Urban Form Relationships with Youth Physical Activity

Typical Dependent Variables in Research on Environmental Correlates of Physical Activity Behavior

Studies of environmental correlates of physical activity specify a variety of different measures of physical activity as dependent variables. While inconsistencies in measure specification often limit direct comparability across studies, groupings of similar variables can be identified, as illustrated in Table 1.

OPEN IN VIEWER

Table 1.

Dependent Variables Commonly Used in Studies of Environmental Correlates of Physical Activity

Type of Dependent Variable	Examples
Overall volume of physical activity	Average daily minutes of light physical activity per day and average daily minutes of moderate-to-vigorous physical activity (MVPA), as in Hume and Ball (2005) and Jago et al. (2005)
Specific types of physical activity
Independent activities or mobility	Child allowed to cross roads or play outside without adults, as in O’Brien et al. (2000) Percentage of trips made alone, as in Clifton (2003)
Active transportation in general	Walking above a threshold distance or more than a minimum number of trips per specified time period, as in Frank et al. (2007) Walking frequency per week, as in Alton et al. (2007)
Active transportation to school	Number of times active commuting per week, as in Kerr et al. (2006) and Schlossberg et al. (2006) Mode of commute to school, as in Cooper et al. (2005)

The dependent variables identified in Table 1 are divided into two broad categories: measures of overall volume of physical activity (e.g., minutes of MVPA per day) and measures of specific types of physical activity (e.g., active transportation to school), each of which have associated benefits and disadvantages. Studies considering specific types of physical activity are important for understanding behavior-specific correlates of physical activity and behavior-specific policy responses. For instance, active transportation to school may be promoted through school siting policies and safe route to school programs, whereas independent mobility may be fostered through policies that address real and perceived safety more broadly, such as traffic calming initiatives. In this vein, de Vries et al. (2010) found different environmental correlates of walking and cycling among children according to trip purpose and commuting mode. However, studies using measures of specific types of physical activity also may need to be interpreted with caution because of the possibility of substitution between different types of physical activity. This possibility can be described in terms of the “activity stat” hypothesis, whereby an individual is hypothesized to maintain relatively constant total physical activity levels over time by compensating for physical activity in one time period with a corresponding reduction in activity in another period (Baggett et al. 2010). The alternative is the “displacement” hypothesis that conversely suggests that sedentary activities may displace physical activity and vice versa (Baggett et al. 2010). Further research is required to clarify which of these competing hypotheses is valid in specific contexts.

One type of physical activity that has been particularly well studied in this regard is active commuting to school. In one study, Cooper et al. (2005) found that both boys and girls who walked to school were more physically active in general than those who traveled by car. Similarly, Cooper found that boys (but not girls) who cycled to school were more physically active than those traveling by car. Cooper noted that the journey to school itself contributed relatively little to overall physical activity levels, possibly suggesting that active commuting might somehow engender in children a motivation to engage in additional physical activity. An alternate hypothesis is that naturally active children choose to travel by active means. Based on an evaluation of weekend physical activity data, Cooper concluded that observed differences in overall activity levels could not be explained by the natural differences in activity behavior of children, but rather was related to active commuting. Faulkner et al. (2009) based on a broader review of studies examining associations between active transportation to school and physical activity outcomes also conclude that children who actively commute to school tend to be more physically active overall than those driven or bused to school. More recent studies generally confirm this conclusion (Abbott et al. 2009; Chillon et al. 2010; Cooper et al. 2010b; Daly-Smith et al. 2010; King et al. 2010; van Sluijs et al. 2009) and thus appear to support the displacement hypothesis. Nonetheless, such findings do not necessarily extend to other activities.

Active commuting to school may be a special case because it is a regular, routine source of physical activity. Other incidental recreational activities may in contrast substitute for each other, with children compensating for an increase in one type of physical activity by limiting other types of activities, consistent with the activity stat hypothesis. Hargströmer et al. (2009) lend support to this possibility, finding that an organized weekly exercise intervention occurring over 13 weeks resulted in decreased total daily physical activity among obese adolescents participating in the trial. Other recent studies appear to support the displacement hypothesis by demonstrating that volume of physical is positively associated with outdoor play (Cooper et al. 2010a; Nilsson et al. 2009) and independent mobility (Page et al. 2009), while still others report null findings (Veitch, Salmon, and Ball 2010). In light of such mixed findings, the possibility of compensation introduces some degree of uncertainty as to the implications for overall volume of physical activity when specific types of physical activity are used as dependent variables. The time scale over which compensatory effects occur in particular requires further study, because they may occur over longer time frames than typically considered (Eisenmann and Wickel 2009) and further longitudinal studies may also be necessary to more conclusively evaluate these effects.

Characteristics of Ecological Models of Physical Activity Behavior

Ecological models share a number of important features in common. First, these models are premised upon the nesting of individuals within multiple levels of environments or sociospatial contexts (Sallis and Owen 2002). These environments may correspond to specific physical settings such as neighborhood parks or they may describe social environments such as workplaces or jurisdictions responsible for policies that influence behavior. For youth, the home and the school are likely particularly important environments because of the proportion of time spent in these settings. Of greater interest to planners, neighborhoods environments may also play an important role in shaping youth physical activity.

Most empirical studies of urban form influences on physical activity behavior focus on the neighborhood surrounding a child’s home (e.g., Roemmich, Epstein, and Raja 2007). However, it is logical that multiple neighborhoods where children spend their time would be relevant. The need to consider multiple neighborhood environments is highlighted by an ecological model proposed by Lee and Moudon (2004), the “Behavioral Model of Environment” or BME. While this model best describes environments relevant to travel between two points, it nonetheless provides a useful illustration of the importance of multiple environments for influencing physical activity behavior more broadly. In this model, environmental settings are identified based on the origin, the route, and the destination of a trip. According to this model, characteristics of each of these are highlighted as potentially influencing physical activity, in addition to characteristics of the areas surrounding the origin and destination (Lee and Moudon 2004). Applied to youth physical activity, the BME might for instance specify the following as important environments: a child’s home, their school, the route between the home and the school, and the neighborhoods surrounding both the home and the school.

