Background:
Organized school sport is a viable medium for promoting physical activity among youth. Participation in youth sport however declines significantly among middle school children. Research suggests that a school sport policy incorporating intramural sports (relative to varsity sports) may introduce more children to a wider variety of sports and perhaps foster increased physical activity during youth and over the lifespan. School sport policies may have particular relevance for girls. Physical activity of middle school girls is influenced to a greater extent by school social climate which in turn is influenced by school policies. Several studies show that girls are less likely to be physically active than boys. Limited empirical research however has addressed the effects of school sport policies on the physical activity of adolescent boys and girls.

Objectives:
This study addressed two objectives: (1) compare physical activity levels of middle school children engaged in two different types of sport programs (Intramural [IM] or Varsity sports [VS]); and (2) examine associations between physical activity and gender, level of program supervision, and other selected program characteristics by type of sport program (IM or VS).

Methods:
Data were obtained from direct observations in 4 schools (2 IM and 2 VS). Between January 2009 and December 2009 1,188 observations (661 VS scans; 527 IM scans) were conducted by trained assessors using the System for Observing Play and Leisure in Youth (SOPLAY). Physical activity levels were recorded in predetermined zones designated for school sports between 2:30-5:00PM during the different school sports seasons. Primary SOPLAY codes accounted for student gender and activity level (sedentary, walking, and vigorous), type of play and level of adult supervision (no, limited, or full supervision). Inter-rater reliability for SOPLAY codes was acceptable (kappa > 0.70). Multinomial logistic regression was used to examine associations between physical activity and the predictor variables. To examine for differences of school policies, separate models were estimated for IM and VM programs.

Results:
Overall, 6,821 children (52% boys) were observed during the study. Across all children, 53 % were sedentary when observed, 29.5% were walking when observed, and 17.5% were engaged in vigorous activity when observed. Across all of the predetermined zones physical activity areas were vacant during 68% of visits to them, with areas in IM schools unoccupied more often (78% vs. 59% of visits). However, students in activity areas in IM schools were more physically active and more likely to be engaged in MVPA than those in VS (51.1 vs. 45.8%). Full supervision was more apparent in varsity schools (88.2%) than in intramural schools (58.5%) although limited supervision was greater among intramural schools (34.8% IM vs 8% VS). An examination of gender based on school type revealed several patterns. Boys participating in intramural sports were more likely to be engaged in vigorous physical activity than girls (OR =1.87, p<.0001). However, no differences in physical activity levels by gender were observed in varsity schools. Boys were less likely to engage in moderate physical activity in varsity school sports (OR =.74, p<.0001). No differences between girls and boys engaged in moderate physical activity levels were observed in intramural schools. Full supervision among intramural sport participants was associated with higher odds of engaging in vigorous physical activity (OR= 2.06, p<.05). Conversely, supervision (limited or full) was associated with lower odds of vigorous physical activity among varsity sport participants (full supervision - OR = 0.49, p<.001; limited supervision OR = 0.34, p<.0001). No significant differences were associated with moderate physical activity in either school type.

Conclusions:
Although data was limited to 4 schools in one geographic region, this study was one of the first to objectively measure the impact of sport policies on adolescent’s physical activity levels. These data show that school sport settings are under-utilized and that school sport policies may impact opportunities for vigorous activity levels among children. Sixty-eight percent of designated sport areas were not being used when observed. The study also showed that intramural sport programs may not be providing as much physical activity opportunities for girls. Finally, supervision was associated with increased activity in intramural sport programs but was negatively associated with varsity sports. Policy measures suggested by our data include: (1) increasing joint programming/joint use of school facilities where there is “excess capacity”; (2) examine the type of adult supervision that occurs during sport to maximize physical activity levels; (3) adopting intramural programming that is more likely to facilitate activity among girls (e.g., sports exclusively for girls and more student involvement in the selection of intramural sports); (4) monitoring time spent in actual activity during sport program and establishing minimum standards; and (5) a 'full' implementation of an intramural program would likely increase physical activity.

Support:
This study was funded by the Robert Wood Johnson Foundation, Active Living Research.