An assessment of schoolyard renovation strategies to encourage children's physical activity

Obesity has become an increasingly troublesome health problem in both wealthy and poor regions around the world [1]; the World Health Organization reports that there are now over one billion overweight adults, at least 300 million of whom are obese [2]. Obesity presents a particularly alarming health concern in the United States, where recent estimates are that approximately one third of children in the United States are considered overweight or obese [3].

Policy designed to change obesogenic environments aims to implement change that both reduce energy intake (by encouraging a healthy diet) and provide opportunities for increased energy output (by encouraging physical activity) [4]. Although behavioral change is a critical component to addressing obesity, interventions designed to modify individual behavior to reduce caloric intake have had limited success in preventing obesity on a long-term basis [3]. Although there is disagreement about which side of the energy equation is the most effective in terms of policy, extensive research has demonstrated that the built environment plays a key role in obesity-related behavior (e.g., [5–8]).

While there may be only a weak to moderate link between physical activity and obesity rates, there are numerous additional health benefits associated with increased physical activity. A recent literature review reports that physical activity is linked with reduced blood pressure, lower levels of cholesterol and blood lipids, reduced incidence of metabolic syndrome, increased bone mineral density, as well as reduced rates of depression [9]. Understanding the relationship between the built environment and physical activity and the specific implications of particular modifications to the built environment may contribute to strategies to reduce obesity prevalence [10] in addition to these other health benefits.

Along with transportation patterns and land-use patterns, design features are one of the key areas of inquiry in studies of the built environment and physical activity [11–13]. A number of specific design features pertaining to children's environments have been investigated in previous research. Time spent outdoors, access to recreational facilities and schoolyards [8, 14] and proximity and number of play spaces and facilities to home [15] are associated with higher levels of physical activity in children and adolescents. Adding an additional recess period each day is associated with greater physical activity [16], while limited outdoor play time has been found to correlate with a high body mass index in young children [17]. If it is designed well, the outdoor built environment can create opportunities for healthy behavior change among children.

Compared to white children, children from African-American and Hispanic ethnicities are particularly likely to suffer from obesity, experience abnormally high glucose levels, and suffer from a higher prevalence of diabetes [18–20]. These observations could be attributed in part to the social and built environments of many minority children living in impoverished urban neighborhoods, which often fail to support healthy development and provide limited opportunities for healthy behavior, especially with respect to physical activity. Poor and minority children often have limited access to outdoor play spaces and structured opportunities for involvement in organized sports and other activities [19] and are more likely to have lower fitness levels. The reasons behind these disparities are nuanced and complex, but it might be possible to develop planning and design policies to address geographic inequality. Some research has shown that parents in low-income neighbourhoods have increasingly restricted their children's activity out of concerns for safety [21], and that design policies can have an impact on this concern [22]. Interventions to improve the safety of schoolyards have been shown to improve schoolyard utilization [23]. Indeed, communities that are designed to support physical activity have been found to have 100% higher rates of sufficient physical activity than those with no supportive attributes [8]. Facets of the built environment, such as the density of residences [24], general walkability [25], and the availability of recreational spaces and facilities [26, 27] may be also linked to physical activity, although research often yields mixed results [21].

With the average child spending 1300 hours at school each year, schools are a valuable physical environment and social resource in efforts to promote physical activity. School wellness policies, community building initiatives, and walk-to-school programs [21] are a few examples of school-based strategies designed to encourage physical activity. In spite of a flourishing body of social-scientific and public health literature examining the relationship between physical activity and urban environments, there remains little work that specifically examines the impacts of recreational space renovation at schools. There is evidence that renovated schoolyards are more widely used by adults and children (especially boys) than un-renovated schoolyards [28]. Small, inexpensive interventions that have changed the structure of the physical environment, especially within the school environment, have shown significant positive correlations with physical activity levels. These interventions include painting the schoolyards [29], providing game equipment [30], and even increasing the number of balls available to youth [31]. Higher physical activity levels have been observed in schoolyards that have multicolored painting compared to those without [32]. The incorporation of culturally-tailored schoolyard elements may also encourage physical activity and ultimately contribute to the reduction of obesity [33], which is especially important in communities with large ethnic minority populations. In a study that used the same observation method as the current one, Sallis et al. conclude that making improvements to school environments could increase the physical activity of students throughout the school day [34]. After observing physical activity across twenty-four middle schools in San Diego, they determined that physical amenities, such as area type, area size, and permanent improvements (such as basketball courts and football goals) were associated with increased physical activity among both boys and girls. The authors conclude that "if we build it, they will come" [34]; that school design and renovation efforts may improve physical activity among children.

Because inclusion in the LL program was initially provided for low-income schools, the schools eligible for this study were located in deprived neighborhoods, populated largely by ethnic minorities. Study schools are therefore located in Denver neighborhoods facing significant social, economic, and educational challenges. Study sites were chosen from three different locations in Denver. Group A schools are from a predominately African-American neighborhoods characterized by significant gang activity. Group B and C schools are in neighborhoods located between three and five miles from downtown Denver and are comprised of poor, predominately Latino neighbourhoods. Schools in groups B and C are similar with respect to income and ethnicity, but are distinguished by school size (group B schools have an average attendance of 559 and group C schools an average attendance of 395). These groups were formed in order to select adequate controls on the basis of income, ethnicity, and school size. Each of the groups contains one school that had LL construction within a year before the data were collected, one school with older LL construction (constructed two or more years prior to the newly-renovated schoolyards) and one school without renovated school grounds (i.e., with no LL construction). Children observed in this study were elementary school students, between six and eleven years old.