An additional complication in identifying environments that may be relevant for influencing physical activity patterns is that some environments may have clearly definable boundaries, while others may not. An example of a clearly defined, discrete environment is the school, which may be delimited by a school building together with the school property it is located on. In contrast, the neighborhood around a child’s home cannot be clearly demarcated. For analytical purposes, neighborhood definition may involve use of predefined spatial units such as census tracts or municipally defined neighborhoods. The alternate, more common approach is to define a “buffer” centered around a child’s home, although there is no consensus on what size or shape this should take.

A second characteristic of ecological models of physical activity behavior, following from the first, is that specific characteristics of environments may be identified which influence physical activity behavior (Sallis and Owen 2002). An ecological model might therefore specify that in addition to individual characteristics (e.g., age and gender), household characteristics (e.g., household income and car ownership) and neighborhood characteristics (e.g., access to parks) all exert independent influences on physical activity patterns. In some cases, similar measures may be used to characterize multiple environments. For instance, measures of SES assessed at both the household and the neighborhood levels may be independently associated with physical activity patterns.

An Ecological Framework for Conceptualizing Correlates of Child Physical Activity

A framework for conceptualizing correlates of child physical activity behavior is presented in Figure 1. This model illustrates: (1) the nesting of children within multiple environments; (2) the hierarchical nesting of proximate within more extensive environments (e.g., the home and the school within neighborhoods); and (3) the social and physical components of environments.

OPEN IN VIEWER

Figure 1.

An ecological classification of correlates of child physical activity. Examples of correlates of physical activity are illustrated in the unshaded boxes to the right, corresponding to the categories indicated by the shaded boxes on the left. This review focuses on neighborhood built environment correlates (corresponding to the highlighted box).

Each of the boxes highlighted in this illustration represents major groupings of correlates of physical activity, corresponding to the individual child or the environments they are situated within. For example, correlates corresponding to the individual child include age, gender, weight status, and a variety of psychological characteristics (Kahn et al. 2008; Van der Horst et al. 2007). Relevant psychological characteristics include self-efficacy, which refers to an individual’s confidence in their ability to engage in physical activity (Strauss et al. 2001), and norms, which refer to the motivation to comply with the perceived beliefs of others (Sallis and Owen 1999). Individual characteristics also include perceptions of environmental characteristics such as perceptions of neighborhood safety, discussed at length below.

Physical activity behavior of children is also likely shaped by proximate environments they are regularly exposed to such as their home and school. Relevant characteristics of each of these environments may be further classified as either social or physical. Physical correlates corresponding to the school environment include availability of recreational features (Barnett et al. 2006; Nichol, Pickett, and Janssen 2009), school and play area size (Cradock et al. 2007), and classroom design (Lanningham-Foster et al. 2008). Social influences related to the school include peer and staff support (Barnett et al. 2006; Hohepa et al. 2007). With regard to the home environment, relevant physical characteristics include yard size (Spurrier et al. 2008), level of access to sedentary activities such as TVs (Singh et al. 2008), and availability of home exercise equipment, although findings in this regard are mixed (Davison and Lawson 2006). Social correlates associated with the home environment largely center around parents, who can influence physical activity patterns of their children through engagement and modeling of physical activity behaviors (Bauer et al. 2008; Hohepa et al. 2007; Ornelas, Perreira, and Ayala 2007).

Children’s homes and schools are in turn situated within neighborhood and regional environments. Relevant characteristics of neighborhood social environments include measures of neighborhood disorder (Molnar et al. 2004), neighborhood SES (Oliver and Hayes 2005), and neighborhood social interactions or social capital (Carver et al. 2005; Hume et al. 2009b). Social aspects of regional environments that may influence physical activity patterns include social marketing campaigns (Mccreedy and Leslie 2009) and school district policies concerning physical education. The focus of the following review is on the neighborhood physical environment, as indicated by the highlighted box in Figure 1. In particular, characteristics of the neighborhood built environment are focused on, in conjunction with correlates that likely interact with these characteristics to shape youth physical activity (e.g., parent and child perceptions and child age). In addition to elements of the built environment (urban form characteristics), the physical environment of neighborhoods and broader regional environments encompasses elements of the natural environment. Characteristics of the natural environment, which have been associated with child physical activity patterns include weather or season (Tucker and Gilliland 2007; Shephard and Aoyagi 2009) and topography (Timperio et al. 2006). Characteristics of the built environment thought to influence youth physical activity are now considered.

Access-Related Characteristics of Urban Form

Access-related characteristics of urban form are hypothesized to influence physical activity patterns by increasing the physical proximity of origins and destinations. This may translate into:

active transportation: for example, in higher density environments, it may be easier for children to walk or bike to a friend’s house than in lower density environments and/or

Design-Related Characteristics of Urban Form

As discussed above, measures of access are used to assess elements of urban form relating to the distribution of land uses and connections between them. In contrast, design characteristics correspond to smaller scale elements of urban form such as the availability of seating, the presence of street trees, safe street crossings, and adequate lighting. Two categories of design elements have received substantial research attention to date: street design and park/playground design. Each of these are now considered.

