As the obesity epidemic continues unabated, the scientific community has struggled to find ways to address it. One question that has dogged researchers is: does where you live influence your weight? For example, does a park nearby promote regular exercise? Switch that park to a fast-food restaurant: Will you now eat more junk food? These are important questions because reducing access to unhealthy foods and building more parks are tangible things that policymakers can do to fight obesity.
Ideally, this question can be answered by conducting an experiment that, for example, randomly assigns one set of people to live close to fast-food restaurants and another set to live far from fast-food restaurants and then tracks what they eat and how much they weigh over time. However, such experiments are infeasible because researchers cannot randomly move people from one location to another at will. Instead, researchers simply compare the diet, exercise and body weight of individuals that have different opportunities for eating and exercising in their neighborhood. But, this approach is problematic because individuals choose which neighborhoods to live in instead of being randomly assigned. And therefore, a finding that people who live close to parks exercise more could arise because – (a) people who like to exercise also like to live near parks, or (b) because living close to a park causes one to exercise more regularly. Only experiments can allow us to distinguish between (a) versus (b). Therefore, scientists have called for more rigorous evidence, including studies of natural experiments – where randomization of subjects is determined by nature or factors outside the control of researchers.
At USC’s Center for Economic and Social Research, an NIH-funded study is taking on this challenge.
The Military Teenagers Environments, Exercise and Nutrition Study, or M-TEENS, is a unique study that leverages a natural experiment where military children frequently relocate to new neighborhoods as a result of their military parents’ transfers. Such relocations are determined by the military to support the unique missions and units at each installation. Consequently, the children of military personnel are “as good as randomly assigned” to varying neighborhood environments, creating a novel opportunity to study the role of place in influencing children’s diet, exercise, and unhealthy weight.
M-TEENS has collected data on eating and exercise behaviors and BMI from over 1500 12-13 year old children from Army families since 2013. Children, parents, and schools attended by the children are surveyed online and children’s height and weight are measured during visits to installations. The study collects data on what types of food and exercise opportunities are available in families’ neighborhoods by linking data from extant sources to the families’ residence location and by asking parents directly. This same information is collected every time the families relocate due to the military parent’s transfer.
M-TEENS has yielded interesting findings so far. In particular, we have learned that food opportunities in the neighborhood do not seem to matter for what children eat or for their body mass index (BMI) – a common measure of unhealthy weight (Shier, V., Nicosia, N., Datar, A. (2016). Neighborhood and home food environment and children’s diet and obesity: Evidence from military personnel’s installation assignment. Social Science & Medicine, 158, 122-131.). A possible explanation is that families do not necessarily eat or shop for food at places in their neighborhood, frequently going to locations further away from home. Similar findings have been reported even in the non-military population (Ilya Rahkovsky and Samantha Snyder. 2015. Food Choices and Store Proximity. USDA Economic Research Service, Report No. ERR-195.).
Instead, we find that what children eat is related to what is available at home, whether parents set limits for children’s access to junk foods at home, and how often they eat meals together as a family.
But then, why do some families have healthier home environments than others, beyond the usual socioeconomic factors? This is a question that a follow-up study is seeking to address. We hypothesize that some of these differences may be explained by preferences, such as future-orientation and risk-aversion. Because eating healthy and exercising involves costs today but yield benefits mostly in the future, individuals who value the future more might be more likely to eat healthy and exercise today. Likewise, healthy lifestyles reduce the risk for chronic diseases (such as diabetes) and so those who are risk-averse would also be more likely to eat healthy and exercise. The M-TEENS follow-up study will measure parents’ and adolescents’ future-orientation and risk-aversion via surveys and choice tasks administered at multiple time points and will examine whether these preferences enhance or diminish the importance of neighborhoods in influencing unhealthy weight in children and related behaviors.
Beyond these questions, M-TEENS data will also allow us to study the influence of other aspects of “place” such as the policies in effect, the social and economic environment, and the culture of health in the community, among other things. These questions are central for improving our understanding of whether, how, and for whom place matters.
Acknowledgement: M-TEENS is funded by a grant from the National Institute of Child Health and Human Development (grant no. R01HD067536). We gratefully acknowledge the Department of the Army for facilitating data collection for this study and the M-TEENS families for their participation.