Children were observed using the System for Observing Play and Leisure Activity in Youth (SOPLAY). SOPLAY is a quantitative method for evaluating levels of physical activity of individuals and associated environmental characteristics in play and leisure environment [36]. Target areas are predetermined and defined as "locations likely to provide opportunities for students to be physically active" [37] in which observers record the number of individuals present, their activity levels, and their gender. Each school yard was divided into activity zones that expressed the ground plane condition and the type of activity occurring. The observation protocol requires the identification of schoolyard variables with the greatest impact on children's physical activity based on area type, size, and permanent improvements. Observers were trained in SOPLAY observation methodology by a certified SOPLAY instructor. Children who are observed to be sedentary or walking are not considered to be physically active and children who are observed engaging in vigorous physical activity or in a "primary activity" such as using the schoolyard equipment such as a swing or jungle gym, are considered to be physically active. While the observers are able to distinguish between children and adults, they were not able to account for the age of specific children.

Observations were conducted over four days at each school to obtain accurate measurements. Two observers simultaneously observed the activity area for 20% of the total data collection time to test the reliability of the data, resulting in a reliability estimate of 87%. Observers were not part of the research team to ensure accuracy of the data collected. Data were collected from schoolyards at participating elementary schools in Denver between September 19 and October 29, 2005 and between September 29 and October 19, 2006 at regular time periods before the beginning of school, during school recess, after the school finished, and during the weekends. Because the observations covered the entire schoolyards, total schoolyard use and the number of children on the schoolyard at any particular observation point could be estimated from these surveys. Observational scans capturing activity on the entire schoolyard were treated as the unit of analysis. Between 28 and 30 schoolyard observations were conducted at each site.

In light of the obesity epidemic and its associated health impacts in the US, parents, educators, and health researchers strive to discover which environmental interventions encourage physical activity among children. Among the ways that design strategies can contribute to addressing this problem, schoolyards can be constructed to encourage children to spend time outdoors and to actively utilize schoolyards. Evidence-based evaluation of the behavioral impacts of specific design initiatives is particularly important in a context in which funding is scarce. Such evidence is often not available or accessible to private- and public-sector policymakers [39].

While urban designers have discussed how specific designs of children's spaces might encourage healthy behavior, few studies have evaluated the effectiveness of these designs once they have been implemented. Outside of some evidence that the addition of color is related to increased utilization [32, 40], little work addresses the impacts of specific design plans, either in terms of particular schoolyard components or as a broader design strategy. With the notable exception of a study from Cleveland [28], there has been little previous work on the impacts of renovation on schoolyard utilization. LL is a program which places high-quality recreational spaces into public schools serving Denver, including those with low-income populations. It was observed that LL schoolyards experienced significantly greater utilization for all observation periods over unrenovated spaces. When "optional" periods (before school, after school, and on weekends) are considered in isolation, greater utilization was generally observed for both boys and girls, with a few exceptions noted in the results.

There is evidence that boys are more likely than girls to exhibit greater utilization and more vigorous physical activity in both schoolyards in general [40] and on renovated playgrounds in particular [28]. While it remains unclear at what age weight-related differences in physical activity are apparent in girls [41], it is widely accepted that physical activity leads to benefits in physical health as well as social development [42, 43]. In this study, greater utilization by girls was observed on LL sites before and after school and during lunch recess, but not on weekends. The reasons behind these findings merit additional scrutiny on several levels. Further analysis might reveal, for example, that specific features of the LL schoolyards encourage girls to use the sites in an active manner. Schoolyard use was generally much higher before school than it was after school. It seems plausible that girls in particular are more likely to utilize these spaces before and after school, in contrast to weekends, as a consequence of their family's perceived safety of having other children and school staff present. Low-income school districts that wish to implement extra-curricular programs to improve physical activity among their students, and among girls in particular, might benefit from research focused around this question. Future work might also investigate how age affects girls' utilization patterns.

In order to address the question of how renovation affected children's likelihood to engage in physical activity, we also examined the percentage of children who were moderately or vigorously active as a proportion of all children observed on the schoolyards. This measure is intended to estimate the way that a particular space influences the type of activity that children engage in; some play spaces might contain features that encourage children to participate in specific activities that are conducive to vigorous physical activity, for example. No statistically significant differences were observed in this measure between any of the comparison groups. It seems plausible that there are differences in how children utilize specific components of various play spaces, but this may not be observable at the scale of an entire school yard. A schoolyard zone containing slides, for example, may encourage more or less activity than another zone containing a jungle gym. Work to compare activity across schoolyard zones that contain different features might be a fruitful avenue for additional inquiry.

An additional feature of this research is the measurement of the impacts of the recency of schoolyard construction. Utilization of LL schoolyards constructed within the previous year was compared with schoolyards constructed between three and four years prior to the study period. Greater utilization was observed on LL sites that newer than one year than on the older LL sites, but the differences were not statistically significant, with the exception of the utilization measures among boys. If the differences observed in this work and evidence from other studies, such as Colabianchi et al.'s [28], are seen to provide evidence that schoolyard construction results in greater utilization and activity, schoolyard renovation itself may be a valid planning strategy for increasing utilization. Additional study is required to determine the duration of the impacts of renovation and whether these differences are the consequence of particular components of the renovated schoolyards.

While encouraging physical activity among children is an important public health goal that may be addressed through a careful planning and design process, it is essential that specific strategies be explored and evaluated in order to determine how and to what extent these strategies encourage physical activity among children. This study provides evidence that schoolyard renovation from Learning Landscapes increases active utilization by school children during both mandatory and optional play periods, contributing to a nascent body of work [28, 44] that suggests that such renovation may be an effective method to encouraging physical activity among children.