Effects of Near-Roadway Air Pollution Exposure on Childhood Weight Gain Trajectories across Early Adolescence

Introduction:

Childhood overweight/obesity has emerged as a major public health crisis. In recent years, growing attention has also been paid to environmental exposures that may increase childhood obesity risk. Such factors include near-roadway air pollution (NRAP), which research suggests has adverse metabolic and disruptive built environmental effects that may deter physical activity and increase sedentary behavior. This secondary analysis tests hypotheses that greater NRAP exposure is associated with higher baseline BMI and waist circumference (WC) as well as with greater increases in BMI and WC trajectories in a cohort of students followed from 4^th-6^th grade.

Methods:

Data were gleaned from Pathways to Health, a cluster randomized control trial for the prevention of childhood obesity and substance use, which was carried out in 24 Southern California schools from 2009-2011. Participants were a longitudinal cohort of 475 students assessed 4 times from 4^th-6^th grade.

Estimated average NRAP exposures from local on-road motor vehicle emissions were assigned to geocoded residence and school locations for the year 2009 using CALINE4 line-source dispersion models for Total NO_X. Exposures to regional air pollutants were modeled using 2009 EPA monitoring data and retrospectively assigned to participant residences. Participant BMI and waist circumference were measured by two trained data collectors at each assessment and obesogenic behavioral covariates were assessed using validated survey instruments.

Linear mixed effects growth curve models were specified in STATA 14 employing random intercepts and slopes by individual and school. The a priori covariate set included participant gender, race, SES, program group, cigarette smoking, regional particulate and ozone air pollution and age. Sedentary behavior and physical activity were tested in subsequent models examining putative behavioral mechanisms.

Results:

At 4^th grade, estimates from adjusted models found that children in the upper 90^th percentile of NRAP exposure had waist circumferences that were on average 3.0 cm wider and BMIs that were 1.2 point higher relative to children in the lowest 10^th percentile of NRAP exposure (p<.05). By 6^th grade, these average differences in Waist Circumference and BMI were further exacerbated, with observed difference of 4.4cm and 1.4 BMI points, respectively. When average annual sedentary behavior and physical activity were entered into the models as covariates, baseline obesogenic effects of NRAP exposure were attenuated by approximately 15% while longitudinal effects were attenuated by 23% for BMI and 49% for waist circumference.

Conclusions:

As hypothesized, students exposed to higher levels of NRAP had higher baseline BMI and Waist Circumference, as well as steeper weight gain trajectories from 4^th-6^th grade, which may result from systematic differences in physical (in)activity associated with residence near busy roadways. However observed effects were small. Nevertheless, given current trends toward greater urbanization, the ubiquity of NRAP exposure and potential population health impacts suggests that larger follow-up studies are warranted to inform prevention efforts.