"Spatial Energetics": Integrating Data From GPS, Accelerometry, and GIS to Address Obesity and Inactivity

Abstract

To address the current obesity and inactivity epidemics, public health researchers have attempted to identify spatial factors that influence physical inactivity and obesity. Technologic and methodologic developments have led to a revolutionary ability to examine dynamic, high-resolution measures of temporally matched location and behavior data through GPS, accelerometry, and GIS. These advances allow the investigation of spatial energetics, high-spatiotemporal resolution data on location and time-matched energetics, to examine how environmental characteristics, space, and time are linked to activity-related health behaviors with far more robust and detailed data than in previous work. Although the transdisciplinary field of spatial energetics demonstrates promise to provide novel insights on how individuals and populations interact with their environment, there remain significant conceptual, technical, analytical, and ethical challenges stemming from the complex data streams that spatial energetics research generates. First, it is essential to better understand what spatial energetics data represent, the relevant spatial context of analysis for these data, and if spatial energetics can establish causality for development of spatially relevant interventions. Second, there are significant technical problems for analysis of voluminous and complex data that may require development of spatially aware scalable computational infrastructures. Third, the field must come to agreement on appropriate statistical methodologies to account for multiple observations per person. Finally, these challenges must be considered within the context of maintaining participant privacy and security. This article describes gaps in current practice and understanding and suggests solutions to move this promising area of research forward.

Figure 1

Hypothetical map of spatial energetics data: GPS data color coded by accelerometry cutpoints (illustration of basic spatial energetics data) overlaid on greenness spatial data.

Figure 2

Maps of spatial energetics data detailing conceptual issues: (a) GPS points and radial/line based network buffers around a home address; (b) Data derived locations.

Figure 3

Examples of technical issues with spatial energetics data: (a) Example of GIS layer error in St. Louis, Missouri park layers versus nationwide park layers; (b) Example of GPS scatter due to physical structures, bodies of water, and urban canyons that block or reflect GPS signals.