Ihuoma Eneli, associate director, American Academy of Pediatrics Institute for Healthy Childhood Weight, and director, Center for Healthy Weight and Nutrition, Nationwide Children’s Hospital, began her presentation by reporting that 24.5 million children in the United States have
overweight, obesity, or severe obesity. According to the U.S. Census, this amounts to one in four children. These children, she observed, are at risk for multiple serious comorbidities that can affect virtually all organ systems. She added that children with obesity have an increased likelihood of having obesity as adults.
Drawing largely on meta-analyses and systematic reviews, Eneli reviewed the treatment of obesity in children and adolescents. In 2007, she noted, an expert committee of the American Medical Association (AMA), Health Resources and Services Administration, and Centers for Disease Control and Prevention issued recommendations on management of childhood obesity that were divided into four stages (Barlow, 2007). In the first and second stages, treatment occurs in the primary care provider’s office, delivered by a pediatrician or nurse practitioner. In the second stage, an allied health care provider is also involved. In the third and fourth stages, a multidisciplinary care team provides more frequent and intensive care focused on lifestyle modification. The fourth stage, Eneli said, involves bariatric surgery, very low-calorie diets, and medications.
With respect to analyzing what works, Eneli distinguished four categories of excess weight: severe obesity, obesity, overweight, and what she called “crossing weight percentiles rapidly.” In 2016, she noted, the U.S. Preventive Services Task Force (USPSTF), updating a 2010 set of recommendations (Whitlock et al., 2010), issued a recommendation that clinicians screen children aged 6 years and older for obesity and offer them or refer them to comprehensive, intensive behavioral treatment aimed at promoting improvement in weight status (USPSTF, 2017). The task force assigned this recommendation a grade of B, Eneli elaborated, meaning that it has moderate certainty of providing a substantial or moderate net benefit. The recommended screening test was BMI, she said, with no recommended interval.
As Eneli noted later in response to a question, every outcome measure has limitations. Nonetheless, she selected a reduction of 0.2 in BMI z-score as a standardized measure of treatment effectiveness, the measure used in the 2016 USPSTF recommendations (O’Connor et al., 2016; Wiegand et al., 2014). The BMI z-score, she explained, allows for the comparison of children of different ages and interventions in different sectors. Studies have suggested that a reduction of 0.15–0.25 in BMI z-score is associated with improved cardiometabolic outcomes, she said. Another study found that a reduction in BMI z-score of 0.2 is equivalent to about a 5 percent decrease in body weight, a benchmark used in the adult population, she noted (see Chapter 2).
Eneli continued by observing that in the 2016 USPSTF draft summary, meta-analyses were grouped by contact hours: 0–5, 6–25, 26–51, and more than 52 hours, in most cases over a 12-month period. She reported that lifestyle behavioral treatments for children that exceeded 52 contact hours over that period generally produced an effective change in BMI z-score, whereas treatments lasting 26–51 hours produced a lower effective change. She explained that these treatments included group sessions, individual sessions, parent-only sessions, child-only sessions, and combinations of these approaches. She added that the interventions frequently involved referral to specialty clinics rather than being primary care–based programs. She noted that they included information on nutrition; physical activity; and behavior modification, such as goal setting, stimulus control, and problem solving.
For children younger than 6, Eneli continued, a Cochrane review (Colquitt et al., 2016) showed that lifestyle interventions are effective in reducing weight, with changes as high as 0.38 in BMI z-score 12–18 months from baseline being seen. For comparison, she noted that an 8-year-old at the 50th percentile in height and the 95th percentile in weight would experience a decline of 0.6 in BMI z-score upon moving into the 85th percentile in weight, which would be equivalent to about an 8.6-pound weight loss. In that case, she observed, a 0.2 decrease in BMI z-score is equivalent to about a 3.3-pound weight loss. Furthermore, she said, because an 8-year-old is still growing, that child could gain 6 pounds over 12 months and still reduce his or her BMI z-score by 0.2 by virtue of gaining 2.5 inches in height. With a 16-year-old who is almost done growing, she added, a decrease of 0.2 in BMI z-score represents about a 9-pound weight loss, or about 5 pounds over 12 months.
According to the 2016 USPSTF report, Eneli continued, weight loss had mixed effects on cardiometabolic outcomes. Blood pressure improved somewhat, as did glucose metabolism, but lipid panel results showed little improvement. However, Eneli said, Ho and colleagues (2012, 2013) found positive effects on the lipid panel, fasting blood sugar, and blood pressure with a decrease in BMI z-score of 0.2–0.3. She added that all studies including bariatric surgery showed a clear benefit of this intervention for cardiometabolic outcomes.