Studies Assessing Physical Activity Differences by Neighborhood Type

In contrast to the above studies focusing on the independent influences of individual built environment characteristics, some studies have examined differences in youth physical activity by neighborhood type. These studies are based on widely varying methodologies, both in terms of sampling strategies, and with regard to the level of complexity used to distinguish between neighborhood types. Some studies for instance rely on simple urban versus rural distinctions (e.g., Huang et al. 2010), while others use more nuanced definitions developed by assessing spatial covariation of built environment characteristics (e.g., Boone-Heinonen et al. 2010c). One example of the latter type of study is Nelson et al. (2006), which relied on cluster analysis to identify neighborhoods characterized by sociodemographic and built environment characteristics. In their final typology, neighborhoods were given labels such as “rural working class,” “exurban,” and “mixed race urban.” Nelson et al. found that adolescents living in older suburban areas were more likely to be physically active than those living in newer suburbs and also that adolescents living in lower SES inner-city areas were more likely to be active than residents of mixed race urban areas. Based on a recent review of such studies, Sandercock, Angus, and Barton (2010) conclude that substantial differences in physical activity levels between rural and urban areas are not generally evident. However, when studies distinguish between urban, rural, and suburban areas, suburban environments are generally characterized by higher physical activity levels in children (Sandercock, Angus, and Barton 2010). Such findings stand in contrast to the results of many studies focusing on individual urban form characteristics, since as described above, many of these imply that dense, mixed use environments with accessible recreational amenities are supportive of child physical activity. One possible explanation for such findings is that combinations of environmental characteristics may relate to activity levels in ways that are not evident when studying individual factors (Norman et al. 2010).

Recent studies not incorporated in the review by Sandercock et al. present complex and varying findings, pointing towards the need for further research in this area. For instance, in a study of adolescents in San Diego County, Norman et al. (2010) use latent profile analysis to define three neighborhood types: “open space,” “residential with cul-de-sacs,” and “housing and facility dense.” Among other findings, they note that boys were less sedentary in the open space and residential with cul-de-sacs neighborhoods than in housing and facility-dense neighborhoods. No differences in MVPA levels of boys were noted by neighborhood type. In contrast, lower MVPA levels were noted for girls living in the residential with cul-de-sacs neighborhoods when compared to those living in the open-space neighborhoods (Norman et al. 2010). Findings from a study of students in Alberta, Canada, indicate that students attending schools in towns and rural areas report more physical activity than students attending urban schools, despite less perceived access to playgrounds or parks and recreational programs (Simen-Kapeu, Kuhle, and Veugelers 2010). Finally, in contrast to these findings, Huang et al. (2010) based on a sample of grade 5 and 6 students in Taiwan found that urban children reported more physical activity compared to rural children.

Recent Methodological Developments

Two recent methodological developments highlight new research directions: the use of Geographic Positioning Systems (GPS) and the use of longitudinal studies and natural experiments as alternatives to typical cross-sectional study designs. A first step in the objective assessment of neighborhood environment influences on physical activity behavior is to identify a specific neighborhood or neighborhoods thought to be of relevance. These are typically defined as an area surrounding the home or school. However, such approaches are generally premised on untested assumptions about where children engage in physical activity. In contrast, the use of GPS enables researchers to track individuals as they move around in the course of their daily activities and thus more accurately identify relevant behavior settings (Cummins et al. 2007). Several recent studies have used GPS for this purpose (Cooper et al. 2010a, 2010b; Jones et al. 2009; Quigg et al. 2010). Certain findings from these studies support existing approaches and assumptions. For instance, Jones et al. (2009) found that children tend to be active close to home. In particular, they found that 63 percent of MVPA bout time occurred inside neighborhoods that were defined as the area within 800 meters from their homes along roads and designated public footpaths. Cooper et al. (2010a) also highlight the importance of outdoor environments as sites for physical activity with their finding that physical activity was more than 2.5 times higher outdoors than indoors. Conversely, other findings appear to challenge current assumptions and suggest the need for further research. In this vein, Quigg et al. (2010) found that only 1.9 percent of total daily physical activity occurred within city parks, which appears to run counter to studies discussed above which have found measures of access to parks or recreation facilities to be important correlates of physical activity. Findings of other studies using GPS also indicate relatively low proportions of physical activity in parks, although not as low as Quigg et al. (2010). Jones et al. (2009) for instance found that 7.3 percent of children’s MVPA bout time occurred within parks. Wheeler et al. (2010) similarly found that for boys, 9 percent of MVPA occurred within greenspace, and for girls, 6 percent of MVPA occurred within greenspace.

While the majority of studies to date have been based on cross-sectional research designs, some recent studies have employed alternative longitudinal or quasi-longitudinal designs tracking changes in behavior of individuals over time. As with studies using GPS, findings of longitudinal studies are relatively preliminary and mixed with regard to built environment influences on child physical activity patterns. Several such studies report complex, and in some cases, unexpected findings. For instance, Evenson et al. (2010) found that many of their hypotheses were not supported by their results that indicate for instance that the perception that streets were well lit was associated with steeper declines in nonschool physical activity. Similarly, the perception that a neighborhood had bicycle and walking trails was associated with a significant decline in physical activity and several other measures of perceptions of neighborhood characteristics were not associated with changes in physical activity (Evenson et al. 2010). Both Boone-Heinonen et al. (2010a) and Crawford et al. (2010) similarly report several null or counterintuitive findings in their longitudinal studies. Conversely, Carver et al. (2009) found several safety-related aspects of urban form to be supportive of physical activity for youth concluding that passive road safety interventions may be particularly important to promote physical activity among less active girls. The most beneficial urban form characteristics for physical activity in this study were speed humps, traffic/pedestrian lights and local (moderate speed limit) streets (Carver et al. 2009). Hume et al. (2009a) similarly report many intuitive findings including that children whose parents knew many people in their neighborhood were more likely to increase active commuting compared to other children. Such mixed findings highlight the need for additional longitudinal studies. Additional longitudinal studies could also benefit from the inclusion of a broader range of built environment measures and in particular, more objective built environment measures to enable further comparisons with findings from cross-sectional studies.

Numerous recent studies have also employed research designs based on natural experiments, which as noted by Petticrew et al. (2005) represent an underutilized tool in public health research, despite their potential to provide insight into the impact of urban form interventions. These are primarily observational studies examining the use of specific facilities following localized retrofits. Several of these have evaluated the influence of park interventions in particular. In one such study, Tester and Baker (2009) found that park playing field renovations including the installation of artificial turf and the addition of lighting, fencing and picnic benches, increased visitation, and overall physical activity. In their quasi-experimental study (employing both intervention and control schools), Brink et al. (2010) found schoolyard improvements including installation of age-appropriate play equipment and tree planting was associated with significantly higher volume of schoolyard use and higher activity levels of students. In another quasi-experimental study, Fitzhugh, Bassett Jr., and Evans (2010) report some null findings with regard to an urban form intervention, retrofitting a neighborhood with a greenway. In particular, although physical activity levels in the intervention neighborhood increased, no changes were noted in active transportation to school for students attending nearby schools.

