Ecological models specify multiple levels of influence on behavior, from individual and social factors, to institutional, community, built environment, and policy factors. A key principle is that interventions should be most effective when they change the person, the social environment, as well as built environments and policies.6
Motivating a person to change in an environment that poses many barriers is not expected to be very effective, nor is providing a supportive environment in the absence of educational interventions to promote use of those environments.
Built environments are the totality of places built or designed by humans, including buildings, grounds around buildings, layout of communities, transportation infrastructure, and parks and trails.7
Policies can be laws and regulations at any level of government, corporate practices, and rules at institutions like schools. Changing built environments and policies is expected to have long-term impact on most or all of the people in those places. Characteristics of built environments, from neighborhoods to cities, have been related to rates of chronic disease and mental health8–10
and risk factors such as obesity11,12
Physical activity is believed to be a critical mechanism by which built environments can affect chronic disease.8,9
Societal changes over decades have dramatically reduced the need for physical activity in daily life while creating ubiquitous barriers to physical activity. Mechanization and computerization have reduced physical activity at work, labor saving devices have reduced activity required for household chores, and investments and policies that favored travel by automobiles have reduced walking and bicycling for transportation. Although these societal changes have had some desirable effects, they have also led to a decrease in daily physical activity.
Physical activity can be classified into four domains of life that describe how people spend their time: leisure/recreation/exercise, occupation (school for youth), transportation, and household.13
The four domains are relevant to and driven by different built environment features and policies. is a simple ecological model of physical activity that identifies institutional and community built environment settings and features, as well as policies, that are relevant to each physical activity domain.
An Ecological Model of Four Domains of Physical Activity
A commonality across all of the environmental settings identified in (i.e., recreation facilities, community design, transportation facilities, workplaces, schools, homes) is that none is controlled by health professionals. Yet these places can affect health. Thus, for both research and practical applications, it is necessary for health professionals to develop partnerships with professionals from diverse and often unfamiliar disciplines and sectors of society.14
The need for a multi-level, multi-sector approach to physical activity promotion, obesity and CVD prevention has been recognized by numerous health organizations.4,15–18
These recommendations justify a vigorous research program to identify modifiable environmental attributes and policies that have the strongest or most widespread effects/associations to guide intervention efforts. The Active Living Research program of The Robert Wood Johnson Foundation has been funding such studies since 2003,19
and the US National Institutes of Health includes environment and policy research in the Strategic Plan for NIH Obesity Research.20
Environmental research on physical activity has grown rapidly since 2000, has been reviewed numerous times,21–24
and is informing policy debates at all levels of government.
Because investigators do not have control over the policy or environment "intervention", it is rarely possible to randomly assign people or places to experimental conditions. Thus, most of the studies are observational, though quasi-experimental studies have become more common. In the present non-systematic review, overviews of the literature are provided and illustrative results are described as related to active transportation and active recreation, because these physical activity domains appear most amenable to environment and policy intervention. Because there is limited evidence about how built environments are related to occupational and household physical activity, these domains were not addressed.