Section
7
Measuring Body Composition in Population Health Research: Case Studies
Case Study 3 Effect of Maternal Gestational Weight Gain on Newborn Adiposity
Background
Several clinical researchers are designing a randomized controlled trial to evaluate whether counseling women with overweight and obesity to eat a healthy diet and maintain an appropriate level of physical activity during pregnancy affects infant adiposity. Preventing excessive gestational weight gain in the women may lead to a healthier body composition in the offspring at birth (less FM, greater FFM). The study aims to determine whether the intervention delivered to women during pregnancy, when the fetus is developing, has a measurable effect on offspring body composition at birth.
Considerations
The researchers are aware that most births will take place in hospitals and there is a limited time window (1–3 days) in which they can measure most infants before the mother is discharged. Conducting the study in the hospital ensures privacy. Additionally, length, weight, and head circumference measurements are taken routinely after birth to assess overall newborn health. Acceptability of the measurements to parents who will provide consent is often high. The researchers will be only required to measure body composition of the newborn one time before the mother and newborn are discharged from the hospital following birth, which eliminates the need for a method that is sensitive to changes over time. However, the method(s) needs to be sufficiently sensitive to detect small differences in FM and FFM between the intervention group and usual care group.
Method Selection
Many methods could be used in this study depending on the resources available to the research team as well as parental preferences and concerns. Length and weight are the most common measurements taken as they are the least intrusive; generally require little skill, training, and equipment; and can be conducted quickly with minimal costs. However, length and weight and their relative indices such as weight-for-length and weight-for-age percentiles do not provide information on body composition, specifically FM or FFM. Such indices assume that a higher percentile reflects additional or excess FM and fail to consider the contribution of the FFM compartment to weight. The study team finds this to be potentially problematic because a higher index could reflect greater lean mass rather than FM.
The team considers other potential methods to more specifically assess total body fat. Skinfold thicknesses of the triceps, subscapular, and iliac crest can be used to assess between group differences in subcutaneous fat, which is a proxy for total body fat. These methods are noninvasive and often acceptable to parents. These methods do require data collector skill and training to conduct and additional time to acquire and can be burdensome to the parent and child.
Methods that have been validated to measure whole-body FM and FFM with high precision in the newborn are air displacement plethysmography (ADP, PEA POD) and dual energy X-ray absorptiometry (DXA). These instruments may be available in the hospital setting but require higher cost, the need for trained/technical staff, and time. They all require that the infant be as still as possible during testing. For DXA, the infant is measured swaddled in a blanket. For ADP, the infant must be naked in a chamber where air temperature is comfortable. DXA also involves a small amount of radiation. For these reasons, these methods may be not be acceptable to the parent who will provide consent.
Ultimately, the study team decides to use DXA for the primary outcome as it is available in this setting and can precisely assess FM and bone-free lean mass. The study will also measure length and weight and use their indices (weigh-for-length and weight-for-age) as a secondary outcome as these measures are commonly used in clinical practice to assess growth and commonly used in other studies for comparison.