Child Perceptions as Correlates of Physical Activity Behavior

Many studies have found that children’s perceptions of urban form characteristics are associated with physical activity behavior. For example, Timperio et al. (2004) found that girls who perceive that there are no parks in their neighborhood are less likely to walk or bicycle in their neighborhood. Similarly, Hume, Salmon, and Ball (2005) conducted a cognitive mapping exercise, in which children were asked to draw and in some cases photograph their home and neighborhood. Hume et al. found that those girls who identified more opportunities for physical activity in their neighborhood engaged in more objectively measured low-intensity physical activity. With regard to perceptions of traffic safety, Carver et al. (2005) found that girls who perceived local roads to be safe spent more time walking for exercise on weekends and for transportation on weekdays. Liu et al. (2007) similarly found a positive association between a child’s perception of safe drivers and pleasant walks in the neighborhood with self-reported physical activity. While not considering physical activity outcomes specifically, Mullan (2003) further highlights the importance of traffic safety in shaping activity patterns. Specifically, Mullan found that adolescents in Wales who reported living with busy traffic and car parking were less likely to have positive perceptions of their local area, effectively viewing it as less safe and pleasant for play. However, not all studies share similar conclusions. In contrast to these findings, several studies have found null associations between child perceptions of safety and physical activity outcomes, including Zakarian et al. (1994) and Mota et al. (2005). Such findings may be a result of conflicting parental perceptions, discussed below.

Children’s perceptions of their neighborhood conditions and opportunities for play are likely also shaped by characteristics of the social environment. In addition to examining perceptions of the traffic environment, Carver et al. (2005) examined associations between measures of physical activity and adolescents' perceptions of their neighborhood social environment. Carver and colleagues concluded that important explanatory factors for walking and cycling within their neighborhoods included adolescents' reports of having friends nearby, having young people of similar age nearby to socialize with, knowing neighbors, and waving or talking to neighbors. Kim et al. (2010) similarly found adolescent’s perceptions of how well residents know each other to be associated with the odds of meeting an MVPA guideline. Findings that adolescents' perception of crime threat in their neighborhoods are negatively related to active transportation to recreational sites (Grow et al. 2008) also highlight the potential importance of perceptions of the neighborhood social environment.

Children’s perceptions of characteristics of the neighborhood built and social environments are not solely shaped by the characteristics themselves but also by the matching of environmental features with individual characteristics such as their physical abilities and social needs (Kytta 2004). Thus, individual children with different interests and physical abilities may perceive their environments in diverse ways. Perceptions of opportunities for physical activity in the neighborhood environment may also be subject to characteristics of the household environment, including level access to sedentary activities in the house, which may moderate the exposure of children to their environment. For example, access to TVs or computers in a house with few restrictions may effectively “pull” children inside, limiting their exposure to characteristics of the neighborhood environment (Epstein et al. 2006; Roemmich, Epstein, and Raja 2007; Wong et al. 2010), although findings on the displacement of physical activity by TV viewing are mixed (Biddle et al. 2004; Smith et al. 2008).

Influences of Parental Perceptions on their Children’s Physical Activity Behavior

As noted above, parents may directly influence physical activity patterns of their children by preferring certain venues for play and accompanying their children to these venues. But parents also influence the independent physical activity patterns of their children through restrictions. As gate keepers for their children, parents are concerned with wide ranging characteristics of the neighborhood built and social environment, centering on safety (Carver, Timperio, and Crawford 2008a; Prezza et al. 2007). With regard to characteristics of urban form specifically, fear of injuries or fatalities from traffic are major concerns that parents have about active transportation to school or leaving children unattended outdoors (Ahlport et al. 2008; Gielen et al. 2004; Ridgewell, Sipe, and Buchanan 2009). Such fears are in turn influenced by urban form and traffic conditions. Timperio et al. (2004) found that parental perceptions of the need to cross several roads to reach play areas (for older girls) and lack of traffic lights or crossings (for older boys) were negatively associated with children regularly walking or cycling to local destinations. Carver et al. (2005) similarly found that parental reports of heavy traffic were associated with lower rates of cycling and walking. Parental concerns about traffic may also help to explain anomalous findings regarding the relationship between street connectivity and physical activity. Specifically, findings that connectivity is negatively or not associated with physical activity-related variables (Braza, Shoemaker, and Seeley 2004; Copperman and Bhat 2007; Timperio et al. 2006) may arise because parents perceive of cul-de-sacs as safer than through streets (Veitch et al. 2006) and may therefore consider neighborhoods with disconnected street networks to be safer for their children than highly connected networks. Further evidence of interactions between environmental characteristics and perceptions is provided by Panter et al. (2010b) who found that distance to school moderated the influence of certain attitudes and safety concerns on active commuting behavior of children.

