What Constitutes Successful Weight Management in Adolescents? FREE

In this issue, 2 reports address the consequences and treatment of adolescent obesity. In the first study (1), van Dam and colleagues report that an elevated body mass index (BMI) based on recalled weight at age 18 years was associated with increased premature death in women enrolled in the Nurses' Health Study II. Mortality was increased among women who were obese (BMI ≥ 30 kg/m²) and those who were overweight (BMI, 25.0 to 29.9 kg/m²) at 18 years of age. In addition, in contrast with recently published data from the current National Health and Nutrition Examination Surveys (2), mortality was increased among women who were overweight and obese as adults. The impact of overweight and obesity at age 18 years on death persisted after control for various variables that are likely to alter the effect of body weight on death, such as smoking, alcohol use, physical activity, and BMI in adulthood (1). Although the reported effect of overweight on death has been controversial (2 - 5), few reports dispute the impact of adult obesity on adult diseases, such as diabetes and cardiovascular disease (6), or the increased risks for adult diseases associated with overweight in childhood.

By convention, I will use the term overweight to describe children and adolescents with a BMI in the 95th percentile for age and sex or greater (the 95th percentile in a young adult is equivalent to a BMI of 30 kg/m²). When I use the term obesity, I refer to the definitions used in the 2 studies.

Overweight in childhood is associated with an increased risk for remaining overweight in adulthood (7) and for becoming more severely overweight in adulthood (8). Overweight in adolescence is associated with increased early all-cause and coronary heart disease mortality rates in adult men and increased risks for coronary heart disease and atherosclerosis in both adult men and women (9).

These observations emphasize the importance of identifying and effectively treating overweight children. Although behavioral interventions directed at increased activity, reduced inactivity, and improved nutrition are associated with the lowest risks for adverse consequences, more aggressive interventions, such as pharmacotherapy, may be warranted for adolescents with comorbid conditions or those at highest risk for adult disease. However, few randomized trials have evaluated aggressive interventions, such as drug therapy, for overweight adolescents.

The report in this issue by Berkowitz and colleagues (10) from a 52-week multicenter placebo-controlled, randomized trial provides important data on the effect of sibutramine therapy on obesity in adolescents. Their data indicate that the addition of sibutramine to behavioral therapy for obese adolescents produced a mean reduction in BMI of 2.9 kg/m² compared with a reduction of 0.3 kg/m² in a group that received placebo plus behavioral therapy. The study cohort consisted of adolescents 12 to 16 years of age whose BMI for the diagnosis of obesity ranged from 2 units more than the 95th percentile for age and sex to 44 kg/m². A BMI that is 2 units more than the 95th percentile for age and sex is roughly equivalent to a BMI at the 97th percentile. Behavioral therapy was provided at 14 visits over 1 year. Study participants who received behavioral therapy plus sibutramine reduced their BMI from 36.1 kg/m² to 33.2 kg/m² (8.2% of starting BMI; mean weight loss, 5.9 kg), and 14.6% reduced their BMI to less than the 95th percentile.

In adults, loss of 10% of body weight improves many obesity-associated comorbid conditions, such as hypertension, dyslipidemia, and glucose intolerance (6). Furthermore, the Diabetes Prevention Program (11) demonstrated that a lifestyle intervention with a goal to achieve modest weight loss and increase physical activity in glucose-intolerant adults statistically significantly reduced the incidence of type 2 diabetes mellitus. Physical activity alone also improves glucose intolerance, as well as blood pressure and dyslipidemia (12). Because we know considerably less about the effect of weight loss or physical activity on obesity-associated comorbid conditions in children and adolescents, Berkowitz and colleagues' study provides welcome new information. Those who lost weight in the sibutramine group of the study had improved levels of triglycerides, high-density lipoprotein cholesterol, total cholesterol, insulin, and insulin sensitivity. Physical activity was a component of the behavioral intervention, but because the authors did not analyze the effect of behavioral intervention separately from that of sibutramine, it is difficult to know what proportion of the improvements in comorbid conditions was attributable to the weight loss and what proportion was attributable to increased physical activity.

Because 17% of adolescents 12 to 19 years of age have a BMI in the 95th percentile or greater (13) and because the prevalence of adolescent overweight has continued to increase, effective treatment for childhood overweight is a public health issue. The success of treatment in Berkowitz and colleagues' study raises a key but still unanswered question: Who needs drug therapy? My clinical experience suggests that the best approach is to stratify treatment by the severity of overweight and reserve drug therapy for very obese patients whose behavioral therapy fails. Within this context, we must still carefully weigh the decision to prescribe drug therapy, because the long-term risks or benefits of drug therapy in children and adolescents are unknown. The short-term benefits are clear: In Berkowitz and colleagues' study, the loss of almost 3 BMI units statistically significantly improved cardiovascular disease and diabetes risk factors. The long-term benefits are less clear: Adolescents who lose weight must maintain the weight loss as adults to reduce their risk for diabetes and cardiovascular disease. Presumably, sustained weight loss will also reduce the risk for increased death observed by van Dam and colleagues. Nonetheless, although overweight adolescents who lose weight may lower their biochemical risk factors for subsequent diseases, they may remain at some increased risk unless they achieve and maintain a healthy BMI. Therefore, as with other chronic diseases, obesity will require ongoing drug therapy, behavioral therapy, or both. Because drug therapy is likely to have higher lifetime risks and costs than behavioral interventions, physicians should aim for sustained behavior change and reserve drug therapy for the severely overweight adolescent who cannot otherwise lose weight.

Further analysis of Berkowitz and colleagues' study may point to more effective behavioral interventions. The behavioral interventions were center-specific rather than standardized across centers. Anonymous, center-specific data for both study groups (provided by Dr. Berkowitz) show that the center-specific changes in BMI varied several-fold across the centers in each study group. The behavioral techniques used by the most successful centers might provide useful insights to be tested in future clinical trials of weight loss in obese people and weight maintenance in normal-weight people.

Although effective weight loss strategies for overweight children and adolescents may reduce risks for obesity-associated diseases or the premature death observed by van Dam and colleagues, preventing overweight should be an even higher public health priority. Most children and adolescents are not overweight. As Berkowitz and colleagues have shown, weight loss requires a substantial investment of therapists' time and substantial drug costs. Prevention of excessive weight gain in those who are not yet overweight may be a more cost-effective approach to weight control in the pediatric population. Similar to tobacco control, prevention of childhood overweight will probably require a comprehensive, multidimensional approach. In contrast to tobacco control, which targets a single behavior, weight control should address both the intake and expenditure sides of the energy balance equation (14 - 15). Strategies for which reasonable scientific evidence exists include increased physical activity ((12), (16)) and breastfeeding (17) and decreased soft drink intake (18) and time spent viewing television (19). Other strategies to achieve caloric balance are under intense investigation.

Implementing weight control strategies requires efforts across several settings. Parents should limit access to calorically dense foods and control television-watching time. Clinicians should promote breastfeeding, reduced soft drink intake, reduced television time, and increased physical activity. Schools can restore physical education and provide nutritionally sound choices. Communities should increase access to recreational facilities. These multisectoral and multicomponent strategies to prevent weight gain in children and adolescents may also help to sustain the weight losses that Berkowitz and colleagues worked so hard to achieve in obese adolescents.

William H. Dietz, MD, PhD

Centers for Disease Control and Prevention

Atlanta, GA 30341