to a specialist.

Lipid level abnormalities are among the most common obesity-related medical conditions.¹⁵ Because of the high prevalence, a fasting lipid profile should be obtained when BMI is 85th percentile, even in the absence of other risk factors. Total cholesterol levels of <170 mg/dL are acceptable, levels of 170 to 199 mg/dL are in the borderline category, and levels of 200 mg/dL are high. Low-density lipoprotein levels of <110 mg/dL are acceptable, levels of 110 to 129 mg/dL are borderline, and levels of 130 mg/dL are high. Dietitians can guide patients and families regarding the reduced-fat and reduced-cholesterol diets recommended by the National Cholesterol Education Panel.⁷¹ If levels are highly elevated and do not respond to diet changes, then a pediatric cardiologist or lipid specialist can assess the benefits and risks of medication use. Abnormal triglyceride levels, defined by the National Cholesterol Education Panel as 110 mg/dL for adolescents, and abnormal high-density lipoprotein levels, defined as 40 mg/dL, respond to increased physical activity.

Psychiatric Disorders
The effects of obesity on quality of life can be severe.⁷² Depression, an important comorbidity of obesity, may precede or result from obesity. Clinicians should look for flat affect, anxiety, body dissatisfaction, excess eating, fatigue, and difficulty sleeping. Sexual and physical abuse may increase the risk of severe obesity.^73,74 Youths with binge eating or purging behavior should be evaluated for eating disorders.

Orthopedic Disorders
Blount disease (tibia vara) occurs more often among obese children, and onset generally occurs after 8 years of age.⁷⁵ Often painless, Blount disease presents as visible bowing of the lower extremity and is diagnosed with anteroposterior radiographic views of the affected knee obtained while the patient is standing. An orthopedic surgeon can determine how to treat this condition, to correct bowing and to prevent progression.

Slipped capital femoral epiphysis occurs between 9 and 16 years of age, affects boys more often than girls, and has an incidence estimated at 11 cases per 100 000 children.⁷⁶ It occurs more frequently when a child is obese.⁷⁷ These children have hip or knee pain and pain with walking. On examination, hip range of motion is impaired. Bilateral frog-leg radiographic views of the hips should be obtained, and the child should be referred to the orthopedic surgery service.

A recent study revealed that overweight children and adolescents reported more fractures and musculoskeletal discomfort.⁷⁸ Because injury and pain interfere with physical activity, early intervention (including physical therapy, when indicated) may reduce weight gain in these children.

Skin Conditions
Acanthosis nigricans is present in 10% of obese white children and 50% of obese black children.⁷⁹ Although it is associated with hyperinsulinemia, acanthosis nigricans is associated more strongly with high BMI. The prominence of acanthosis nigricans diminishes with weight loss.

Severely obese children can have chronic irritation and infection in the folds of the skin, especially in the lower abdomen and axilla. This intertrigo and furunculosis requires good hygiene, use of topical antibiotic and anticandidal ointments, and sometimes systemic antibiotic therapy.

Genetic Syndromes
Well-defined genetic syndromes that cause obesity, such as Prader-Willi syndrome, are very rare. The assessment report lists some of these syndromes and their presentations. Clinicians should consider referral for genetic testing, especially when the obese child is short and has developmental delay. Unfortunately, diagnosis of these genetic syndromes does not modify treatment options.

Laboratory Testing
History and physical examination cannot effectively screen for abnormal cholesterol levels, NAFLD, and type 2 diabetes mellitus. Therefore, these conditions must be identified with laboratory tests. The expert committee recommends that children with BMI of 85th to 94th percentile should undergo lipid panel testing and, if risk factors are present, then fasting glucose, ALT, and AST levels should be measured every 2 years for individuals 10 years of age. For children with BMI of 95th percentile, the committee suggests that fasting glucose, ALT, and AST levels be measured every 2 years starting at 10 years of age, regardless of other risk factors. Elevation of ALT or AST levels above 60 U/L on 2 occasions may indicate the need for additional evaluation, probably with guidance from pediatric gastroenterology/hepatology experts.

The results of the primary care provider's history, physical examination, and screening laboratory tests may indicate the need for additional diagnostic tests. A table of more-specialized diagnostic testing to be performed after initial positive screening is presented in the assessment report.² Table 7 summarizes the medical assessment according to BMI category.

Behavior Assessment
Goals
The purpose of the behavior assessment is twofold. The first goal is to identify the child's dietary and physical activity behaviors that may promote energy imbalance and that are modifiable. The second goal is to assess the capacity of the patient and/or the patient's family to change some or all of these behaviors. Families must have both the means and the motivation to make changes. For instance, a child may benefit from increased outdoor play but, if no safe play area exists or if the parents do not perceive the benefit of this behavior change, then no change will occur and the child will "fail treatment." The clinician should work with the family to target behavior changes that are appropriate and possible.

Dietary and Physical Activity Assessments
Because comprehensive dietary and physical activity assessments, such as diet or physical activity diaries, are impractical in a typical office setting, the expert committee recommends a focused assessment of behaviors that have the strongest evidence for association with energy balance and that are modifiable. It should be noted that current evidence generally reveals an association between specific behaviors and energy consumption or expenditure or between a behavior and weight status, leaving the direction of the relationship unknown.

For eating behavior assessment, the following behaviors should be addressed:

• frequency of eating food prepared outside the home, including food in restaurants, school and work cafeterias, and fast food