As with children, parental perceptions of the neighborhood environment are not solely a function of urban form characteristics but likely depend on a number of other factors including perceptions of safety of the social environment. Major parental concerns relating to social safety include fears of “stranger danger” and bullying (Freeman 2006; Carver, Timperio, and Crawford 2008a). McDonald (2007b) and McDonald, Deakin, and Aalborg (2010) for instance found measures of social control and cohesion to be important explanatory factors for active transportation to school, suggesting that parents require a neighborhood with high levels of social trust to allow their children to walk or bike to school. Cradock et al. (2009a) similarly found social cohesion to be positively associated with participation in general physical activity. Another important characteristic that may influence parent’s perceptions about the safety of their children is the perception of the presence of other children (Salmon et al. 2007; Timperio et al. 2006). The presence of other children might for example make parents feel more confident about neighborhood safety because it suggests a level of confidence in other parents regarding neighborhood safety (Tranter 2006). Alternately, the presence of other children might be seen as important as it implies more “eyes on the street” (both those of the children and of other parents watching after them). The amount of time parents spend in their neighborhood may also influence their perceptions of the neighborhood environment. Lam (2001) for instance found that parents with full-time jobs felt their local environment was safer than those with part-time jobs. Lam hypothesized that this was because parents spending more time at home might be more likely to observe firsthand the dangers that their children are exposed to. Gebel, Bauman, and Owen. (2009) found that simply having children is associated with differences in environmental perceptions of adults. They noted that adults with children were significantly more likely than other adults to misperceive a high walkable environment as low walkable, speculating that due to time constraints, such adults would be less familiar with their environment. In general, it is likely that perceptions of risk and fears are modified by a range of individual factors including prior experiences and preconceptions (Loukaitou-Sideris 2006) as well as parent’s gender (Cloutier, Bergeron, and Apparicio 2010). In contrast to studies examining individual components of parental concerns, Kerr et al. (2006) developed a composite parental concerns scale incorporating measures of fears of stranger danger, gangs, bullying, traffic, availability of sidewalks, among others. Kerr and colleagues found that parental concerns as gauged by this measure were associated with active commuting.

Subsequent to forming perceptions about the environment, parents may choose to influence how their children engage with the environment through either defensive behavior (e.g., accompanying children during play or teaching children how to cross streets safely) or avoidance behavior (e.g., restrictions; Carver, Timperio, and Crawford 2008a). Parental restrictions on the independent mobility of their children are often conceptualized in terms of “activity or mobility licenses,” which include licenses to cross roads, use buses, go to school on their own, go to other places on their own, cycle on the road, and go out after dark (Hillman 1993; Jago et al. 2009). Specific restrictions that parents place on their children are highly dependent on the age and gender of their children, with greater restrictions on independent mobility generally imposed on younger children (Veitch, Salmon, and Ball 2008) and girls (Kytta 2004; Carver, Timperio, and Crawford 2008a; O’Brien et al. 2000). The influence of parental restrictions on child physical activity patterns was studied by Evenson et al. (2006) who found that girls were more likely to report physical activity when they had parents who allowed them to walk on their own, use public transit on their own, or walk/bike to transit from their home. Similarly, Carver et al. (2010) found constrained behavior to be negatively associated with active transport for male and female children as well as female adolescents. While such findings intuitively confirm the importance of parental influence on physical activity patterns, even this relationship may not be direct, because children may perceive and respond to their parental restrictions in a variety of ways.

Intersection of Parent and Child Responses to Environmental Characteristics

Given parental restrictions, children may respond in a variety of ways. Although parents may in some respects act as gate keepers, children may also extend the spatial extent of their activities through the influence of peers or completely independently (Cornell et al. 2001). Further, parents often do not have a good sense of what their children are actually doing, including how far or where they are going independently (Cornell et al. 2001). In many instances, children do not accept their parents' restrictions and may even prefer places outside of the control of their parents (Korpela, Kytta, and Hartig 2002). In a study of favorite places, Korpela, Kytta, and Hartig (2002) did not find associations between mobility licenses given by the parents and the distance of children’s favorite places from home. One reason for noncompliant reactions to parental restrictions may be that children perceive their environment in very different ways than their parents. Carver et al. (2005) found that adolescents had more favorable views of neighborhood safety than did parents and speculate that some discrepancies between parental concerns and adolescent behavior can be explained by the desire of adolescents to seek increased autonomy. Finally, parents may also not be able to enforce restrictions on their children, as a result of scheduling issues relating to their job or for other reasons. Such factors may underlie null findings of several studies with regard to associations between perceptions of safety and physical activity (Carson et al. 2010).

Aside from enforcing restrictions, parents may also influence the independent physical activity behavior of their children in other ways. For example, as illustrated in Figure 3, mode choice decisions of parents influence the extent to which children are chauffeured or engage in active transportation on their own. A complex suite of factors in turn influence parental travel habits and preferences. Once the sunk costs of car ownership are incurred, driving may be seen as a cheap and highly convenient mode of transportation for families with children. Johansson (2006), for example, found that the number of cars in a household was one of the most important variables correlating positively with chauffeuring of children. Similarly, Wen et al. (2008) found that number of cars in the household was associated with children being driven to school. Traveling by transit, bike, or foot with young children may be seen as tiresome for parents (Dieleman, Dijst, and Burghouwt 2002), while driving may be seen as relatively convenient.

Perhaps indicative of such mode choice preferences, a number of studies have found associations between car ownership and physical activity-related variables or found that car ownership moderates the relationship between urban form and physical activity (Frank et al. 2007; Kerr et al. 2007; Timperio et al. 2004). Frank et al. (2007), for instance, found that children in households with two or less cars were more likely to walk than those in households with three or more cars. Kerr et al. (2007) found that car ownership rates strongly moderated the influence of urban form on youth physical activity patterns. In general, Kerr and colleagues found weaker associations between built environment characteristics and physical activity for children in households with fewer cars, suggesting that youth in such households were much more likely to walk for transportation, regardless of the environmental conditions.

A variety of other factors likely also influence parental mode choice decisions, including household size. In households with children, the coordination of a variety of activities becomes necessary, reflecting complex intra-family dynamics and trip planning, which in turn may lead to complex trip chaining patterns (Lee and McNally 2006; Srinivasan and Ferreira 2002). Thus, parents with more children may be more likely to chauffeur their children because of the convenience of trip planning. Srinivasan and Ferreira (2002) provide evidence of substantial trip chaining in families with children, finding that less than 6 percent of tours in two-worker households with children were without nonwork activity chaining. Finally, an extensive body of research has found that urban form also influences mode choice of adults (Frank, Engelke, and Schmid 2003), including parents. In general, a highly dispersed urban form with little mixing of uses is associated with higher rates of car use and lower rates of active transportation (Frank, Engelke, and Schmid 2003). With regard to the driving of children in particular, trends toward increased structured activities rather than unstructured play (Prezza 2007) may result in more adults chauffeuring their children as such activities are often located at specialized facilities that may be seen as being too far from home for children to transport themselves. In a study of children’s mode choice for getting to soccer games, for example, Tal and Handy (2008) found that driving was the predominant mode of transport, with 76.8 percent of players driven versus 18.4 percent biking and 4.8 percent walking. Major factors underlying this modal split included distance to the game and logistical barriers such as the need to carry equipment and snacks (Tal and Handy 2008).

