Associations between the built environment and physical activity from analyses of spatial clusters, trail use, and locations where physical activity occurs

Kosuke Tamura, Purdue University


Over the past two decades, an increasing number of scientific studies have examined associations between the built environment and physical activity and obesity. These studies have documented positive associations between environmental variables, such as population density, street connectivity, and composite measures of neighborhood walkability and physical activity. Studies have also shown inverse relationships between the presence of neighborhood grocery stores and recreational facilities and obesity. Despite this evidence, there continues to be limitations in built environment studies conducted to date. The three dissertation studies described here were designed to address several different aspects of built environment research that warrant greater attention. The first study addressed the issue of whether physical activity and obesity is spatially clustered in relation to certain attributes of the built environment. Many previous built environment studies used geographically referenced data, such as geocoded home addresses and locations of facilities. The use of these types of data neglected spatial relationships among observations. The second study examined whether objectively measured trail use was associated with physical activity and sedentary behavior. The study also quantified on-trail physical activity using two approaches: accelerometer counts only and both counts and GPS data. Participants (N = 141, mean age = 44.1 ± 13.0) were recruited on five trails in Massachusetts. They were asked to wear accelerometer and GPS devices for four days. Total physical activity, and daily minutes of light, moderate, and vigorous physical activity, and sedentary behavior were derived from accelerometer counts. A trail-use day was defined as a minimum of two consecutive monitoring minutes occurring on-trail. Linear mixed models were used to examine whether trail use was related to physical activity and sedentary behavior. Overall, statistically significant positive associations were found between trail use and physical activity. Trail use was associated with about 28 minutes of moderate physical activity per day compared to no trail use. On-trail vigorous physical activity minutes increased by 346%, based on accelerometer and GPS data compared to accelerometer counts only. This trail study provided evidence that adults engaged in more physical activity when they use trails. In addition, this study indicated that the use of both accelerometer and GPS data may be a useful method for classifying intensity of physical activity occurring trails; particularly facilities where bicycling is a common activity. (Abstract shortened by ProQuest.)




Klenosky, Purdue University.

Subject Area

Public health

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server