In this chapter, the committee presents its detailed consensus findings, conclusions, and recommendations. First, however, it should be emphasized that research on the relationship between the built environment and physical activity is at a pivotal stage. A growing body of empirical evidence, primarily from cross-sectional studies reviewed in Chapter 6, suggests an association between the built environment and physical activity levels. The science, however, is not sufficiently advanced to support causal connections or identify with certainty those characteristics of the built environment most closely associated with physical activity behavior. Thus, the committee is unable to provide specific policy guidance, although it offers several recommendations for strengthening theory, research, and data that should provide a firmer basis for future policy making and intervention.
The committee believes that the importance of physical activity to health warrants a strong and continuing research effort to further understand the relationship between the built environment and physical activity. If the field is to move forward, however, different kinds of collaboration and research are needed. First, a more interdisciplinary approach to research would help bring together the needed expertise of the public health, physical activity, urban planning, and transportation communities, among others. The committee found that the interdisciplinary character of its own membership greatly facilitated its understanding of the issues. Second, researchers should broaden their areas of inquiry to address the knowledge gaps identified in Chapter 6. Third, additional funding is needed to support difficult-to-finance multiyear longitudinal
studies, rapid-response capability to evaluate natural experiments as they arise, and extensions of national databases if important causal connections are to be researched.1 Among the committee’s key recommendations is support for a collaborative effort by the leadership of the Department of Health and Human Services and the Department of Transportation through an interagency working group to help develop and fund an appropriate research agenda for this purpose.
Finally, the committee wishes to emphasize that, as noted in Chapter 6, modifications to the built environment alone are unlikely to solve the public health problem of insufficient physical activity. Increasing populationwide levels of physical activity will require a range of approaches. Complementary strategies addressing the individual and social as well as the environmental determinants of physical activity behavior need to be the subject of future research and interventions. Such complementary strategies need to encompass leisure-time, home-based, transport, and occupation-based physical activity, given that a combination of physical activities in a variety of settings and locations can provide individuals with a feasible way to reach the goal of at least 30 minutes per day of moderate physical activity.2 The fact that this 30 minutes can be accumulated in segments of at least 10 minutes (see Chapter 2) means that all the activity need not be accrued in leisure time, at home, in transport, or at work, but can be spread across a range of locations where individuals spend their time.
Physical activity levels have declined sharply over the past half-century because of reduced physical demands of work, household management, and travel, together with increased sedentary uses of free time. Labor-saving technological innovations have brought comfort, convenience, and time for more leisure activities. They have resulted as well in more sedentary lifestyles with adverse health effects for many Americans. Changes in land use and travel may also have contributed to the decline in physical activity levels. For example, the steady dispersion of both population and employment to low-density suburban locations has increased reliance on the private vehicle as the dominant and most convenient travel mode. Rebuilding physical activity into the daily routine is a public health priority, but the specific contribution that the built environment could make is not well understood.
The built environment can facilitate or constrain physical activity. The built environment can be structured in ways that give people more or fewer opportunities and choices to be physically active. The characteristics of the built environment that facilitate or constrain physical activity may differ depending on the purpose of the activity. For example, ready access to parks and trails may facilitate walking for exercise; sidewalks and mixed-use development are likely to be more important to encourage walking for local shopping and other utilitarian purposes. The built environment can be changed in ways that increase opportunities for and reduce barriers to physical activity. The paradigm of the Robert Wood Johnson Foundation’s “active living” concept, for example, is to make opportunities for physical activity so pervasive that such activity is integrated into daily routines.
The relationship between the built environment and physical activity is complex and operates through many mediating factors, such as sociodemographic characteristics, personal and cultural variables, safety and security, and time allocation. Whether an individual is physically active is determined largely by his or her capacity, propensity, and willingness to make time for physical activity. For example,
while public health surveys have found that on average physical activity levels decline with age, many senior citizens remain physically active. Individual behavior is also influenced by the social and physical environment (see Figure 1-1 in Chapter 1). For example, the social disorder and deteriorated physical condition of poor inner-city neighborhoods deter physical activity for many residents. These neighborhoods have some of the physical characteristics thought to be conducive to walking and nonmotorized transport—sidewalks, multiple destinations within close proximity, and mixed land uses—and indeed, low-income urban populations report high levels of walking for utilitarian trips. However, they also report low levels of discretionary physical activity. Crime-ridden streets, littered sidewalks, and poorly maintained environments discourage outdoor physical activity other than necessary trips. Time is another mediating factor, cited by many as a reason for not being more physically active. For some (e.g., single parents, those holding two jobs), making time for physical activity is difficult. For others, particularly those who spend large amounts of leisure time on such sedentary pursuits as watching television, sedentary behavior may reflect the low priority given to physical activity. Time constraints often dominate the choice of travel mode, particularly for destination-oriented trips such as commuting and shopping, and influence destination and activity choices. In general, the role of time has not been well accounted for in examining the relationship between the built environment and physical activity.
