Research on the relation between the built environment and physical activity and nutrition is still in an early stage. Several issues need to be resolved before the utility of these measures in various contexts is understood.
First, is the perceived environment or the objective environment more strongly associated with physical activity and eating patterns? As Brownson et al. (2004) states, "because it is not clear whether perceived or objectively measured environmental variables provide more explanatory power, the use of triangulation … is recommended" (p. 479). One of the few studies that examined both objective and perceived environments found that different aspects of the perceived and objective environments were related to different types of physical activity (Hoehner et al., 2005). For example, walking and bicycling for transportation were positively associated with objective measures of the number of destinations and availability of public transit, and negatively associated with both objective and perceived neighborhood aesthetics. In contrast, recreational activity was positively associated with objective measures of neighborhood aesthetics such as shade trees and minimal litter, and perceived access to recreational facilities. We know of no such studies to date addressing the nutrition environment.
Second, is there a causal relationship between neighborhood environment and behaviors of physical activity and/or healthy eating? In the healthy eating arena, there is limited evidence of environment-behavior associations from a few cross-sectional studies (Glanz et al., 2005). Most of the research done to date in physical activity environments and behaviors has been cross-sectional and numerous researchers have called for longitudinal studies (Brownson et al., 2004; Humpel et al., 2002; Owen et al., 2004).
Third, how context-dependent are the observed relationships between neighborhood environment and physical activity? Sallis and colleagues (2006) developed an ecologic model for active living that underscores the complexity of understanding and increasing physical activity in populations. The model includes four active living domains (i.e., recreation, household activities, occupational activities, and transportation activities) and posits that different constellations of factors affect each. Similarly, there are no studies that simultaneously examine healthy eating and active living environments – which can be hypothesized to operate together as contributors to overweight and obesity.
A final issue that remains unexamined relates to the potential for environmental changes to influence changes in behaviors – in particular, little is known about how sensitive the available measures of activity and eating environments are to change. Even less is known about how much environmental change might be necessary to achieve meaningful effects on behavior and health outcomes.
There is much more work to be done in designing and testing measures of food and activity environments that are adaptable to a variety of locations and health issues. Developers of these measures will be challenged to be attentive to the meaningfulness of indicators, relevance and feasibility of measures, and potential for linking environmental and individual assessments in subsequent studies. A range of psychometrically sound measures are needed to obtain accurate and reliable estimates of the relation between nutrition and physical activity environments and individuals’ health behaviors and weight status, as well as to evaluate change in these environments secondary to intervention. Despite numerous research challenges in this line of inquiry, development and dissemination of valid and reliable measures is a critical early step.