Physical Activity During Recess Outdoors and Indoors Among Urban Public School Students, St. Louis, Missouri, 2010–2011

We measured the quantity and intensity of physical activity in 106 urban public school students during recess outdoors, recess indoors in the gym, and recess indoors in the classroom. Students in grades 2 through 5 wore accelerometer pedometers for an average of 6.2 (standard deviation [SD], 1.4) recess periods over 8 weeks; a subsample of 26 also wore heart rate monitors. We determined, on the basis of 655 recess observations, that outdoor recess enabled more total steps per recess period (P < .0001), more steps in moderate-to-vigorous physical activity (P < .0001), and higher heart rates than recess in the gym or classroom. To maximize physical activity quantity and intensity, school policies should promote outdoor recess.

(Health-o-Meter, Bridgeview, Illinois), height with a stadiometer (Seca, Birmingham, United Kingdom), and waist circumference at the superior border of the iliac crest. HealthWatch Pro 3.1 software (Seattle, Washington) was used to compute sex-and age-specific percentiles for body mass index (BMI) (5) and waist circumference, and sex-, age-, and height-specific percentiles for blood pressure (6). This study was approved by the Washington University School of Medicine Human Research Protection Office and the school district's Research Review Committee.
Generalized estimating equations with an exchangeable correlation matrix were used to evaluate the effects of recess location and sex on total steps and steps in rated MVPA (SAS version 9.3 [SAS Institute, Inc, Cary, North Carolina]). Analysis of variance was used to compare heart rate between recess locations.
On the basis of 655 recess observations of 106 students (6.2 recess periods per student; SD, 1.4), we determined that the greatest quantity and intensity of physical activity were achieved during outdoor recess. There was a significant interaction between recess location and sex on steps per recess period (P = .007). Girls took an average of 976 steps per recess period outdoors (SD, 385), 692 steps in the gym (SD, 330), and 328 steps in the classroom (SD, 465) (P < .0001 for all comparisons) ( Figure 1). Boys took 1,281 steps per recess period outdoors (SD, 566), 824 steps in the gym (SD 459), and 378 steps in the classroom (SD, 433, P < .001 for all comparisons). Boys accumulated more steps than girls during recess outdoors (P < .001). More steps during outdoor recess were rated MVPA (56%) than steps during recess in the gym (36%) or classroom (15%) (P <.001 across recess locations).
We acquired individual and mean heart rate data for 111 recess observations

Discussion
Our study revealed that location is a major determinant of the quantity and intensity of physical activity achieved during recess. Students accumulated more total steps and more steps rated MVPA during recess outdoors than during recess in a gym or classroom. Consistent with findings in other studies (7), boys were more active than girls.
High levels of MVPA by children are associated with reduced cardio-metabolic risk factors (8). The Institute of Medicine recommends that schools provide opportunities for students in grades kindergarten through 12 to participate in 60 minutes of physical activity each school day (9). A national sample of children aged 6 to 11 years found that only 42% achieved the recommended amount of daily physical activity (10). Children who live in low-income neighborhoods may have limited opportunities outside school to engage in physical activity. Reznik et al found that students are more active on school days with a physical education class than on those without and on days with outdoor recess than on those without (11). With physical education curricula limited in many public school districts, recess may account for a large portion of school-based physical activity.
Recess provides students with opportunities for important physical, cognitive, social, and emotional benefits (12). Although our results highlight the importance of outdoor recess to optimize physical activity, indoor recess is often necessary and therefore should include creative activities that promote movement of moderate-to-vigorous intensity, whether in a gym or a classroom (13).
Study limitations are inclusion of only 2 schools and the possibility that study participants were more or less active than nonparticipating students, which would limit the generalizability of our results. Furthermore, accelerometer pedometers provided only estimates of students' MVPA.
In summary, outdoor recess provided the optimal environment for children to engage in quality physical activity. When outdoor recess is not feasible, district and school policies should mandate indoor recess opportunities that promote active play, particularly in urban settings.