Section
7
Measuring Body Composition in Population Health Research: Case Studies
Case Study 6 A Clinic-based Intervention to Promote Weight Loss in Adolescents with Severe Obesity
Background
A research grant is awarded to a team of clinical investigators to conduct a randomized controlled trial to evaluate the effects of Liraglutide, a glucagon-like peptide-1 receptor agonist drug, when compared to a lifestyle counseling intervention to promote weight loss in adolescents aged 13–18 years with severe obesity. The trial is being implemented in a multidisciplinary weight management clinic located in a tertiary care hospital. The project team would like to evaluate the extent to which the drug leads to greater weight loss over 52 weeks compared to the lifestyle intervention. The team must also assess relevant safety parameters.
Considerations
The clinical investigators expect that treatment with the drug will result in significant weight loss, which raises concerns about normal growth in this study population. Many adolescents are undergoing puberty and rapid growth during these ages, therefore normal growth in height/stature and normal bone development are critical to monitor. In addition, given that the study is targeting significant weight loss, it is important to assess whether favorable changes in body composition occur during growth (fat mass relative to muscle mass) so that muscle mass is maintained or not adversely affected while loss of fat mass takes place.
Adolescent females face additional considerations. Girls who are sexually active and not using a reliable form of contraception may be at risk of pregnancy. Therefore, they need to be excluded from both the drug and lifestyle interventions due to weight gain associated with pregnancy and because the drug intervention may put the fetus at risk of harm.
The study needs to assess changes in body weight and composition, and the measure of body composition needs to be conducted at baseline and at multiple time follow-up points. Therefore, the measure needs to be sensitive to detecting changes in body composition over time.
Although the multiple assessments will add to study costs and participant burden, smaller samples of participants are needed in comparison to prevention trials because effect sizes are expected to be larger between the drug and lifestyle treatments. In this trial, the study staff anticipate about a 10% reduction in BMI, on average, in the drug + lifestyle arm and a 5% in the lifestyle only arm. This trial requires about 100 adolescents per study arm at completion as compared to hundreds or thousands of participants in population-based studies focused on the prevention of obesity. Therefore, costs of more expensive and accurate body composition methods may be less prohibitive.
Unlike research conducted in community settings, the clinical setting offers privacy and exam rooms for research teams to conduct measurements. Participants can wear hospital gowns to facilitate examination. Even so, adolescents who have severe obesity and are developing secondary sex characteristics, may feel embarrassed during measurements or examinations. They often prefer to wear their own loose-fitting clothing with minimal to no intrusive physical measurements.
In addition, preparation for clinic visits may be difficult (such as fasting or limited physical activity), and visits need to avoid time away from school and are often scheduled in the afternoons after school. Acceptability of the measures with regard to safety, burden, time of day, and intrusiveness to both parents who will provide consent and adolescents who will need to assent is a key consideration.
Method Selection
Given the considerations outlined above, the researchers agree it will be important to have a method, such as DXA, that can separate body weight into fat mass and fat-free mass as well as measure bone density. The hospital in which the clinic is located offers the availability of sophisticated measures of body composition, such as DXA, and well-trained staff/technicians to conduct the measurements. DXA costs are higher than other measurement methods, but the relatively small samples of adolescents may make it feasible. The test takes about 15 minutes and the adolescent can remain fully clothed. On the other hand, DXA machines have size limitations that may make it infeasible to measure participants with extreme obesity.
The research team also debates other methods that can estimate whole-body fat mass and fat-free mass without radiation exposure, such as ADP (BOD POD) and bioelectrical impedance analysis (BIA), if they choose not to measure bone density. The BOD POD assessment is simple and less expensive than DXA and is fast and safe; however, special compression clothing is necessary. Teens with severe obesity may be deterred by this requirement or find it uncomfortable. Participants also need to fast, which may be feasible in the morning, but morning visits require missing school. Similarly, BIA is non-invasive, safe, quick, and relatively inexpensive but use of BIA requires standardized participant conditions, such as fasting and lack of strenuous activity for 12 hours before the test, which are difficult to implement. In addition, prediction equations used to estimate fat-free mass and fat mass are population specific. Therefore, equations for estimating fat-free mass in the study population would have to be available for the estimates to be valid.
Ultimately, the investigators select DXA as the method for this study because it separates body weight into fat mass and fat-free mass as well as measures bone density. With many adolescents in the study expected to be undergoing puberty and rapid growth during these ages, the investigators conclude that the benefits of monitoring body composition and bone development outweigh potential radiation risks. The investigators also opt to include BMI as a study method. Even though BMI does not provide information on body composition, specifically fat and fat-free mass, this method will be included in the trial because it can be used to monitor normal growth and to assess the efficacy of the weight loss intervention as a secondary outcome. Because BMI is used extensively in assessments of other weight management interventions, it also would be useful as a common measure for comparison to these other trials.
In terms of outcomes, the study team will be able to compare the changes from baseline to end of intervention in whole body fat or BMI outcome between the two study arms. The team also may examine changes in the percentage of participants classified as having obesity or severe obesity (using BMI percentiles) from baseline to the end of the intervention.