Recognizing that environments influence physical activity levels, many community leaders, especially in low-income areas, are considering shared or joint use of school grounds and facilities to increase physical activity in the community and to improve a community’s social cohesion (1–5). A joint-use agreement employs a contract or written agreement between a school and a city or community entity outlining specific terms and conditions for shared use (6). To date, few studies have examined the use of these agreements to create physical activity opportunities in under-resourced communities (1,4,5,7).

Although many public health experts consider joint-use agreements a viable strategy for increasing physical activity, some schools remain reluctant to enter into such agreements because of concerns about liability and additional maintenance costs (6). To address these concerns, groups specializing in public health law have begun to create resources to support communities that are considering adopting joint-use agreements (6–8).

At the local level, joint-use agreements are an attractive approach for increasing physical activity among residents, especially for communities with limited parks or open spaces and high rates of obesity (9). A recent report by the Los Angeles County Department of Public Health (DPH) showed that cities in the county with the least available amount of park space experienced more economic hardship and had higher obesity rates than cities in the county with the most available amount of park space (10).

With the support of the Communities Putting Prevention to Work obesity prevention program of the Centers for Disease Control and Prevention (http://www.cdc.gov/communitiesputtingpreventiontowork/), DPH established the Joint Use Moving People to Play (JUMPP) Task Force to address the lack of physical activity opportunities by making system and environmental changes, such as joint-use agreements with schools. Our study examines the JUMPP Task Force’s work from January 2010 to December 2012. Our study’s objective was to assess whether joint-use agreements can increase community members’ use of school grounds outside of school hours.

We used mixed methodology to evaluate community member use of school grounds following execution of JUMPP-assisted joint-use agreements in Los Angeles County from January 2010 to December 2012. Methods included cross-sectional observations of community members’ use of school facilities and environmental assessments of facilities and exercise equipment at each site. Although our evaluation team envisioned pre- and post-test design data collection (ie, onsite observations of physical activity before and after implementation of joint-use agreements), school grounds or facilities were not opened to the community outside of school hours until joint-use agreements were executed, making pre-implementation observations unnecessary. Agreements were executed throughout Los Angeles County; however, these agreements were concentrated in areas with the least amount of park space per 1,000 residents (Figure).

The combined total amount of space available for physical activity at all 12 observed sites was 43,101 square meters; available space for each site ranged from 155 square meters to 10,787 square meters. Eleven of the 12 sites were deemed conducive to physical activity per SOPARC criteria and had equipment that was in working condition. The 1 school site deemed not conducive to physical activity had an area with excessive mud and some standing water during 1 of the days of observation. Of all spaces observed, 83% were outdoor areas and 17% were indoors; spaces were all open and accessible during the time of observation. On most days of site visits, the weather was mild; observers did not encounter rainy or excessively hot conditions.

In total, the evaluation team completed 172 SOPARC observations and environmental assessments for the 12 school sites with joint-use agreements. Observers recorded information for a total of 1,669 community members; nearly half (42.0%) were adults and almost all (95.0%) were nonwhite (Table 1). The age and race characteristics of site users (Table 2) were similar to those of the population in the surrounding communities (Table 3).

Ten of the school sites had physical activity programs underway some or all of the time that their facilities were open to shared use; 2 sites did not offer any programs. Observers conducted nearly half (47.0%) of the 172 observations and environmental assessments while physical activity programs was underway at the site. The spaces across all sites were empty one-third (35.0%) of the time, but only at times when no programs were underway. The 2 sites that did not offer physical activity programs were empty more frequently than sites that offered programs. Community use was 16 times higher at joint-use sites when observations were conducted during physical activity programs than when no structured programs were underway (P < .001). In observations made during programs, an average of 8.4 observed adults and 11 children used the school grounds or facilities; in observations when there was no program underway, 0.3 adults and 0.9 children used the school grounds or facilities.

About two-thirds (68.0%) of all community members using school facilities were observed participating in moderate (51.0%) or vigorous (17.0%) physical activity. The remaining community members were inactive at the time of observation (32.0%). The evaluation team did not find differences in observed physical activity levels (ie, moderate vs vigorous) between users at sites with physical activity programs and users at sites without programs.