Behavioral Reinforcement through Feedback: the potential for Social Traps

Figure 3 illustrates two feedback loops, one corresponding to parental perceptions and the other to child perceptions. Each of these feedback loops may operate through two mechanisms. First, they may arise as a result of the behavior of individual children and their parents. For example, the more mobile individual children are, the more likely they are to acquire environmental knowledge, which in turn creates an increased motivation to explore and be more mobile (Kytta 2004). Similarly, increased exposure to rewarding activities in the neighborhood may increase the motivation of children to be outside (Epstein et al. 2006). In contrast, restrictions on mobility may set in place a vicious cycle, whereby individual children lose interest in their environment and become less motivated to explore it (Epstein et al. 2006). Similarly individual parents accustomed to supervising their children may find it difficult to let them go out on their own. As hypothesized by Prezza et al. (2001), parents' confidence in their child’s ability to be responsible on their own may set in motion a virtuous cycle whereby acquisition of self-confidence on the part of children increases in tandem with parental increases in confidence.

An alternate mechanism through which these feedback loops may operate is in response to the aggregate behavior of other children and parents in the neighborhood. Neighborhoods with many children playing outdoors may, for instance, be more conducive to independent play, since children have more options to go outside and play with other children at their leisure than children living in neighborhoods with few other children playing outdoors. Carver et al. (2005), for instance, found various physical activity-related variables to be associated with child reports of many peers to hang out with locally. Santos et al. (2009) similarly found that seeing people being physically active in the neighborhood is positively related to physical activity for boys. While the presence of children in a neighborhood might thus set in motion a positive feedback by encouraging more parents to let their children out, the absence of children might conversely set in motion a vicious cycle, whereby parents are afraid to let their children out as long as no one else does. This interaction may be further nuanced, reflecting the demographic composition of children outdoors. Veitch et al. (2006), for instance, note that parents report the presence of teenagers at parks as a deterrent to the use of these facilities by younger children, as a result of fears of undesirable behaviors such as bullying, swearing, and drinking alcohol.

Concerns about traffic safety may similarly lead to a vicious cycle. In this case, increasing numbers of parents chauffeuring their children reduces the safety of the neighborhood environment for child pedestrians and cyclists, further encouraging other parents to drive their own children (Kearns and Collins 2006). Aside from the increased risk of injury or death that chauffeuring creates for children not being driven, such activity also contributes to decreased air quality and fears about exposure to air pollution may further contribute toward parental restrictions on their children’s outdoor physical activity (TRB 2005). The net effect of all of these feedback loops may be what has been described as social traps (Carver, Timperio, and Crawford 2008a) whereby chauffeuring and mobility restrictions become entrenched behaviors over large areas. In parallel with such changes, children may also become increasingly involved in structured activities after school in place of unstructured activities (Prezza 2007). These activities may be characterized by high levels of parental chauffeuring by car (Tal and Handy 2008), further reinforcing social traps that discourage independent activities by children within the built environment.

The Case for Urban Form Interventions

Despite the uncertainty resulting from limitations of existing research, a number of important considerations lend support to the potential for urban form interventions to influence youth physical activity. First, in contrast to many other correlates of physical activity (e.g., age, gender, and weather), urban form characteristics are generally modifiable. Existing urban environments are changing all the time and decisions affecting the design of new communities are made on a regular basis, presenting opportunities for interventions. Second, even if urban form changes cannot be immediately causally linked to behavior changes, this may be a result of social traps or other factors, and urban form changes may at least be enabling of future behavior change by creating more settings and opportunities for physical activity. Conversely, urban form may act as a barrier to future behavior changes, and so urban form interventions may be a prerequisite for the success of other joint interventions targeting behavior change more directly. For example, initiatives persuading parents to encourage their children to substitute outdoor play for indoor sedentary activities might be futile and potentially dangerous if the targeted families live in neighborhoods with heavy traffic and few parks. Third, although long periods of exposure to environmental changes may be required before behavior change occurs (Spence and Lee 2003), urban form interventions may also have longer lasting effects than many alternative interventions because of the permanency of the built environment (Saelens Sallis, and Frank 2003). Further, the types of physical activities that occur within the built environment might also be easier to adopt and maintain than other activities. These include activities such as walking and cycling to school, which can easily be incorporated in daily routines as well as other more incidental activities like unstructured play. As noted by Deforche, De Bourdeaudhuij, and Hills (2007) such lifestyle activities provide the opportunity for children to divide energy expenditure into several small bouts as opposed to a single bout as in structured activities, a pattern of activity more consistent with children’s observed physical activity patterns. Fourth, ecological models suggest that higher level environmental interventions influence greater numbers of people than lower level environmental and individual interventions (Spence and Lee 2003). Urban form interventions can thus influence entire populations, in contrast to interventions which target individuals. Urban form interventions to develop freely accessible public infrastructure (e.g., creating new parks or installing traffic calming as opposed to investing in structured recreational facilities that charge entry fees) may be especially important in this regard, since people from all socioeconomic backgrounds may benefit from the changes. Finally, urban form interventions to promote youth physical activity may also promote physical activity among adults, as many of the correlates are shared. To highlight the potential for such broader benefits, Watson and Dannenberg (2008) estimated that in large and small urban areas in the United States, approximately 65.5 million people could benefit from Safe Routes To School projects implemented within 0.5 miles of schools.

The Environment or Perceptions of the Environment—Which Should be Targeted?