In the studies that reported psychosocial outcomes, Eneli elaborated, comparison of intervention and control groups revealed no consistent differences in quality of life, self-esteem, or depression, according to the USPSTF. On the positive side, she added, children who received the intervention compared with the control group showed no increase in disordered eating. She asserted that effectiveness measures would be improved if psychosocial outcomes could be measured objectively and incorporated.
Other studies, Eneli noted, have looked at ethnicity, sample age, intervention length, intervention structure, and effectiveness measures. Summarizing these studies, she observed that interventions appear to be equally effective for Latinos, African Americans, and non-Hispanic whites. Intervention length does matter, she added, in that a 12-week intervention appears to be inadequate, and most studies looked at 6-month interventions at a minimum. Data on the effectiveness of parent-only versus parent-and-child interventions are inconclusive, she said.
Evidence on the effectiveness of working in the primary care setting also is mixed, Eneli reported. Mitchell and colleagues (2016) found a small effect size in terms of improving weight status, with a larger effect from more treatment contact and from including the pediatrician in that contact. However, Eneli said, a more recent meta-analysis by Sim and colleagues (2016) found a marginal effect. “The way they described this marginal effect,” she explained, “is that if you compare a 10-year-old at the 90th BMI percentile in the control group with a 10-year-old peer in the intervention group, what you would find is a 1 kilogram difference in weight over 3 years.”
Eneli noted that no evidence currently supports the ability of school-based interventions to decrease BMI z-score. “It can be an adjunct treatment,” she said, “but just solely on its own, there is no evidence.” She reported that the Mind, Exercise, Nutrition, Do It! (MEND) program, which the YMCA is adapting to community settings, involved more than 52 hours of contact and showed a decrease of 0.2 in BMI z-score at 6 months that carried through to 12 months (Sacher et al., 2010).
Specialized diets for children who have severe obesity can be effective, noted Eneli. As an example, she cited a high-protein, low-carbohydrate diet that decreased BMI z-score by an average of 0.25 at 12 weeks (Krebs et al., 2010). She added that the protein-sparing modified fast, a very low-carbohydrate diet including 20–40 grams of carbohydrates a day, is used for specific conditions (Ebbeling et al., 2003; Kirk et al., 2012; Schwingshackl et al., 2015). In one case she described, a 16-year-old with a BMI of 99 who was put on the diet decreased his BMI to 85 before undergoing surgery (Watowicz et al., 2015). “There are opportunities to look at some of these treatment options,” she said.
Eneli reported that information on the use of obesity medications with children is sparse compared with their use in adults. The published trials, she noted, show a small effect (Mead et al., 2016), but the effect is not as large as for lifestyle interventions. “There is a lot of work to be done,” she asserted. “There are no good answers. That is why we are envious of our
adult population. It would be nice to look at drugs. It would be nice to look at a combination of drugs and lifestyle.”
With regard to weight loss surgery, Eneli observed, the gastric sleeve is now the most common such surgery performed in the pediatric population, as it is in the adult population. In the Teen-LABS study, she noted, 97 percent of participants had a BMI of 40 or higher at baseline, while 3 years later, only 37 percent of them had that high of a BMI (Inge et al., 2016). “The median decrease with bariatric surgery is 30 percent,” Eneli said. “That brings up [the question of] what age do we do it?”
Extrapolating from a study by Steele (2013), Eneli noted that, despite its effectiveness, weight loss surgery occurs in only about 1,200 children per year. “That means that we still have 24.5 million children who get lifestyle interventions or medical treatments,” she said, “and a large proportion of them get no treatment at all.”
According to Eneli, one prominent research need is for longer followup studies. For example, she said, one long-term study showed that approximately 48 percent of participants no longer had obesity after 10 years (Epstein et al., 1994). “There are possibilities,” she stated. “We just need more studies.”
Eneli cited additional potential opportunities, including using the right effectiveness measure, looking at a set of effectiveness measures rather than a single measure, making effectiveness measures more relevant to stakeholders in the field, providing better training, supporting care in clinical and community settings, focusing on children with severe obesity as well as on children who are rapidly crossing percentiles, integrating programs across settings, supporting extended research funding, integrating technology, and linking outcomes to policy on reimbursement and population health (Dietz et al., 2015; Foltz et al., 2015). “We have to find a consistent and systematic way in which [care] is paid for,” she insisted, “because our children deserve it.”
This page intentionally left blank.