Our study showed that adopting joint-use agreements is feasible for schools and can successfully attract community members to use school facilities for physical activity outside of school hours. The study also showed that conducting physical activity programs for adults and children at schools increases community use of school facilities. Because most JUMPP-assisted agreements were adopted and implemented in cities or communities with disproportionately high obesity rates, this strategy could be a model of practice for similar communities. It would align with the emerging viewpoint among community leaders that joint-use agreements are worthwhile investments that can increase physical activity opportunities in under-resourced areas of a city or community at a relatively low cost (6,8).

Physical activity programming has been found in several studies to be a major factor in promoting moderate to vigorous physical activity and community use of parks and open spaces (17,18). In this study, school sites with physical activity programs attracted 16-fold more community members than school sites that had no structured programs. With guidance from the JUMPP task force and support from partnering organizations (eg, city and county parks and recreation departments, the American Youth Soccer Organization), physical activity programs can become an integral part of interventions, helping to sustain and accentuate the health and social benefits of joint-use agreements over time.

Another important finding of our study is the value of adult-inclusive programs for attracting both adults and children to school sites during joint-use hours. This finding alludes to a family unit effect, suggesting that adult users will bring their children along for the structured activity even if specific child-centered activities are not offered. Interestingly, both adults and children accessing school recreational facilities during joint-use hours engaged primarily in moderate to vigorous physical activity regardless of the presence or type of program. Among these users, Hispanics remained the largest group seen engaging in moderate to vigorous physical activity. This observation has important implications for reducing health disparities given the disproportionately high rates of obesity and weight-related chronic conditions in Hispanic populations (19,20).

Our study has several limitations. First, the observational nature of the evaluation design limited collection of individual-level data from community members, including such measures as self-reported sociodemographics, customary recreational habits, and intentions to change physical activity behaviors. This limitation, however, should not be viewed as a weakness of the observational data, because these observational data are generally regarded as more reliable than self-reported behaviors, avoiding, in many instances, social desirability and selection biases (21). Second, because of limited resources and time constraints, only 1 trained observer was dispatched to each site for the SOPARC observations. For this reason, inter-rater reliability could not be meaningfully assessed. Although this may represent a notable unknown about the adapted SOPARC tool’s internal validity, the concern is likely minimal because the tool has been previously validated by its designers (14), and the adaptations that we made did not deviate significantly from the original instrument. Third, the cross-sectional data collection approach offered no temporal information about changes in physical activity levels among community members. For example, there is no way to determine whether community members observed using the school facilities actually increased their net amount of physical activity as a result of the joint-use agreements. Users may possibly have substituted physical activity space provided at each school site for another physical activity space, yielding no net increase in overall minutes of physical activity. Fourth, because of the mixing of adult- and youth-centered programming at several of the school sites, observers in several instances could not determine with a strong degree of certainty whether the children observed participating in moderate to vigorous physical activity were part of a family unit (brought along by parents) or there for the youth-oriented activities. Finally, we observed community members using the school facilities at only 1 point in time. Therefore, whether these community members will continue to use the opened spaces for the longer term is difficult to predict.

In spite of analytic limitations, findings from the present study suggest that implementing school-based joint-use agreements is a promising strategy for increasing open space for moderate to vigorous physical activity among adults and children in under-resourced communities. Although Los Angeles County school districts have a history of creating such agreements for promoting physical activity among youth, they have not opened school facilities for use by adults. School officials have historically viewed children as the intended beneficiaries of school space. A new paradigm in which adults are welcomed to access school facilities for exercise is one step closer to the much-promoted model of schools as centers of the community (22,23). Agreements that stipulate physical activity programs for adults and youth, for example, may draw a much greater number of users from the community than agreements that do not.

Although our study highlights the potential of joint-use agreements for promoting physical activity in under-resourced communities, numerous questions remain. For example, what are the administrative and maintenance costs of these agreements for schools? How can local jurisdictions use joint-use agreements more effectively to engage community members in physical activity over longer periods of time, without significant attrition? What role does the family play in influencing community member participation at joint-use school sites? The latter question speaks to the interplay between adults and children as a family unit, which may offer social reinforcement for sustaining moderate to vigorous physical activity and can lead to multiplicative health benefits in the long term (17,18,22,23). Answers to these questions could help inform public health efforts to promote physical activity in disadvantaged communities.