Background and Purpose
Schools are important environments for children to accrue health promoting physical activity. In elementary schools, physical education (PE), recess, and other before/during/after school programs all have been identified as being important contributors to physical activity accrual. Of these, PE is the only physical activity program institutionalized in U.S. schools. It is a mandated curricular area in nearly all 50 states, and is often referred to as the foundation of school physical activity. Although it is the foundation, the broader implications of PE relative to other school physical activity programs remains unclear. There have been a few recent studies of school PE policies, but an understanding of the broad based function of these policies as they relate to the policies and practices of other school physical activity programs  (e.g., recess) is limited. The purpose of our study was to explore these relationships.

Objectives
The specific objectives of our study were to assess the relationships of various district- and school-level elementary school PE policies to school-level recess policies and practices.

Methods
We analyzed existing School Physical Activity Policy Assessment (S-PAPA) data completed by a key informant in 65 elementary schools in 7 states. S-PAPA is a validated instrument that provides information on the adoption and implementation of district- and school-level policy areas and related school environmental variables at the school site. It has three distinct sections: 1). PE; 2). Recess; and 3). Other Before, During, and After School Programs. This study focused specifically on the S-PAPA data for district- and school-level PE policies and recess policies and practice. We used descriptive statistics (percentages) to describe PE and recess policy adoption and recess practice at the school level. We used logistic regression to assess the odds of a school having individual recess policies or practices when specific district- or school-level PE policies were adopted.

Results
The most prevalent PE policies pertained to the assignment of grades, standards/guidelines for PE content, and PE minutes per week; these were more prevalent at the district level than at the school level. Recess policies were less common than PE policies, and the most prevalent of recess policies related to playground facility maintenance, recess minutes per day, and the training of playground supervisors. Results showed that 68% schools were in districts that had a policy requiring a specific number of minutes or days of PE per week. Compared to those without the district policy, schools located in districts with a minutes or days of PE per week policy had nearly 5 times greater odds of having a district policy specifying the number of recess minutes per day students should receive (OR = 4.8, 95% CI = 1.32 – 17.44), 5 times greater odds of having a district policy requiring that recess supervisors receive supervision training (OR = 5.17, 95% CI = 1.04 – 25.85), and over 10 times greater odds of providing students with scheduled recess each day (OR = 10.67, 95% CI = 1.1 – 103.27). Fifty-six percent of schools reported this same PE policy at the school level (requiring a specific number of minutes or days of PE per week). Compared to schools without a policy requiring a specific number of minutes or days of PE per week, schools with that policy had over 6 times the odds of having a policy specifying the number of recess minutes per day for students (OR = 6.30, 95% CI = 1.94 – 20.38).

Conclusions
This is one of the first studies to examine the impact of PE policies on physical activity opportunities outside of PE, and our results showed that adoption of PE policy by a school has important implications for recess policies and practices. In particular, a PE policy requiring a specific number of minutes or days of PE per week at both the district and schools levels appears to be extremely important relative to schools adopting recess policies and practices.

Support / Funding Source
Data were from a project funded by grant #67113 from Active Living Research, a national program of the Robert Wood Johnson Foundation.