Nancy Humpel, Eva Leslie, and I were fortunate to be well placed to synthesize information from new studies and to be part of identifying the research agenda for a rapidly developing research field. Nancy had done excellent work in this area for her Ph.D. studies, and we realized that we had something important to say about the pattern of findings in recent empirical studies. We are very pleased indeed that research on how the built environment influences physical activity is receiving the attention that it deserves—the American Journal of Preventive Medicine has been particularly insightful and proactive in identifying this work as important for science and for public health. Our review was the first paper to synthesize the initial pattern of findings in a hot new sub-field of research on physical activity and health. It was only as recently as 1996 that we saw the first US Surgeon General’s Report on Physical Activity and Health—particularly focusing on the prevention of metabolic disorders like Type 2 diabetes and hyperlipidemia, cardiovascular disease and musculoskeletal disorders. More recently, cancer research and advocacy bodies have begun to disseminate new scientific information on the importance of weight control and physical activity for reducing the risk of breast and colon cancers. Physical activity is now, after tobacco, understood to be the most important determinant of chronic disease risk in the populations of industrialized countries—and also in many very large, rapidly urbanizing populations in developing countries. The detrimental health effects of sedentariness and the benefits of physical activity are well documented; and we know quite a lot about the distribution of physical activity and inactivity in populations. There is thus a new public health research agenda to understand the determinants of physical activity, so that appropriate interventions and policies can be developed. The Robert Wood Johnson Foundation has a large-scale initiative on "Active Living Research," led by Jim Sallis at San Diego State University. Robert Wood Johnson are funding new research on the measurement of environmental factors that may influence physical activity and other studies to determine how built environments are objectively related to activity as people go about their day-to-day lives. There is also a new set of initiatives by the Centers for Disease Control in Atlanta, drawing together researchers and practitioners from urban planning and transport with behavioral scientists, epidemiologists, and public health experts. The CDC is active and effective in stimulating new policy initiatives and supporting their related research agendas. There is also much good new research going on in several European and Scandinavian countries. There have been two recent special sections on physical activity and the built environment in two of the key journals in the field—the American Journal of Public Health and the American Journal of Health Promotion. It is definitely a hot research topic. It is also gratifying to know that there are not too many steps necessary to follow from our research findings in order to arrive at initiatives that will have a sustained public health benefit.

Doing behavioral studies in population health, we are busy mapping and systematizing much new territory. New discoveries will more likely come from syntheses of the accumulated efforts of networks of researchers, rather than from dazzling individual findings. Our work applies a conceptual model—an "ecological" perspective, emphasizing direct environmental influence on particular behaviors—to synthesize some of the key elements of the research literature. At this stage, our theoretical framework is still too broad and fuzzy in places, but we are moving towards greater theoretical specificity and have ideas for studies that might start to point to underlying mechanisms. In our research in Australia, some of which is covered by our review paper, we are working to link Geographic Information System databases to the behavioral measurements of physical activity. We use surveys, accelerometers, and pedometers to capture this behavior as objectively as is possible in large groups of people—often with samples from whole populations or defined communities. We are also building on some excellent new findings emerging from the work of Billie Corti’s group at the University of Western Australia in Perth. We have been using an ecological model of direct environmental influence on behavior to guide two new studies—one in Seattle and Baltimore and one in Adelaide, South Australia. These will allow us to better identify the particular influences that attributes of the physical environment can have on choices of whether to walk for exercise or recreation and for transport. We will also be looking at how environmental factors interact with cognitive attributes such as efficacy and outcome expectations and social support and we will be explicitly controlling for the effects of socio-economic status. The overall scientific goal is to identify the role environmental factors play in determining the different facets of adults’ physical activity habits.

It is crucial that we build a strong body of scientific evidence that shows how people’s community environments (for example, presence of sidewalks, green open space, destinations to walk or bike to, active commuting options, etc.) can determine how physically active they are. Adults who spend a lot of time watching TV, using computers and driving cars have poor metabolic control and are more likely to develop diabetes. They are also at an increased risk of heart disease or cancers of the breast and colon. It is not enough to tell people that they should exercise more—real opportunities for active living have to be available for them. Unfortunately, in many communities, adults have very few realistic active choices available, given their physical environments and the demands of their work and family lives. The goal of this research is to gain objective evidence on whether environments do act directly to influence people’s choices to be physically active.

This research reflects my central scientific preoccupation since I started my psychology doctorate back in 1969. I was fascinated by the ideas of B. F. Skinner—that cues and rewards from people’s environments are what primarily shape their habitual behavior. This simple, but fundamental, idea is at odds with conventional views of what social cognitive theories of behavior are telling us; there is a strong tendency to overemphasize personal responsibility and individual decision making as the main determinants of behavioral choice. As an Australian, I have always felt that these notions of personal responsibility, and the theories growing out of them, had a peculiarly North American flavor. Theories, especially theories of human nature, tend to reflect the cultures that generate them, and we remain skeptical of the cognitively-focused theorizing about the determinants of habitual health behaviors that comes through in much of the work done in health psychology in the USA. With Jim Sallis, our colleague in San Diego, we have started to articulate a conceptual framework that emphasizes direct environmental influences on health-related behavioral choices. In developing these ideas, I have been lucky to benefit from the thinking of many excellent research colleagues. With Ron Borland, David Hill, and Melanie Wakefield, I was able to work with some nice data on the effects of the Australian workplace smoking bans in the early 1990s. This was a case in point for environmental influence, in what might seem to be a context of individual choices to smoke, driven by emotion, thoughts, or the underlying pharmacology of addiction. I have also been fortunate to work with my epidemiologist colleague Adrian Bauman, exercise physiologist Wendy Brown, and many excellent postgraduate students and postdoctoral fellows from behavioral science, exercise science, and public health disciplines. I am particularly pleased to have worked with Nancy Humpel and Eva Leslie, my co-authors on the review paper—their energy and insight has been central not just to that paper, but to the groundwork on which our new program of studies is built. We are fortunate also to have a solid base of research funding from the Queensland government, through Queensland Health; and, more recently, the National Health and Medicinal Research Council has provided support through program and capacity building funding. Their support is invaluable and much appreciated—it allows us to approach our research in a programmatic manner. Indeed, our work is becoming even more interdisciplinary, through our collaboration with Graeme Hugo at the National Centre for Social Applications of Geographic Information Systems in Adelaide and Larry Frank (an urban planning expert) at the University of British Columbia in Canada. Working with experts from social geography, demography, urban planning, and transportation has helped to broaden our thinking about how environmental factors might act to influence physical activity.