Everyday environmental exposures and mid-life dietary and physical activity variations: E3 study protocol.

Background: Poor dietary and physical activity (PA) behaviors increase risk for obesity, cancer, other chronic diseases, and contribute to disparities. Mid-life remains a vulnerable life stage, where obesity rates peak, and chronic diseases emerge. Neighborhood environments provide opportunities, barriers, and cues/triggers to engage in healthy or unhealthy behaviors. However, most research is limited to consideration of residential neighborhoods proximal to an individual's primary residence and does not consider the environment's role in within-person daily and momentary differences in behaviors. Furthermore, current research tends to ignore questions about for whom the environment matters and under what conditions. Objective: To address environmental exposures and provide a test of activity-space environmental explanations for between-and within-person diet and PA variations during mid-life. Our central hypothesis is that activity-space environmental exposures contribute to both between- and within-person variations in dietary and PA behaviors and influence these behaviors, more than residential-neighborhood environments alone. Additionally, we hypothesize that activity-space environmental exposures are more consequential for diet and PA when self-regulatory capacity, trait or state factors that affect a person's ability to make efforts to regulate behavior are diminished. Methods: A total of 400 adults ages 40-64 were enrolled. We used a combination of geographically-explicit ecological momentary assessment (GEMA) methodologies: global positioning system (GPS) location tracking; smartphone-based mini-surveys of diet, PA, and state factors; and accelerometry, as well as three 24-h dietary recalls, anthropometric measurements, and questionnaires of trait and other factors. Routine, daily, and momentary activity-space measures were derived based on the spatial extent of their movement and duration of exposure. Residential and activity-space environment features were measured using GIS, including absolute and relative availability of healthful and unhealthful foods, walkability, recreational resource availability, and crime. Conclusion: This research employed a dynamic environmental exposure approach using GEMA to supply evidence on the environment's role in between-and within-person variations in diet and PA during mid-life, a pivotal time, in a racially/ethnically diverse sample. As such it will contribute to the understanding of how environmental determinants of behaviors are studied, informing new targets for lifestyle and place-based interventions to improve health during mid-life.