Abstract Obesity is an established driver of cancer risk, and substantial evidence has identified inadequate physical activity and sleep as common risk factors. Inadequate physical activity contributes to over 12% of breast and colon cancers each year in the U.S. Furthermore, 10-20% of the general population experiences curtailed sleep and/or sleep disturbances, which have been associated with higher risk of breast, colon, prostate, and endometrial cancers. Geographic contextual measures, including neighborhood walkability and access to green space, have been demonstrated to affect physical activity, sleep patterns, and obesity. Both behaviors and environmental influences are most commonly measured with questionnaires, which almost certainly have substantial error, but this error has not been quantified. Novel mobile health technologies (mHealth), such as global positioning systems (GPS)-enabled smartphones and consumer-wearable accelerometry devices, can provide efficient, rigorous, and objective measures of geographic context, physical activity, and sleep with high spatio-temporal resolution. However, managing, processing, and analyzing streaming high-dimensional data has presented significant logistical and analytical challenges, especially when linking these data to existing data from large prospective cohorts. In this talk, I will discuss how we are measuring the interdependent relationships between geographic context, physical activity, sleep, and obesity by deploying smartphone applications and wearable devices within a subsample (n=500) of the Nurses’ Health Study 3 (NHS3). NHS3 is a web-based, nationwide, prospective open cohort with a current enrollment of ~46,000 male and female nurses aged 19-46 years old, a critical period for behaviors that influence cancer etiology. We are using mHealth approaches to collect streaming, high spatio-temporal resolution measures of geographic context (walkability and green space) based on smartphone GPS, physical activity, and sleep (based on Fitbit wearable devices) over a 7-day monitoring period, four times over one year. We are developing statistical methods to examine the interrelationships between these high-dimensional, intensive measures of context and behavior. Our long-term goal is to assess the effect of minute-to-minute exposure to geographic context on objective measures of physical activity and sleep, as well as subsequent obesity and cancer risk, within the full NHS3 cohort. This study is enabling us to rigorously quantify contextual exposures, physical activity, and sleep, and to identify the influence of geographic contextual factors on these interdependent behavioral risk factors for cancer and obesity. As we are advancing research on the objective measurement of geographic context, physical activity, and sleep through mHealth technology, we are bridging the gap between precise studies with small sample sizes and large prospective studies with abundant error. The approaches discussed here will aid in advancing epidemiologic research to better evaluate causal mechanisms between geographic context and behavioral risk factors for cancer. Citation Format: Peter James. Incorporating mobile health technology into a prospective cohort study to measure environment, physical activity, and sleep [abstract]. In: Proceedings of the AACR Special Conference on Modernizing Population Sciences in the Digital Age; 2019 Feb 19-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(9 Suppl):Abstract nr IA05.
Read full abstract