Improving Built Environments Could Help Address Adolescent Obesity


Built environments, but not social and economic environments, were negatively associated with BMI and overweight or obesity status, with a stronger association among adolescents with longer exposure to their built environment.

Maria J. Prados, PhD - University of Southern California

Maria J. Prados, PhD

Courtesy of University of Southern California

Built environments, not social and economic environments, are a strong predictor of body mass index (BMI), overweight or obesity status, and eating behaviors in adolescents, according to new research.1

iThe investigative team at the University of Southern California indicated the association was stronger among those with greater or longer exposure to their built environments. According to investigators, these results help provide the first quasi-experimental evidence of these environments on the study factors.

“Our research suggests that strategies for addressing childhood and adolescent obesity should focus on improving built environments more comprehensively,” said Maria J. Prados, PhD, Center for Economic and Social Research, University of Southern California.2 “Adolescents represent an important target for potential health policy interventions because they are at an age when their health behaviors, preferences, and interactions with the environment are evolving.”

Environmental domains, including built, social, and economic environments, may simultaneously influence obesity-related behaviors and outcomes, based on theory. As Prados and colleagues indicate, a critical challenge arises when assessing the causal effects of these multifaceted environments on obesity to better inform policy. Individuals generally self-select into areas, meaning observational studies may not allow for causal inference on the effects of place-specific environments, according to investigators.

Moreover, studies designed to address causality, including the Moving to Opportunity housing experiment and other quasi- and natural experiment studies, have not directly or simultaneously assessed the independent roles of these environments on obesity. These studies have instead focused on a narrower set of environmental characteristics, including supermarkets, restaurants, or peers, or looked at the effects of the neighborhood.

The current natural experiment study relied on the routine assignment of military personnel to different installations based on the needs of the Army, exploiting quasi-exogenous variation in families’ exposure to communities with varying rates of obesity. Three indices for the built, social, and economic environments characterized 35 county-level environments based on 19 indicators. Using data from the Military Teenagers Environments, Exercise, and Nutrition Study (M-TEENS), the analysis sample consisted of 1111 adolescents with data on BMI, diet, and activity behaviors, as well as individual characteristics.

Around half of the analysis sample was female, with an age range of 12 to 14 years, and substantial racial and ethnic variation. Prados and colleagues identified significant heterogeneity in the two measures of exposure to the civilian environment: 61% of families had been assigned to the military installations for more than 2 years and 54% were living off the installation in the surrounding community.

Upon analysis, data showed the exposure to a more advantageous built environment for more than 2 years was associated with lower probabilities of obesity (0.18; 95% confidence interval [CI], 0.34 to 0.026). In addition, investigators identified the marginal effects on the probability of overweight or obesity as larger in magnitude, with a 33.7-percentage-point lower probability for a one-unit improvement in the built environment index for those with more than 2 years at the installation (95% CI, –0.56 to –0.12; P = .003). The results for the adolescents living off-installation were similar, according to the analysis.

The analysis showed the estimated coefficients for the association between built environment and BMI z scores were –0.76 for the subsample with more than 2 years at the installation (95% CI, –1.45 to –0.02; P = .04) and –0.72 for the off-installation subsample (95% CI, –1.37 to –0.07; P = .03).

Prados and colleagues noted more advantageous built environments were associated with lower consumption of unhealthy foods, but built environments were not associated with the weekly amount of time spent in physical activity. Additionally, the estimates do not support associations between social economic environments and adolescent’s diet or physical activity.

“Further research is needed to understand whether social and economic environments may be more consequential among adolescents and adults in civilian populations,” investigators wrote.1


  1. Prados MJ, Nicosia N, Datar A. Impact of built, social, and economic environments on adolescent obesity and related health behaviors. Obesity (Silver Spring). 2023;31(4):1085-1094. doi:10.1002/oby.23682
  2. ObesitySociety. (2023, March 21). Built environment strongest predictor of adolescent obesity, related health behaviors. EurekAlert! Retrieved March 21, 2023, from
Recent Videos
1 KOL is featured in this Insights series.
Charles C. Wykoff, MD, PhD: Interim Analysis on Ixo-Vec Gene Therapy for nAMD | Image Credit: Retina Consultants of Texas
Sunir J. Garg, MD: Pegcetacoplan Preserves Visual Function on Microperimetry | Image Credit: Wills Eye Hospital
Edward H. Wood, MD: Pharmacodynamics of Subretinal RGX-314 for Wet AMD | Image Credit: Austin Retina Associates
Dilsher Dhoot, MD: OTX-TKI for NPDR in Interim Phase 1 HELIOS Results  | Image Credit: LinkedIn
Katherine Talcott, MD: Baseline EZ Integrity Features Predict GA Progression | Image Credit: LinkedIn
Veeral Sheth, MD: Assessment of EYP-1901 Supplemental Injection Use in Wet AMD | Image Credit: University Retina
Discussing Post-Hoc Data on Ruxolitinib for Nonsegmental Vitiligo, with David Rosmarin, MD
1 KOL is featured in this series.
1 KOL is featured in this series.
© 2024 MJH Life Sciences

All rights reserved.