American Heart Association Childhood Obesity Research Summit Report

Abstract

HomeCirculationVol. 119, No. 15American Heart Association Childhood Obesity Research Summit Report Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBAmerican Heart Association Childhood Obesity Research Summit Report Stephen R. Daniels, Marc S. Jacobson, Brian W. McCrindle, Robert H. Eckel and Brigid McHugh Sanner Stephen R. DanielsStephen R. Daniels Search for more papers by this author , Marc S. JacobsonMarc S. Jacobson Search for more papers by this author , Brian W. McCrindleBrian W. McCrindle Search for more papers by this author , Robert H. EckelRobert H. Eckel Search for more papers by this author and Brigid McHugh SannerBrigid McHugh Sanner Search for more papers by this author Originally published30 Mar 2009https://doi.org/10.1161/CIRCULATIONAHA.109.192216Circulation. 2009;119:e489–e517is corrected byCorrectionOther version(s) of this articleYou are viewing the most recent version of this article. Previous versions: March 30, 2009: Previous Version 1 TABLE OF CONTENTSI. The Problem: Childhood Obesity— A Burgeoning Epidemice490II. Childhood Obesity Research Summit…e490III. Pathophysiology and Morbidity of Childhood Obesity…e490 A. Pathophysiology of Obesity…e490 B. Medical Consequences of Childhood Obesity…e490 C. Psychosocial and Societal Consequences of Childhood Obesity…e490IV. Current Healthcare Practices in Childhood Obesity Prevention and Treatment…e492 A. Lifestyle Assessment: Diet and Physical Activity…e492 B. Clinical Assessment…e493 C. Psychosocial/Behavioral Assessment…e493 D. Implementing Effective Prevention and Treatment Options: Diet…e494 E. Implementing Effective Prevention and Treatment Options: Exercise and Sedentary Behaviors…e495 F. Implementing Effective Prevention Treatment Options: Family and Behavioral Approaches…e496 G. Behavioral Management…e497 H. Medical and Pharmacological Management…e497 I. Surgical Management…e498V. Barriers to Optimum Care…e498 A. Healthcare System Factors…e498 Preparedness of Providers…e498 Models of Coverage and Care…e499 Uninsurance and Underinsurance Among Children and Families…e500 Healthcare Providers and Settings as Role Models…e500 B. The Challenge of Behavior Change…e501 Motivational Interviewing for Pediatric Obesity…e501 Psychosocial Factors Influencing Behavior Change…e502 Environmental Factors Influencing Behavioral Change…e503VI. Practice-Based Resources for Prevention and Management…e503 A. Models From Other Disciplines…e503 Adult Chronic Care Model…e503 Treatment of Heart Failure…e504 Treatment of Type 2 Diabetes Mellitus…e504 Treatment of Childhood Asthma…e505 B. Public Health Measures for Childhood Obesity…e506 Tobacco Control: Implications for Childhood Obesity…e506 C. Specific Research Issues…e506 Specifying Priorities in the Context of a Complex, Multifactorial Problem…e506 Taking a Developmental Approach…e507 Engaging Stakeholders…e507 Targeting the Appropriate Outcome…e508VII. Research Challenges…e509 A. Research in the Busy Practice…e509 B. Bring Research Into Practices: Web-Based Assessment for the Pediatric Obesity Clinic…e509 C. Research Agenda: NHLBI and the Pediatric Heart Network…e510 D. Research Agenda: The Children’s Oncology Group…e510 E. Research Agenda: CDC…e511 F. Research Agenda: Foundations…e511VIII. Research and Policy: Knowledge Translation…e512 A. From Practice to Policy to Practice…e512IX. Concluding Remarks…e512Appendix A: Planning Committee…e513Appendix B: Presenters…e513Over the past 3 decades, the prevalence of obesity has increased among children of all ages. Today, one third of American children and adolescents are obese or overweight. Childhood obesity is one of the most pressing health threats facing the United States.I. The Problem: Childhood Obesity—A Burgeoning EpidemicOver the past 30 years, the prevalence of obesity has nearly tripled for children 2 to 5 years of age and youth 12 to 19 years of age, and it has quadrupled for children 6 to 11 years old.1 Data from 2 National Health and Nutrition Examination Surveys (1976 to 1980 and 2003 to 2004) show that for children 2 to 5 years of age, the prevalence of overweight increased from 5% to 13.9%; for those 6 to 11 years of age, prevalence increased from 6.5% to 18.8%; and for those 12 to 19 years of age, prevalence increased from 5.0% to 17.4%.2Overweight children and adolescents are at risk for significant health problems both during their youth and as adults. For example, during their youth, overweight children and adolescents are more likely than other children and adolescents to have risk factors associated with cardiovascular disease (eg, high blood pressure, high cholesterol, and type 2 diabetes mellitus).3 Overweight children and adolescents are also more likely to become obese as adults.4,5 For example, 1 study found that approximately 80% of children who were overweight at 10 to 15 years of age were obese adults at 