The available empirical evidence shows an association between the built environment and physical activity. However, few studies capable of demonstrating a causal relationship have been conducted, and evidence supporting such a relationship is currently sparse. In addition, the characteristics of the built environment most closely associated with physical activity remain to be determined. Preliminary research does provide some evidence suggesting that such factors as access and safety and security are important for some forms of physical activity, such as walking and cycling, and for some population groups. However, the findings are not definitive because it is not known whether these characteristics affect a person’s over-
all level of physical activity or just his or her amount of outdoor walking and cycling. Furthermore, the literature has not established the degree of impact of the built environment and its various characteristics on physical activity levels; the variance by location (e.g., inner city, inner suburb, outer suburb) and population subgroup (e.g., children, the elderly, the disadvantaged); or the importance to total physical activity levels, the primary variable of interest from a public health perspective.
Weaknesses of the current literature include the lack of a sound theoretical framework, inadequate research designs, and incomplete data. The current state of knowledge in this area is limited in part by the lack of a sound theoretical framework to guide empirical work and inadequate research designs. As noted, most of the studies conducted to date have been cross-sectional. Longitudinal study designs using time-series data are also needed to investigate causal relationships between the built environment and physical activity. Studies that distinguish carefully between personal attitudes and choices and external influences on observed behavior are needed to determine how much an observed association between the built environment and physical activity—for example, in an activity-friendly neighborhood—reflects the physical characteristics of the neighborhood versus the lifestyle preferences of those who choose to live there. Appropriate measures of the built environment are still being developed, and efforts to link such measures to travel and health databases are at an early stage. For example, none of the national public health surveys on physical activity report its location or include such activity within buildings, although physical activity often takes place in workplaces, homes, and schools. Travel surveys are typically focused on purposeful travel and ignore physical activity for exercise or recreation.3 Another issue is the appropriate scale of the built environment to which physical activity data should be linked and the relevant environmental characteristics that should be included at each
scale. For example, design features that may encourage walking within buildings or at building sites are likely to differ from those that may encourage physical activity in a neighborhood or the larger community.
The built environment in place today has been shaped by long-standing policies and the practices of many decision makers (e.g., policy makers, elected officials, planners, developers, traffic engineers). Many existing development patterns have resulted from zoning and land use ordinances, design guidelines and funding criteria for transportation infrastructure focused primarily on motorized transportation, values and preferences of home owners and home buyers (e.g., suburban lifestyles, single-family housing), and racial and economic concentration of the poor and disinvestment in their neighborhoods. At the same time, the built environment is constantly changing as homes are renovated and new residences, developments, and office complexes are constructed.
Regular physical activity is important for health, and inadequate physical activity is a major, largely preventable public health problem.
The committee concurs with the strong and well-established scientific evidence linking physical activity to health outcomes and supporting reversal of the decline in overall physical activity levels as a public health priority. The connection between regular physical activity and health, although not the primary focus of this study, has clearly motivated interest in examining the built environment as a potential point of intervention to encourage more active behavior.
Built environments that facilitate more active lifestyles and reduce barriers to physical activity are desirable because of the positive relationship between physical activity and health.
Achieving this goal is challenging in a highly technological society with a built environment that is already in place and often expensive to change. Nevertheless, even small increases in physical activity levels can have important health and economic benefits. Moreover, the built environment is constantly being renovated and rebuilt and new developments are being constructed; these changes provide opportunities to incorporate more activity-conducive environments. In the committee’s judgment, such changes would be desirable even in the absence of the goal of increasing physical activity because of their positive social effects on neighborhood safety, sense of community, and quality of life.
Continuing modifications to the built environment provide opportunities, over time, to institute policies and practices that support the provision of more activity-conducive environments.
