Can weight at age 5 predict obesity in adulthood?
Obesity in children and adults is a major public health challenge worldwide. Recent research has shown that it is possible to predict with great accuracy the evolution of individuals’ body mass index from childhood to adulthood. To achieve this, scientists used an innovative approach combining artificial intelligence with genetic, family, and environmental data.
The study relied on data from nearly three thousand participants followed from age 8 to 27. Researchers found that weight at age 5, measured by body mass index, is the most determining factor in anticipating the risks of obesity later in life. This result is particularly pronounced during childhood, although its influence slightly decreases during adolescence and adulthood. On the other hand, the impact of genetic predispositions, assessed using scores calculated from DNA, becomes more significant after age 17.
Other factors also play a significant role. The weight and height of parents, their level of education, and body fat measurements taken during early childhood also influence weight development. For example, excess weight in the mother before or during pregnancy increases risks for the child. Similarly, a disadvantaged family environment can limit access to healthy food and regular physical activity, which promotes weight gain.
Scientists used an artificial intelligence model capable of analyzing hundreds of variables and extracting the most relevant ones. This model highlighted complex links between these different factors. It reveals, in particular, that genes associated with obesity in adults gain importance with age, while living conditions during early childhood remain crucial in the early years.
These findings underscore the importance of acting early to prevent obesity. Careful monitoring of weight from age 5 could help identify children at highest risk and implement appropriate measures. Interventions could target dietary habits, physical activity, and family support, taking into account parental history and living conditions.
The use of artificial intelligence in this study offers a new way to understand how the risk of obesity develops over time. It paves the way for more personalized and effective prevention strategies, based on data available from early childhood.
Sources and Credits
Source Study
DOI: https://doi.org/10.1038/s41366-026-02050-1
Title: Longitudinal prediction of BMI using explainable AI: integrating polygenic scores, maternal, early-life and familial factors
Journal: International Journal of Obesity
Publisher: Springer Science and Business Media LLC
Authors: Fuling Chen; Phillip E. Melton; Kevin Vinsen; Trevor Mori; Lawrence Beilin; Rae-Chi Huang