Association of Exposure to Communities With High Obesity With Body Type Norms and Obesity Risk Among Teenagers

This cross-sectional study examines whether teenagers’ exposure to communities with high obesity rates is associated with ideal body type and risk of obesity.


Introduction
There is substantial evidence of clustering of obesity within geographic and social networks. 1 Designing effective policies requires understanding whether this clustering is owing to shared environments, homophily, or social contagion. However, conclusive evidence is limited by the lack of quasi-experimental or natural experiment studies.
In a 2018 study, 2 we analyzed data from a natural experiment, the Military Teenagers Environments, Exercise, and Nutrition Study (M-TEENS), in which US Army families were exposed to communities with varying rates of obesity as a result of their assignments to specific installations.
We found that teenagers assigned to counties with higher obesity rates were more likely to have overweight or obesity. The study design ruled out homophily as an explanation, and we found no evidence that shared environments explained these results. We suggested that our findings may be consistent with social contagion with the goal of exploring this issue in future work. In the current cross-sectional study, we used newly collected M-TEENS data to examine whether teenagers' exposure to communities with high obesity was associated with their ideal body type and obesity risk.

Methods
The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies. It was approved by the University of Southern California institutional review board.
Between Dec 2017 and July 2018, 401 M-TEENS participants (aged 16-19 years) and their parents completed online surveys that included self-reports and parent-reports regarding teenagers' height, weight, and other individual and household covariates. Parent consent and child assent were obtained online. Height and weight measurements, available for approximately half the sample, were    obesity (figures E-I). 3 We focused on whether they chose an IBT that reflected overweight or obesity, defined as figures D through I.

County obesity rates (COR) for their current installations were obtained from the Robert Wood
Johnson Foundation County Health Rankings Data. 4 In line with prior research suggesting a tipping point in social norms, 5 we captured high obesity exposure via an indicator for whether the COR exceeded the median (ie, 30.5%). We hypothesized that teenagers exposed to counties with high obesity rates would tend to choose an IBT reflecting overweight or obesity because of changes in their descriptive and injunctive norms about body size and would have higher obesity risk.
Multivariable logistic models estimated the association between exposure to high COR and selecting an IBT reflecting overweight or obesity and then between high COR and obesity, adjusting for IBT. All analyses were conducted in Stata version 14 (StataCorp). No prespecified level of statistical significance was set.

Results
Sample characteristics are reported in Table 1. Teenagers exposed to high COR had significantly higher odds of choosing an IBT reflecting overweight or obesity compared with those exposed to low COR (adjusted odds ratio, 1.96; 95% CI, 1.18-3.24) (

Discussion
Teenagers' exposure to high obesity was associated with selecting an IBT reflecting overweight or obesity and greater obesity risk. These findings suggest that high obesity rates may normalize unhealthy weight, 6 making obesity prevention more difficult. While exposure to high obesity rates primarily because of military parents' installation assignment was a key strength of the study, the cross-sectional design and generalizability were limitations. Longitudinal data on IBT norms and obesity risk from future M-TEENS waves can help assess the temporal association of obesity exposure with IBT norms and risk of obesity among teenagers.