The recent enthusiasm internationally for investigating the impact of the environment on health augurs well for efforts to broaden and advance our understanding of the determinants of nutrition and physical activity behaviors. Such research has a range of important practical and policy implications. "One size fits all" blanket approaches to promoting nutrition and physical activity assume that 'background' components are equal, whereas a growing body of research demonstrates that people live in a range of environments, not all of which are conducive to healthy eating and physical activity. In addition to informing environmental planning and policy efforts, knowledge of key environmental determinants may be vital in information and awareness-raising approaches to promoting healthy behavior, for example by helping people to make the most of the environments in which they live, work and play. Findings may also inform community advocacy efforts. If residents are aware that traffic calming, green spaces or fresh fruit availability are important, for instance, they can lobby to have these elements of their neighbourhoods modified.
What are the implications for research? There is little consensus as to the most appropriate approaches for defining environments or the selections of environmental exposures that might be important for nutrition and physical activity behaviors. Much research has been opportunistic, focussing on existing measures of facility availability only. Some researchers have attempted to develop composite indices taking into account that accessibility of facilities depends not only on distance but also attractiveness, importance to the user, and so forth [39
], and this developing field of work might address some of these complexities. With few exceptions [58
], much research in the field to date has also been observational, which does not provide strong evidence as to the temporal or causal nature of associations of environmental factors with behavior.
In our opinion, however, the most significant challenge for researchers in this field is to clearly articulate theoretical models from which to develop testable hypotheses regarding the influence of environmental exposures on nutrition and physical activity behaviors. Conceptual development currently lags behind analytical advances, and there remains considerable complexity and little evidence to guide most appropriate definitions of environment, or identification of important influences on behavior. There remains a critical need for continued development and application of theoretical models linking environmental and individual exposures to health behaviors.
Future research: a proposed agenda
Based on the overview of the literature and the methodological and conceptual challenges outlined here, we propose the following key priorities as part of a research agenda for advancing our understanding of environmental determinants of nutrition and physical activity behaviors.
Future research should:
• Focus strongly on progressing development and testing of clear conceptual models. Hypothesize specific, testable hypotheses that explicitly link different levels of influence from social ecological models to context-specific nutrition or physical activity behavioral outcomes in particular population groups; and test these models with appropriate empirical data (which may require new data collections): for example, a model of adults' leisure-time walking; of children's active free play; of adolescents' sport participation; of adolescents' soft drink consumption; or of older adults' fruit and vegetable consumption.
• Consider the relative influence of factors within different environments in which people live and function (home, work, school, neighbourhood).
• Focus on progressing methodological efforts, including refining environmental measures. Further develop, reliability and validity test environmental audit instruments. Investigate the implications of using different sized-buffer zones for different environmental exposures, behaviors and across different target groups.
• Incorporate less well-studied nutrition and physical activity behaviors (eg other than walking).
• Capitalize on available analytical techniques to test theoretically-based hypotheses regarding the cross-level interactions of individual, social and environmental determinants, and mediating and moderating relationships amongst determinants from different levels of analysis.
• Progress conceptual understanding and empirical evidence regarding the role of the social, cultural and policy environments.
• Extend primarily observational research studies, for example, by utilizing opportunities for 'natural experiments' such as assessing the impact of new food stores or physical activity facilities
There seems little doubt that aspects of 'the environment' have a potent influence on nutrition and physical activity behaviors. As yet, research in this area is in its infancy. Much remains to be done, creating exciting opportunities for innovative methodological, conceptual and empirical research in the coming years.