The long-term decline in physical activity among the U.S. population has been the cumulative result of many changes; thus there are many opportunities for intervention. However, some interventions will be easier to effect than others. For example, formidable hurdles would have to be overcome to substantially modify long-standing policies, such as the current system of zoning regulations and land use controls that reflects the preferences of many suburban home owners and buyers, to allow greater density of development and more mixed land uses. Similarly, many barriers persist to ending concentrations of minority populations and underinvestment in poor neighborhoods and the accompanying social and economic isolation of the poor. More flexible and targeted approaches—context-sensitive design, special overlay districts, traffic calming measures, community policing—have a better chance of gaining support. Construction of new buildings and developments offers promising opportunities for creating more activity-friendly environments. A wider range of such environments should become available as more neotraditional communities prove financially successful and employers embrace more walking-friendly office complexes to encourage healthier workforces.
Opportunities to increase physical activity levels exist in many settings—at home, at work, at school, in travel, and in leisure. The built environment has the potential to influence physical activity in each of these settings.
Each setting is characterized by different environmental opportunities and constraints that could affect physical activity levels. In some neighborhoods, for example, residents walk for utilitarian purposes. Keeping these neighborhoods safe and providing desirable destinations should help reinforce and perhaps enhance this behavior. In other neighborhoods, walking for utilitarian purposes is limited. In these settings, recreational walking and cycling may offer the greatest potential for increasing physical activity in the daily routine. Trend data from national public health surveys suggest that in fact leisure-time physical activity has increased slightly over the past decade, and the literature reveals that many characteristics of the built environment at the neighborhood level are significantly correlated with leisure-time physical activity and exercise. Of course, individuals can also obtain their daily physical activity by exercising at home. Most Americans spend the majority of their day at home, at work, and at school, and these are important but understudied locations for physical activity, particularly in view of the guidelines, which suggest that the daily 30-minute minimum of moderate physical activity can be accumulated in many locations and in small (10-minute) time increments.
Many opportunities and potential policies exist for changing the built environment in ways that are more conducive to physical activity, but the available evidence is not sufficient to identify which specific changes would have the most impact on physical activity levels and health outcomes.
Research has not yet identified causal relationships to a point that would enable the committee to provide guidance about cost-beneficial investments or state unequivocally that certain changes to the built environment would lead to more physical activity or be the most efficient ways of increasing such activity. Effective poli-
cies to this end are likely to differ for different population groups (e.g., children, youths, the elderly, the disadvantaged), for different purposes of physical activity (e.g., transportation, exercise), and in different contexts (e.g., inner city, inner suburb, outer suburb, rural). For example, much recent research has focused appropriately on physical activity at the neighborhood level, where many opportunities exist for walking and cycling for recreation or errands. However, home and work can also be locations for physical activity, as can travel itself between home and work or other destinations.
Given the current state of knowledge and the importance of physical activity for health, the committee urges a continuing and well-supported research effort in this area, which Congress should include in its authorization of research funding for health, physical activity, transportation, planning, and other related areas.
Priorities for this research include the following:
Interdisciplinary approaches and international collaboration bringing together the expertise of the public health, physical activity, urban planning, and transportation research communities, among others, both in the United States and abroad.
More complete conceptual models that provide the basis for formulating testable hypotheses, suggesting the variables and relationships for analysis, and interpreting the results.
Better research designs, particularly longitudinal studies that can begin to address causality issues, as well as designs that control more adequately for self-selection bias.
More detailed examination and matching of specific characteristics of the built environment with different types of physical activity to assess the strength of the relationship and the proportion of affected population subgroups. All types of physical activity
should be included because there may be substitution among different types. The goal from a public health perspective is an increase in total physical activity levels.
National public health and travel surveys should be expanded to provide more detailed information about the locations of physical activity and travel, which is fundamental to understanding the link between the built environment and physical activity in all potential contexts.
