Background:
The effect of playground equipment without organized activity on the physical activity of children is not well understood. After Hurricane Katrina, play structures were built on a number of schoolyards in New Orleans. We assessed physical activity levels of children in three schoolyards before and after the play structures were installed.
Objectives:
Our objective was to determine whether installed play equipment has any effect on physical activity among children during free play.
Methods:
Observations
We conducted direct observations of children playing in the yards of three elementary schools, using the SOPLAY method, in which children are classified at a single point in time as sedentary, walking, or very active. Each schoolyard was divided into 5-6 “target areas” for observations, counting boys and girls separately. Observations took place during the spring of 2007 five to eight days immediately before the schoolyards were closed to prepare them for the installation and eight to 11 days immediately after the installation.
Analysis
For each school and each time period, the number of children at each level of activity was summed across target areas, and the percentage of all children at each activity level was then calculated.
We used logistic regression to test whether the change in proportion of children who were very active increased between pre- and post-installation periods. We analyzed each school separately because the schoolyards were very different both before and after the installations.
Results:
Before and after the installation, the School A schoolyard had three slides and four climbers, so the installation was an improvement in quality rather than a change in the equipment available. At School B, there were three slides and one climber before the installation, and four slides, four climbers and four swings afterwards. School C had no installed play equipment before the installation, which included three slides, two climbers, and four swings.
There was no change in physical activity of children at School A, with the percent of children “very active” 31% before the installation and 32% afterward.
In contrast, there were large changes in physical activity at School B, with the percent of children “very active” only 14% before the installation but 30% afterward (p<0.001)
There were small and inconsistent changes in physical activity at School C. While the percent of children “very active” increased from a low 13% before the installation to 23% after the installation, the percent of children “sedentary” also increased from 50% before the installation to 55% afterwards. The target area which the playground equipment was installed was used by very few children before the installation, but was by far the most heavily used after the installation. Within that target area, 30% of children were “very active” and only 47% were sedentary.
Conclusions:
The effect of installed play equipment on physical activity levels was inconsistent across the three schools. The increase in physical activity at School B, in which there was a substantial increase in the amount of play equipment, may have been caused by the installed play equipment. The lack of effect at School A may be explained by the fact that the amount of play equipment did not change much.
The lack of an increase in physical activity at School C is surprising. This schoolyard had no play equipment before the installation and a large installation afterward, so we expected the largest change in physical activity at this school. The children at School C were not very active either before or after the installation. Observers recorded the predominant activities of children in each target area, and there were notes that children at School C were often waiting in line to use the swings, so it is possible that there simply was not enough equipment for the children even after the installation. Another possibility was that teachers at School C were imposing order on children’s play in ways that reduced activity. Still, it remains possible that the play equipment at School C simply did not induce children to become more physically active.
Because the results in these three schools are inconsistent, it would be valuable to conduct similar before-and-after observations at other schools to see if a pattern emerges. We are conducting additional observations in the fall of 2007 and spring of 2008. For these additional observations, we will be using both SOPLAY and accelerometers to measure children’s physical activity.
9/14/2007
Support:
Centers for Disease Control and Prevention, Prevention Research Centers Program, through the Tulane University Prevention Research Center, cooperative agreement # 1-U48-DP-000047.