Research to date suggests the importance of interventions that target both objective characteristics of the environment and perceptions. In particular, it may be necessary to target perceptions that diverge widely from objective characteristics of the environment. Kerr et al. (2006), for example, note that their findings suggest the possibility of overly concerned parents preventing their children from walking despite supportive built environments. Carver, Timperio, and Crawford (2008a) similarly observe that parents often exaggerate the risk of stranger danger and become overly anxious about the safety of their children as a result. As indicated above, such fears may lead to social traps whereby built and social environments become progressively less conducive to physical activity. Interventions addressing perceptions may therefore play an important complementary role to built environment interventions. Conversely, objective characteristics should clearly be targeted where these characteristics present real risks or barriers to children safely engaging in physical activity. As noted by Kerr et al. (2006), interventions addressing parental perceptions without ensuring the safety of children through environmental interventions may put children at unnecessary risk. For example, while programs promoting active transportation to school through encouragement by parents or teachers may be desirable, simultaneous improvements to street design to enhance safety may also be necessary. Child pedestrian or cyclist injury rates need to be interpreted with caution since these reflect both the traffic risks and the level of exposure of children to these risks (Sonkin et al. 2006). Thus, low rates of injury or death do not necessarily reflect a safe environment but may also arise from few children being exposed to these risks. Finally, programs designed to enhance safety need to be carefully designed to ensure that the desired safety benefits are well established. Dumbaugh and Frank (2007) note that many safety benefits associated with Safe Routes to School countermeasures are assumed rather than known, suggesting the need for further research in this area.

Planning Policies to Promote Physical Activity

Based on an ecological model of behavior, interventions to promote physical activity by modifying urban form may occur at multiple levels of environmental influence, including for example at the state level via growth management legislation, at the city level via zoning bylaws, or at the neighborhood level via installation of local traffic calming infrastructure. Table 2 categorizes and presents examples of such policies. Additional examples of planning policies designed to promote physical activity are summarized by Frank and Kavage (2009). Many of these may be applicable to children either directly or via their influence on parents. Barriers to implementing such policies are highlighted by Grant et al. (2010), while Mccreedy and Leslie (2009) and Guldbrandsson, Wennerstad, and Rasmussen (2009) present specific examples of policy implementation and evaluation.

OPEN IN VIEWER

Table 2.

Examples of Planning and Complementary Policies to Support Child Physical Activity

Type of Policy	Possible Health-Oriented Benefits	Elaboration
Built environment—access
School location policies	To enable active transportation to school and greater access to school sites as recreational destinations outside of school hours	Current school siting policies often favor large schools located on the edge of urban areas (Vincent 2006). Alternate policies might be developed to take into account impacts of siting decisions on the ability of children to actively commute, as recommended in the recent White House Task Force on Childhood Obesity Report to the President (2010). School siting standards and recommendations for planners are discussed below and by McDonald (2010)
Comprehensive plans and zoning bylaws	To enable active transportation in general, for both children and adults, and therefore to create a safer traffic environment	Comprehensive plans and corresponding land use regulations may be designed to promote high density, mixed use development that support greater use of non-motorized modes of transportation. Examples include the City of Portland’s comprehensive plan (City of Portland 2006) and urban growth boundary (Jun 2004)
Built environment—design
Street design retrofits	To promote active transportation, and/or to increase safety for children playing outdoors	Wide ranging policies have been implemented to retrofit streets for increased safety and to promote active transportation, including complete streets ordinances (National Complete Streets Coalition 2010) and traffic calming or home zone initiatives (Biddulph 2010)
Safe routes to school (SRTS)	To promote safe active transportation to school	These are primarily transportation infrastructure improvement programs such as sidewalk gap closures and intersection safety upgrades, and have been implemented extensively throughout the United States (discussed below). Martin, Moeti, and Pullen-Seufert (2009) presents an overview of SRTS implementation considerations
Park or recreational area investment	To promote the use of parks in general and/or to support higher intensity activities by users	Park and playground renovations may be used to improve existing underutilized recreation amenities. These include improvements in playing surfaces, addition of lighting, installation of age appropriate play equipment and tree planting (Brink et al. 2010; Tester and Baker 2009)
Process-oriented policies
Health impact assessments (HIAs)	To account for health benefits and costs that might otherwise be unaccounted for in planning projects	HIAs are formal tools that can be used to link public health and planning through the evaluation of health impacts of planning initiatives such as comprehensive and transportation plans. Forsyth, Slotterback, and Krizek (2010a, 2010b) present options for the incorporation of HIAs in planning processes. In addition, While not presented within the context of HIAs, Boarnet, Greenwald, and McMillan (2008) and Guo and Gandavarapu (2010) provide examples of specific economic evaluation tools that might be used as part of the evaluation of health impacts of planning initiatives
Participatory planning	To identify and address unique local concerns regarding health and safety	Given the diversity of factors potentially influencing physical activity and travel behavior in neighborhood environments, participatory planning efforts may be critical to identify barriers and opportunities unique to a particular locale (Morabia and Costanza 2009). Such initiatives might also be designed to simultaneously address both perceptions of environmental characteristics and the environmental characteristics themselves. Participation of youth themselves may be particularly valuable (Bridgman 2004; Francis and Lorenzo 2002)
Complementary policies—supporting but not directly influencing built environment characteristics
Joint use agreements	To enable broader community access to existing recreational resources	Joint use agreements are formal agreements between two government bodies establishing terms for the shared use of public properties (NPLAN 2009). Such agreements may for example be established to enable public use of school recreational facilities after hours (NPLAN 2009). Such initiatives may be especially important by enabling the greater use of existing recreational resources in built out areas where development of new recreational facilities is not an option
Walking school buses	To complement safe routes to school and related policies by addressing parental safety concerns and time constraints	While SRTS and related initiatives may help to enable active commuting, they may not be adequate on their own because of parental time constraints and safety concerns (McDonald 2008b). Walking school buses are programs in which a group of children walk to or from school attended by one or more adults, thus reducing the time commitment of individual parents while supporting active transportation (Kong et al. 2009)

Ecological models also suggest the possibility of multisectoral interventions, involving agencies dealing with transportation, land use planning, parks and recreation, and more diverse stakeholders such as public health agencies, schools, and media (Pate et al. 2000; Sallis and Owen 2002). Collaborative efforts among diverse stakeholders may lead to more effective or complementary policies. Several examples illustrated in Table 2 might be developed through such collaborative efforts, including the evaluation of health impacts as part of a comprehensive planning process and the implementation of WSB programs in conjunction with SRTS programs. Additional examples of policies that might complement built environment interventions include the use of recreation facilitators to connect children to existing recreational resources (McNeil et al. 2009) and programming of recreational activities to further support the use of these resources (Cohen et al. 2009; Tester and Baker 2009).

