Setting girls on a healthy diet trajectory from an early age

Globally, poor diet is the second biggest risk factor for early death after smoking. According to the World Health Organization, approximately 1.7 million (2.8%) of deaths worldwide are attributable to low fruit and vegetable consumption. Nutrition studies suggest that to combat the ever increasing risk of poor diet and obesity, interventions need to target the development of healthy eating behaviors in early child-hood. «Diet quality tracks across the life-course,» Dr. Megan Jarman explains. «Children with poor quality diets, characterized by frequent intakes of added sugar and fewer fruits and vegetables, have a higher risk of becoming obese». An AXA grantee at the Aston University in Birmingham, Dr. Jarman is leading a highly innovative transdisciplinary investigation into the different factors that influence children’s diet quality at home. The ultimate objective is to create a computer model, that will allow for intervention strategies to be simulated and tested which will identify key levers for future intervention.

The study will pay particular attention to gender differences. Indeed, as the researcher explains, «estab-lishing a healthy diet early in life is particularly important in women, notably because during pregnancy the way in which women eat can influence the life-long health of her offspring ». Indeed, she reports that poor quality diets, and excess weight in women going into pregnancy are associated with poorer quality diets and excess weight in offspring. On top of that, it has been observed that by mid-childhood, the diet quality of girls declines more rapidly than boys, and by adolescence, this may contribute to a higher risk of being overweight or obese in girls compared to boys. « This is globally true across Europe », Dr. Jarman specified.

A transdisciplinary approach: a synergy between genetic, psychological, behavioral and nutrition data

« There has been quite a bit of research already on how young people eat, but most of the time, these studies are mere snapshots of the problem. As we know, eating is a behavior, so it’s very complex. It is not just about preferences, for instance. A myriad of factors come into play: what we like or don’t like, of course, but also our food environment, for instance, whether we eat at a table or in front of the televi-sion, if parents model good eating behaviors , etc. » This collaborative project aims to be the first in the world to synergize genetic, psychological, behavioral, and nutrition data, to assess the interrelationships between these factors and their influence on children’s diet quality.

« An emerging method in public health is to use complex systems modeling. It allows us to consider the complexity of the interactions between different factors for one outcome, which in the present case, is the quality of eating of young children», Dr. Jarman explains. Specifically, the computational model the project will build is called an agent-based model (ABM). It simulates the actions and interactions of au-tonomous individuals, called agents, within a given environment. « The first stages will consist of pulling together all the data sources we have, meeting with experts, and brainstorming about all the influences children face in the home environment. Then, we will list these influences, determine a hierarchy, devel-op a system map and finally, build, test and refine the model.»

By bringing together a pool of experts from various fields, this project’s comprehensive take on the pressing issue of obesity is both ambitious and innovative. By targeting young girls in particular, and thus future mothers, it has the potential to address «the transgenerational cycle of increased risk of poor health.» Additionally, her collaborative work with the local government and other key stakeholders holds great potential for effective application of the project’s findings. « From a research point of view, the sustainable impact of past interventions has been disappointing, she reflects. Perhaps that’s because we haven’t hit the nail on the head, so far. Complex modeling will be helpful in that manner, so we can tailor the interventions and identify what is likely to be effective for who and under what circumstances.»