Propagation of the US Obesity Epidemic Across Generations: Maternal Obesity and Future Obesity Prevalence

Keisuke Ejima
(University of Alabama at Birmingham)

2015/12/21, 13:30 - , at W1-C-909

      Abstract: It has been known for some time that there is genetic transmission (heritability) of obesity across generations. Evidence has also been accumulating that there can be mechanisms by which mothers transmit obesity to their offspring beyond transmission of the nuclear genome sequence. However, the approaches to quantifying the effects of such mechanisms on the changes in BMI distribution or obesity prevalence across generations are limited. We attempted to quantify the genetic and phenotypic obesity transmission from mother to child with readily available data, and subsequently predict the future obesity prevalence. First, we divided the total population into two sub populations, obese and non-obese, and construct a mathematical model that describes the time evolution of each of the sub population. We parameterized the genetic inheritance of obesity gene (FTO) (genetic transmission), the excess mortality associated with obesity, and differential chances of childhood obesity associated with maternal obesity (phenotypic transmission) in the model. Second, combining the readily available data (The U.S. Census Bureau and NHANES, NHS) and our model, we estimated the parameters in the model including phenotypic and genetic transmission. Finally, we projected the future obesity prevalence based on the estimated parameter set. We found that 49.8% of children of obese mothers develop obesity in their childhood (on contrary, 11.6% of children of non-obese mothers develop obesity in their childhood), and relative obesity hazard per A-allele of FTO gene was estimated as 1.77. Additionally, we estimated the time-variant obesity hazard and found that it is growing continuously in these 50 years. Running the model with the estimated parameters, the obesity prevalence in US is expected to reach and saturate at 53.4%, and can potentially be suppressed by 13.2% if the phenotypic transmission is perfectly interrupted.

Back: Japanese / English