25 years of age.4 Another study found that 25% of obese adults were overweight as children.3 The latter study also found that if overweight begins before 8 years of age, obesity in adulthood is likely to be more severe. Studies have also documented the link between obesity and poor school performance and unhealthy or risky behaviors such as alcohol use, tobacco use, premature sexual behavior, inappropriate dieting practices, and physical inactivity.6,7Overweight children and adolescents may experience other health conditions associated with increased weight, which include asthma, hepatic steatosis, sleep apnea, and type 2 diabetes mellitus. Obesity also puts children at long-term higher risk for other debilitating chronic conditions such as stroke; breast, colon, and kidney cancers; musculoskeletal disorders; and gall bladder disease.Obesity-associated annual hospital costs for children and youth have more than tripled over 2 decades, rising from $35 million in 1979 to 1981 to $127 million in 1997 to 1999. After adjustment for inflation and conversion to 2004 dollars, the national healthcare expenditures related to obesity and overweight in adults alone range from $98 billion to $129 billion annually.8 The obesity crisis and the public health threat it represents mandate immediate action from the healthcare community to provide clinical intervention and to serve as advocates for the development of community programs that help teach and support healthy behaviors.II. Childhood Obesity Research SummitAs part of its strategic focus on childhood obesity, and particularly the prevention of childhood obesity, the American Heart Association convened a Childhood Obesity Research Summit to examine research opportunities where the obesity epidemic intersects with the healthcare system. The outcomes of the conference help inform the American Heart Association and the Alliance for a Healthy Generation broadly in terms that help to focus ongoing research and public policy initiatives.III. Pathophysiology and Morbidity of Childhood ObesityA. Pathophysiology of ObesityThe pathophysiology of obesity in childhood can be viewed from 3 perspectives: a homeostatic or energy-balance perspective, an epidemiological perspective, and a pathological perspective examining the consequences of excess fat on risk for future disease. The fundamental cause of obesity is a greater imbalance between energy intake and expenditure than is expected for normal growth and development. Usually, this occurs over a period of time and in the setting of a susceptible genetic background and environmental factors. Epigenetic factors, defined as the changes in gene function that do not relate to changes in DNA sequence, begin in utero (or in some cases in previous generations) and also contribute.9 Infants of diabetic mothers and of mothers who smoke during pregnancy have increased risk of subsequent obesity. Infant feeding practices may also play a role, particularly a shortened period of breast-feeding. A reduced amount of sleep during infancy is another potential risk factor for obesity. Some medications have been clearly demonstrated to cause excess weight gain.B. Medical Consequences of Childhood ObesityThe increasing prevalence and severity of obesity in children and adolescents have resulted in a higher prevalence of comorbid conditions, including high blood pressure, early development of atherosclerosis, type 2 diabetes mellitus, nonalcoholic fatty liver disease, polycystic ovary disorder, and disordered breathing during sleep.10 These complications can occur both in the short-term and in the long-term. Some complications, such as type 2 diabetes mellitus, previously thought to only occur in adulthood have now been shown to occur in children and adolescents. This has raised concerns about whether the obesity epidemic might shorten the lifespan of the current generation of children.Olshansky et al11 evaluated the potential effect of childhood obesity on lifespan. Their analysis predicts a shorter lifespan for the current generation of children, in large part because of obesity and its related comorbidities, including cardiovascular disease and metabolic, gastrointestinal, pulmonary, orthopedic, neurological, psychological, and social disorders. However, others have speculated that improved medical care will ameliorate these effects.There is still much to be learned about the mechanisms for obesity development and the related comorbid conditions. Translational research perhaps will lead to information that will help to identify children at higher risk for excessive weight gain and for the development of specific adverse outcomes due to obesity. This type of mechanistic research could also inform more specific strategies for treatment of comorbidities when weight management cannot be accomplished or is less than adequate.Selected Unanswered Questions Related to the Medical Consequences of Childhood