Using a siblings reared apart and together design to parse genetic vs. environmental influences on children's body mass index

Childhood obesity is associated with adverse outcomes that can persist into adulthood, including continued obesity, type 2 diabetes, and hypertension. Understanding the underlying mechanisms that contribute to childhood... [ view full abstract ]

Childhood obesity is associated with adverse outcomes that can persist into adulthood, including continued obesity, type 2 diabetes, and hypertension. Understanding the underlying mechanisms that contribute to childhood obesity is critical for prevention efforts. Research has identified both genetic and environmental influences on children’s risk for obesity, although most existing work confounds genetic background with rearing environment when comparing genetically-related individuals who are living in the same home (e.g., traditional family studies). Novel approaches such as the children of twins design and discordant twin design provide unique methods for examining shared environmental and genetic influences on child obesity, but have design-specific limitations related to shared rearing environment of family members. To address these confounds, the present study uses biological and adoptive parents and their children to examine how the rearing environment and genetics may contribute to the body mass indices (BMI) among siblings reared apart and together. The study sample was drawn from the Early Growth and Development Study and includes 899 linked children with different levels of relatedness (unrelated siblings, half-siblings, full siblings) residing in the same home or different homes. Parent-reported height, weight, age, and child sex were used to calculate BMI scores for all children using the US CDC 2000 norms. We used age-adjusted Z-score estimates of BMI for children ages 2 to 20. The complex nature of the study design with many multi-level connections among participants – both in relatedness and shared environment – provides a unique opportunity to employ a novel statistical approach – a profiled restricted maximum likelihood model (pREML) with pairwise relatedness and environmental matrices to partition variance into that which can be explained by genetic background (V_A), shared environment (V_C), and residual variance (V_E). Heritability was estimated as the proportion of variation in BMI that can be explained by genetic factors over the total variation in BMI. We observed a heritability estimate of 0.63 for BMI in this dataset, and an effect of shared environment of 0.26. The obtained heritability estimate of 0.63 falls in the middle range for other genetic studies that have used more traditional twin methods (e.g., range 0.40-0.85; Feng, 2016), suggesting that prior studies of genetically-related children reared together may underestimate the role of shared and unique environment on child BMI, or may include the effects of non-additive genetic factors in the heritability estimate. The results of this study confirm the unique influence of genetics while simultaneously highlighting the important role that the rearing environment can have on childhood weight status.

Reference

Feng, R. (2016). How much do we know about the heritability of BMI? American Journal of Clinical Nutrition, 104(2), 243-244.