Geocoding the data on physical activity and health collected in large surveys, such as the Behavioral Risk Factor Surveillance System, the National Health and Nutrition Examination Survey, and the National Health Interview Survey, could help link these rich data sets with information on the built environment and the specific locations where physical activity is occurring. Similarly, travel surveys, such as the National Household Travel Survey, as well as regional travel surveys, should be geocoded to provide more fine-grained geographic detail so researchers can link these surveys and diary data with characteristics of the built environment. In addition, data that reflect a more comprehensive picture of physical activity should be provided. For the public health databases,this means capturing more than leisure-time physical activity; for the travel databases, a more complete accounting should be provided of walking and other forms of nonmotorized travel. More reliable and valid measures of the built environment, both objective and subjective, are also needed. Technologies are available to help verify the accuracy of self-reported data automatically and objectively. Ideally, both self-reported and objectively measured data should be collected. Self-reported data provide qualitative insights, such as trip purpose, that cannot be determined through technical measurement, while objective measures reduce the risks of respondent bias and provide a cross-check of survey responses. Finally, a new database—the Bureau of Labor Statistics’ American Time Use Survey—provides an opportunity to track detailed types and durations of respondent activities in many locations. With the collection of extensive demographic and socioeconomic data on
the respondents, the database offers researchers a more comprehensive picture of activities and time-use trade-offs by various subgroups of the population than has previously been available. Because the survey is new, opportunities exist to add questions related specifically to physical activity levels.
When changes are made to the built environment—whether retrofitting existing environments or constructing new developments or communities—researchers should view such natural experiments as “demonstration” projects and analyze their impacts on physical activity.
Numerous such opportunities exist, ranging from the construction of new, neotraditional developments to projects of the Active Living by Design Program of the Robert Wood Johnson Foundation.4 To take advantage of these natural experiments, baseline data must be collected. A “rapid-response” capability is needed so that timely funding can be made available to gather the appropriate data when opportunities arise. This might mean gathering data in both treatment and comparison communities prior to an intervention to provide before and after data for assessing impacts.
Leadership of the Department of Health and Human Services and the Department of Transportation should work collaboratively through an interagency working group to shape an appropriate research agenda and develop a specific recommendation to Congress for a program of research with a defined mission and recommended budget.
An interagency approach is needed because the necessary research does not fall within the purview of any one agency. The committee recognizes that funding for research is currently being provided by the Robert Wood Johnson Foundation and encourages its continuation. Additional funding is needed to enhance research and data
collection in several areas and provide a more solid foundation for policy making. An interagency initiative is likely to encourage a more interdisciplinary approach to the problem.
Federally supported research funding should be targeted to high-payoff but difficult-to-finance multiyear projects and enhanced data collection.
The highest priorities, in the committee’s judgment, include funding for multiyear longitudinal studies, a rapid-response capability to take advantage of natural experiments as they arise, and support for recommended additions to national databases. The federal government should supplement funding provided by foundations to ensure that this high-payoff research is conducted. The new National Institutes of Health initiative on obesity and the built environment is one possible funding source.5
The committee encourages the study of a combined strategy of social marketing and changes to the built environment as interventions to increase physical activity.6
The research should be designed to study these approaches both separately and in combination so that the influence of individual factors can be evaluated. To be effective, social marketing campaigns should be tailored to different population subgroups with relatively homogeneous characteristics and linked with other interventions involving the built environment for evaluation. For example, a social marketing campaign targeted to low-income, minority popula-
tions could be combined with a community policing effort to create safe havens for walking and studied for the effect on increasing physical activity levels in these communities. This targeted approach should prove more effective than mass messages about the benefits of being physically active. Possible audiences include but are not limited to (a) subgroups of the population segmented by gender, age, income, and race; (b) public and private officials responsible for community design, development, safety, and public health (e.g., elected officials, planners and planning boards, parks departments, local police, local public health officials, developers); (c) transportation infrastructure planners and providers (e.g., metropolitan planning organizations, traffic engineers and consultants), and (d) private employers responsible for workplace design and employee information programs and incentives.
Universities should develop interdisciplinary education programs to train professionals in conducting the recommended research and prepare practitioners with appropriate skills at the intersection of physical activity, public health, transportation, and urban planning.
Ideally, new interdisciplinary programs should be developed with a core curriculum that brings together the public health, physical activity, transportation, and urban planning fields in a focused program on the built environment and physical activity. At a minimum, existing programs in public health, transportation, and urban planning should be expanded to provide courses related to physical activity, the built environment, and public health. Similarly, practitioners in the field—local public health workers, physical activity specialists, traffic engineers, and local urban planners—could benefit from supplemental training in these areas.
Those responsible for modifications or additions to the built environment should facilitate access to, enhance the attractiveness of, and ensure the safety and security of places where people can be physically active.
Even though causal connections between the built environment and physical activity levels have not been demonstrated in the literature to date, the available evidence suggests that the built environment can play a facilitating role by providing places and inducements for people to be physically active. Local planning officials, as well as those responsible for the design and construction of residences, developments, and supporting transportation infrastructure, should be encouraged to provide more activity-friendly environments.