ObesityWho is at risk for development of the medical consequences of childhood obesity?How soon should surveillance of children begin to identify obesity-related comorbidities?When weight loss is inadequate or fails to modify the medical consequences of obesity, how aggressive should healthcare providers be in the institution of pharmaceutical agents to treat these consequences?Under what conditions should children with obesity-related comorbidities be evaluated or managed by subspecialists?C. Psychosocial and Societal Consequences of Childhood ObesityHistorically, studies to evaluate the psychosocial consequences of pediatric overweight have focused primarily on comparing the functioning of overweight children and adolescents with that of nonoverweight peers on dimensions of health-related quality of life, self-concept, body image, depressive symptomology, and peer relationships. Cross-sectional studies comparing either clinical or community samples of overweight children and adolescents with nonoverweight comparison samples demonstrate fairly consistent support for decreased health-related quality of life associated with increasing weight, with those with the highest weight demonstrating the greatest impairment in health-related quality of life. There is also evidence indicating lower body satisfaction and physical appearance–related self-concept in overweight children and adolescents.12 Longitudinal studies suggest decreases in self-esteem from childhood to early adolescence among overweight children,13 and there is some evidence from prospective studies indicating increased risk for development of obesity among depressed adolescents.14 Overweight children experience more teasing and both overt and relational victimization than normal-weight peers. Longitudinal research suggests that weight status is predictive of vulnerability to bullying in peer relationships.15,16 Finally, some studies document a relationship between overweight status and decreased probability of employment and less financial support for college among women, as well as lower household incomes for both men and women. Findings from other domains of psychosocial function, including depressive symptomatology, externalizing behaviors, and self-concept, are more heterogeneous, and even when samples of overweight children and adolescents yield lower scores, they are often still in the clinically normal range.More recent studies have focused on understanding the variability in outcomes among overweight children and adolescents, with the potential for identifying risk and protective factors. Social stigma and weight-based teasing have been identified as key dimensions that may increase risk for negative psychosocial outcomes among overweight pediatric populations.17 Cross-sectional and longitudinal studies have demonstrated a relationship between weight-based teasing and a number of negative psychosocial outcomes, including unhealthy weight-control behaviors, decreased body satisfaction, decreased self-concept, and depressive symptoms. Body dissatisfaction and weight/shape concerns have also been associated with lower self-esteem, as well as higher depression and anxiety scores. A final set of findings with regard to psychosocial consequences relates to changes observed through participation in weight-control interventions. Improvements in psychosocial functioning have been observed in children and adolescents participating in weight-control treatments. These changes are typically independent of weight-loss outcomes.Selected Unanswered Questions Related to Psychosocial and Societal Consequences of Childhood ObesityWith regard to severely overweight youth, to what extent do negative psychosocial outcomes, such as depressive symptoms and disordered eating behaviors, relate to degree of overweight, and to what extent do these dimensions improve as a result of intensive weight-control interventions?What role do moderators such as age, gender, ethnicity, and socioeconomic status play in the relationship between overweight and psychosocial outcomes?To what extent do cultural norms regarding weight status and the associated stigma of overweight impact psychosocial outcomes such as self-esteem and body satisfaction?What protective factors, such as degree of family connectedness and family behaviors, buffer the potential negative psychosocial consequences of pediatric overweight?To what extent can protective factors be enhanced through intervention?What impact do psychosocial outcomes, such as social stigma, have on weight status, risk behaviors, and health outcomes?RecommendationsEducation: Given the heterogeneity of outcomes for overweight youth, it is important for professionals to objectively evaluate psychological and social correlates and not assume maladjustment in these patients. A second potential area for professional education relates to the importance of efforts to decrease weight-related teasing and