Policies to Promote Active Transportation

Initiatives targeting active transportation are particularly widespread and evidence to date generally confirms their effectiveness. Key among these are SRTS programs administered by State Departments of Transportation (DOTs) and funded through the Federal Highway administration. Evaluations of California’s SRTS program have found certain traffic safety improvement projects (e.g., sidewalk gap closures and replacement of four-way stops with traffic signals) to be associated with increased active transportation to school (Boarnet et al. 2005a, 2005b). As part of the nationwide SRTS program, $370.6 million has been awarded up to January 1, 2009, to more than 4,566 schools throughout the United States by State DOTs (National Center for Safe Routes to School 2009). According to the National Center for Safe Routes to School, there is substantial unmet demand for SRTS programs, with only 37 percent of programs applying to state DOTs for funding being selected in the last quarter of 2008 (National Center for Safe Routes to School 2009). Further, even among successful applications, many schools have only received a portion of the requested funds (National Center for Safe Routes to School 2009).

Although such initiatives are important and in need of additional funding, their effectiveness may be limited without supporting policies. In addition to design-related retrofits supported by SRTS initiatives, those initiatives targeting access need to be addressed. Improving access implies more comprehensive land use initiatives such as revised school siting policies. As noted by McDonald (2007a), distance to school has increased and may account for half of the declines in active commuting to school in recent decades. The increasing distance that youth have to walk reflects policies and decisions that have encouraged the closure of old neighborhood schools and construction of new schools on large campuses, typically sited on the outskirts of urban areas (MacDonald 2007a). Specific reasons include school siting standards (e.g., minimum acreage standards), school funding formulas, and lack of coordination between school officials and planners (National Center for Safe Routes to School 2007). Such practices have resulted in an approximately 58 percent reduction in the number of all elementary and secondary schools between 1940 and 2003, even while the population of students almost doubled (National Center for Safe Routes to School 2007). The effects of these practices are dramatically highlighted by Falb et al. (2007), who found that in Georgia, only 6 percent of elementary school students, 11 percent of middle-school students, and 6 percent of high school students could reasonably be expected to walk to school. The location of schools may also raise safety concerns since schools located near multilane high-speed roads may result in high risk of involvement of child pedestrians/cyclists in traffic accidents (Abdel-Aty, Chundi, and Lee 2007). The need to revisit school location policies sooner rather than later has been highlighted by McDonald (2008a) who notes that substantial new school construction and renovations are currently planned across the United States. McDonald considers the increased development of community schools as one possible solution (McDonald 2008a).

Options for Focusing Policies

Given limited public resources, it may be necessary to focus policies by targeting specific populations of youth at risk of physical inactivity, including those living in low SES neighborhoods (Poulsen and Ziviani 2004). Because access to recreational areas may be an especially important correlate of youth physical activity, spatial patterns in the distribution of these areas has been examined, with some evidence indicating that fewer recreational resources are located in low SES areas (Gordon-Larsen et al. 2006), although evidence todate is mixed (Dowda et al. 2007). Nonetheless, even if recreational resources are distributed relatively evenly across neighborhoods by SES, these spaces may vary in quality or safety, with the implication that some areas may benefit from interventions targeting design of recreational facilities rather than access to them. Crawford et al. (2008), for instance, found that public open spaces in high SES neighborhoods had more amenities, were more likely to have shade trees, water features, walking and cycling paths, and lighting among other features than those in lower SES neighborhoods, concluding that open spaces in high SES neighborhoods are equipped with more features likely to promote physical activity among children. Neckerman et al. (2009) similarly found that low-income neighborhoods in New York City are less conducive to walking than they would appear if only considering measures of access like density and land use mix. Among other findings, Neckerman and colleagues note that poor census tracts had fewer street trees, landmarked buildings, and clean streets, further highlighting the importance of urban form interventions in low SES neighborhoods in particular. Similar findings indicating that recreational amenities in low SES or inner-city neighborhoods are less safe or of lower quality have been noted by Cutts et al. (2009), Franzini et al. (2010), and Holt et al. (2009).

Initiatives might also be focused by specifically targeting older youth and girls since these too constitute populations at risk of inactivity. Urban form interventions could thus adopted to design recreational facilities specifically for these populations. Ries et al. (2008) highlights the need for such policies in a study of use of recreational facilities by urban African American adolescents. In this study, Ries et al. noted that many small neighborhood parks are designed for young children and are lacking in amenities desired by older youth. Ries et al. also found that young women rarely use outdoor facilities on their own but that indoor facilities provide positive social settings for both young men and women (Ries et al. 2008).

Finally, given the possibility of social traps arising from feedback loops as illustrated in Figure 3, initiatives may only be effective if they change the behavior of a critical mass of the population of an area. This also implies the need to prioritize environmental interventions, although in a potentially different sense than targeting at-risk populations. Specifically, this implies the need to target interventions spatially. That is, in order to reverse cycles that lead to social traps, resources should be geographically targeted in areas likely to benefit the most. Such interventions may have the added benefit of enhancing overall safety of area residents because the likelihood of a pedestrian or cyclist being struck by a motorist varies inversely with the amount of walking or cycling (Jacobsen 2003).