stigma and enhance self-concept in overweight children and adolescents.Public policy/advocacy: Policies to decrease tolerance for teasing and stigma in schools may improve the emotional climate for overweight children and adolescents. Education of the public regarding physiological aspects of overweight may enhance these efforts.Research: (1) The focus should be on severely obese youth, with attention to documentation of emotional and social functioning, as well as change in psychosocial correlates with decreases in obesity. (2) Moderators including gender, race, ethnicity, and socioeconomic status should be examined in the relationship between weight status and psychosocial outcomes. (3) Longitudinal studies to examine relationships among weight status, psychosocial correlates, and other health risk behaviors would be informative in elucidating temporal relationships and identifying intervention targets. (4) A research agenda concerning weight stigma as it relates to psychosocial, academic, and physical health outcomes should be pursued.IV. Current Healthcare Practices in Childhood Obesity Prevention and TreatmentA. Lifestyle Assessment: Diet and Physical ActivityLimited data have been published to assess the specific goals that pediatric primary care providers hope to accomplish when performing lifestyle assessment of diet and physical activity; however, routine care and anecdotal reports suggest that diet and physical activity assessments are usually the first step in counseling in the primary care setting. In routine care, assessment questions are intended to bring up the issue and give families an opportunity to ask questions or raise concerns. Lifestyle assessment is also an opportunity to identify potential targets for prevention and increase families’ self-awareness of current behaviors. The situation may differ between routine well-child care visits and visits aimed at addressing a specific health issue, with childhood obesity lifestyle assessment requiring specific and more in-depth questions. In preparing for diet and physical activity counseling, it is helpful to assess the current situation to determine whether the patient meets recommendations, not only in areas of obesity prevention or treatment but also in other areas of pediatric nutrition.Some published data address how frequently lifestyle assessment is performed in the pediatric primary care setting. In response to a mailed questionnaire, in a random sample of pediatricians, nurse practitioners, and dieticians, more than 90% reported routine assessment of diet history, and more than 95% reported assessment of physical activity.18 In a separate study, 56% of family practice physicians and 71% of pediatricians reported routinely performing diet recall; 54% and 79%, respectively, reported assessing sedentary activities; and 88% and 98%, respectively, reported assessing exercise and sports.19 The main limitations of these and similar studies include the usual low response rate achieved in questionnaires of healthcare professionals and, even more concerning, the likely bias that results because providers who perform lifestyle assessments are more likely to respond, thus overestimating how frequently such assessment actually occurs.Even more important than how frequently assessment is performed is how it is actually conducted. One study reports that pediatricians assess diet using questions about usual or typical food intake (33.5%), diet diaries (31.2%), 1-day diet recall (15.3%), or eating practices and patterns (14.1%). They assess physical activity using questions about organized physical activity (97.4%), sedentary activities (93.6%), or unstructured physical activities (91.6%).14 Because dietary or physical activity assessments are not typically taught in medical or other health-profession schools, a wide variability exists in how they are performed in primary care. Most assessment methods are not standardized, and it remains unclear how often standardized assessment methods are used.Several tools have been developed in the research setting to assess dietary intake that could potentially be used in the primary care setting. These include weighted food records, semiquantitative food records, and standardized diet history, all of which are reliable but likely unfeasible in the primary care setting. Less reliable but likely more feasible methods include informal diet history and standardized or informal food-frequency questionnaires. For physical activity research, methods of assessment include doubly labeled water and calorimetry, heart rate monitors, and accelerometers. Perhaps more feasible in the primary care setting are pedometers, physical and sedentary activity diaries, and formal validated or informal physical activity questionnaires. Feasibility issues in the primary care setting include time constraints, cost, training, and expertise in the use of these methods. Emerging technologies may provide novel methods of diet and physical activity assessment that potentially may be helpful in the primary care setting. These include Internet-based diet and physical activity assessment; PDA (personal digital assistant) or cell phone–based dietary assessment; accelerometers or motion sensors in cell phones, clothing, or shoes; and use of a GPS (global positioning system) to assess physical activity. Another interesting concept is the placement of risk-assessment (including diet and physical activity) kiosks in the waiting rooms of primary care practices to allow families to provide important and standardized information that can be analyzed and used by the clinician in the visit after the assessment.Selected Unanswered Questions Related to Lifestyle Assessment: Diet and Physical ActivityWhich goals in lifestyle assessment should be prioritized in the pediatric primary care setting? This question should be addressed by assessing the feasibility of each method in terms of time, cost, and expertise, as well as by assessing the expected benefits in terms of quality and usefulness of the information collected and its utility as a first step in counseling.How is lifestyle assessment currently conducted in the pediatric primary care setting? Because of the issues related to response rate and bias, as well as issues of social desirability in responses, methods other than questionnaires should be considered, including direct observation or patient interviews.What are the feasibility and the validity of tools currently used in research settings in the office setting? More promising tools for the clinical setting include targeted food-frequency questionnaires, brief physical activity questionnaires, and pedometers.How can emerging technologies be used in the primary care setting to assess diet and physical activity?Can lifestyle assessment without effective counseling have a potentially negative impact?RecommendationsEducation: Pediatric primary care providers should be trained in brief, validated methods of diet and physical activity assessment.Public policy/advocacy: Reimbursement of diet and physical assessments in the pediatric primary care setting should be promoted and supported.Research: More research is needed in the areas of clinical and psychosocial/behavioral assessment. The clinical effectiveness and cost-effectiveness of physicians versus dieticians, nutritionists, and certified diabetes educators in lifestyle assessment should be examined.Prevention in office-based settings: Effective prevention and treatment options related to diet, exercise, sedentary behaviors, and family and behavioral approaches should be implemented, as well as management of obesity and morbid obesity and behavioral, medical, pharmacological, and surgical management.B. Clinical AssessmentEvaluation of obesity begins with calculation of body mass index (BMI), which has clinical validity because it correlates with adiposity,20 adult adiposity,21 cardiovascular risk factors,22 and long-term mortality.23 Unfortunately, there is no perfect cut point for BMI that identifies all children with elevated body fat. BMI correlates with total fat; however, there may be as yet unidentified or unquantified race and gender differences that need to be considered.Because BMI norms change with age and differ between boys and girls, absolute BMI is not an appropriate screen in children. Practitioners need to plot BMI on the Centers for Disease Control and Prevention (CDC) percentile curves to identify the BMI percentile category (www.cdc.gov/growthcharts). The BMI z score is an appropriate alternative but requires further calculation, which is cumbersome for busy pediatricians to perform. Skinfold-thickness measurement to confirm increased body fat is not currently recommended, because accurate, reproducible measures are technically difficult to perform. Waist circumference is also not recommended until research establishes how this measure adds information about risk to that already provided by BMI.Clinicians should assess obesity risk in all children, integrating information about the patient’s BMI, medical risk, and current eating and physical activity behaviors and attitudes. This assessment should guide initiation of preventive or treatment strategies. Current definitions use the 85th percentile to define overweight and the 95th percentile to define obesity.23aSelected Unanswered Questions Related to Clinical AssessmentEvaluation of current recommendations implementation is needed. This evaluation should determine which recommendations (eg, National Heart, Lung, and Blood Institute [NHLBI] and the American Academy of Pediatrics) are being implemented, the barriers to implementation, and the costs and benefits of implementing screening practices.Should screening and management differ by level of obesity? In the patient with mild BMI elevation, how should clinical information affect management, and what additional measures would help delineate risk (such as visceral adiposity assessment)? Is early treatment more successful? When should intervention take place?For patients with BMI >99th percentile, should the evaluation be modified?RecommendationsEducation: Primary care providers and third-party payers should be educated about existing guidelines, as well as about the costs and benefits of screening.Public policy/advocacy: Systems to coordinate screening efforts among various healthcare providers, such as obstetricians, pediatric physicians, and adult physicians, should be developed to promote BMI screening across the lifetime of individuals.Research: (1) Barriers to implementation of screening guidelines should be studied. (2) Screening guidelines should be evaluated on the basis of severity. (3) Research to collect more epidemiological data on anthropometrics early in life (normative data on length for weight) should be funded. (4) More epidemiological data are needed on early childhood obesity, including normative data on high weight-for-length and risk for persistence and poor health outcomes. (5) More data are needed on the success of various management strategies.General: A network of obesity researchers, educators, clinicians, and policy makers should be developed to increase communication and the sharing of tools and techniques for performing research.C. Psychosocial/Behavioral AssessmentCurrent practices for psychosocial screening assessments include quality-of-life assessments, readiness assessments, and family-context assessments. From studies of psychosocial assessment, it is known that the prevalence of clinical-range depression symptoms is low, although there is a substantial prevalence of a subclinical range of depressive symptoms; there is a low prevalence of comorbid eating disorders; and impairments such as poor health-related quality of life, social isolation and stigmatization, adolescent low self-esteem and poor body image, and maternal distress are prevalent.The available data indicate that greater depressive symptoms at baseline are associated with higher dropout rates in a clinical program,24 and greater parent distress and child social problems at baseline are associated with poorer weight-loss outcomes.25 The existence of fewer social problems at baseline predicted better long-term maintenance of weight loss at 2 years,26 and improvements in weight status were shown to be associated with improvements in the child’s social problems and psychological adjustment, with reduction in maternal distress.27 In addition, participation in intervention alone has a positive effect on self-esteem.28To help clinicians tailor treatment approaches, randomized clinical trials are needed on the appropriateness of psychological treatment before clinical treatment, referral for concurrent psychological and clinical treatment, or treatment of psychosocial issues within a weight-management model. In addition, an understanding of the readiness to change in a family can be useful. This can be accomplished in the framework of “stages of change,” with assessment of where the parent and child are on a continuum of readiness to make lifestyle changes. Motivational interviewing techniques may be helpful to move patients and families toward greater readiness. Parents need to be better educated about the definition and health consequences of obesity. Currently, there are no well-validated measures in widespread use to evaluate readiness to change. Studies of the potential for harm from intervention in a family that is not ready for an intervention are also needed.The final area that requires evaluation is the family structure. A family assessment is recommended to determine who will do what in providing/monitoring care, what the favored parenting style is, and how the family functions. The existing literature suggests that higher family conflict/lower cohesion, ineffective parenting style, and maternal distress are more prevalent in families of children who present for specialized obesity treatment. An improved understanding is needed in relation to how family/parenting factors bring about lifestyle change, whether family/parenting factors lead to obesity development, and the best methods to assess and monitor patients in a clinical setting.Selected Unanswered Questions Related to Psychosocial/Behavioral AssessmentHow should psychosocial assessment be performed? Should patients be referred for psychological treatment first or be referred for concurrent psychological treatment after BMI has been measured and interpreted, or should psychosocial issues be treated within a weight-management model?How does psychosocial impairment pose barriers to adherence to a weight-management prescription?Can obesity-specific instruments to assess child/adolescent lifestyle behavior and the psychometrics of these psychosocial instruments be developed and evaluated?Recommendation