School Characteristics and Physical Activity among Grade 1-4 Students Scott T. Leatherdale, PhD Objectives: To examine school and student characteristics associated with physical activity in grade 1 to 4 students. Methods: Multi-level logistic regression examined factors associated with being moderately and highly active. Results: A student was more likely to be moderately or highly active if he/she attended a school that provides student access to facilities and equipment outside of school hours. Male students were more likely to be moderately or highly active
R
egular participation in physical activity (PA) is critical for healthy child development. PA is associated with a decreased risk for chronic illness and obesity, and improved physical, mental, and social well-being among children.1,2 Unfortunately, PA levels are declining among children3 as time spent in sedentary behaviors is becoming more common.4-6 Considering that health behavior patterns of children may become resistant to change around the 6th grade,7 understanding the factors that impact PA among young children should be a priority. Existing research suggests that student-level characteristics (eg, sex, body mass index (BMI), sedentary behavior, and activity preferences) are associated with PA levels in older children.5,8-14 However, because school-based PA can account for up to 40% of the total activity among youth populations in Canada,15 it suggests that there is a need to move beyond only examining individual student characteristics associated with youth PA if we want to understand how best to target school-based PA interventions so that they are most likely to have impact.16,17 It may be just as important to identify the school-level correlates of children’s PA. For instance, whereas there is evidence demonstrating that characteristics of the school PA policy environment or the built environment surrounding a school are important predictors of PA for students
Scott T. Leatherdale, University of Waterloo, School of Public Health and Health Systems, Waterloo, ON, Canada. Correspondence Dr Leatherdale; [email protected]
200
and overweight or obese students were less likely to be moderately or highly active. Conclusions: Research should evaluate if implementing a school policy that provides student access to facilities and equipment outside of school hours is effective in promoting physical activity in grade 1 to 4 students. Key words: physical activity, children, school policy Am J Health Behav. 2014;38(2):200-207 DOI: http://dx.doi.org/10.5993/AJHB.38.2.5
in higher grades (5 to 8) even when controlling for student characteristics,16-20 similar evidence does not yet exist for children in grades 1 to 4. Using a similar approach as research published with elementary school students in grades 5 to 8,16,17 this exploratory study is the first to examine how school-based PA policies and the number of recreation centers surrounding a school were associated with the PA levels of a sample of elementary school students in grades 1 to 4. METHODS Design This cross-sectional study used data collected in 2007-08 from a convenience sample of grade 1 to 4 students attending 30 elementary schools (Ontario, Canada) as part of the PLAY-On study.21 The purpose of the PLAY-On study was to examine the school-level factors associated with PA and overweight among a sample of elementary school students in grades 1 to 8. Given the different data collection protocols required for respondents of this age range, there were 2 separate study arms of the PLAY-On host study: the first arm of the study was among respondents in grades 5 to 8 (aged 10 to 13 years) and the second arm of the study was among respondents in grades 1 to 4 (aged 6 to 9 years). Research has previously examined the school-level factors associated with overweight5,22 and PA16,17 among the grade 5 to 8 respondents in PLAY-On. The following study examines the PA data collected from the sample of grade 1 to 4 respondents in PLAY-On. In this component of PLAY-On, studentlevel data were collected from eligible students
Leatherdale in grades 1 to 4 using 2 methods: data provided by parents, and data provided by students to the PLAY-On study staff. School-level built environment data were provided by the Enhanced Points of Interest (EPOI) data resource. The EPOI data file is a database of the type and location of different opportunity structures within the built environment (eg, recreation facilities grocery stores, fast-food restaurants). Additional details about the DMTI-EPOI resources are available online (http://www.dmtispatial.com). The school-level PA policy data were provided by the elementary school version of the Physical Activity Module (PAM) of the Healthy School Planner (HSP). The HSP-PAM is a tool designed to assess policies, activities, committees, facilities and guidelines surrounding PA in the school environment. Additional details about the HSP-PAM measures and assessment categories are available online (http:// www.healthyschoolplanner.uwaterloo.ca).
Measures Outcome – physical activity level. Consistent with evidence that parental proxy report of young children’s PA levels is a valid and reliable measure of youth PA,24,25 parents were asked to report: “On
average, how many hours per day is your child physically active? Please include both moderate activity (eg, walking, biking to school, recreational swimming) and vigorous activity (eg, jogging, active team sports, fast dancing).” Based on Canadian PA guidelines for children,26 PA was coded as follows: low active (3 hours per day). Student-level predictor variables. Body Mass Index (BMI) was calculated for each student using the objective measures of weight (kg) and height (m) (BMI=kg/m2). Weight status was then determined using the BMI classification system of the International Obesity Task Force27 based on age and sex adjusted BMI cut-points for individuals at 18 years of age. Parents were asked to report: “On an average day, how much time does your child spend watching TV/movies, playing video/ computer games, surfing the internet, instant messaging or talking on the phone?” Based on Canadian sedentary behavior guidelines for children,28 sedentary behavior was coded as follows: low sedentary (≤2 hours), moderately sedentary (2 to 4 hours), and very sedentary (>4 hours). Students were asked 6 questions by the nurse collecting the Play-On data. Compared to other kids in your class, when you are at school (during recess, lunch, nutrition breaks) are you…(more active, less active, about the same)? Do you prefer to play alone or do you prefer to play with other children (alone, other children, both)? Do you prefer active games (eg, tag, kickball) or do you prefer quiet games (eg, board games) (active games, quiet games, both)? Do you like playing sports (eg, soccer, basketball) or do you dislike playing sports (likes sports, dislikes sports)? Do you like to read or do you dislike reading (likes reading, dislikes reading)? Do you like to play outside or do you like to play inside (outside, inside, both)? School-level predictor variables. The 2008 EPOI data were used to identify recreational facilities (includes dance studios, fitness/gym facilities, and sport and recreation clubs) within a 1-km buffer surrounding each of the 30 schools using the same method as Pouliou and Elliott.29 As described and presented elsewhere,16 the HSP tool measured PA policies in the 30 schools. In brief, the HSP PA tool measures school PA policies aligned with the 4 components that form the basis of the Foundations for a Healthy School (FHS): Healthy Physical Environment (availability of, access to, and adequacy in meeting student needs for, indoor and outdoor facilities, equipment and resources for safe, quality PA on or near school grounds, both during and outside of school hours); Instruction and Programs (availability, delivery and characteristics of curricular physical education, extracurricular PA programs, and active transportation to school, including barriers to implementing such programs); Supportive Social Environment (characteristics of the school’s
Am J Health Behav.™ 2014;38(2):200-207
DOI:
Data Collection All students in grades 1 to 4 at the participating schools were eligible to participate. Prior to participating, active consent from parents (or guardians) was required and at any time students were able to decline participation. During the active consent process, parents were asked to report their child’s PA. During the data collection from students, the registered nurse working for PLAY-On asked each eligible student 6 questions pertaining to their physical activity routines and preferences. After their responses were recorded by the nurse, students were then asked to remove their footwear, and height and weight were objectively measured and recorded by the nurse. Students were asked to stand upright with their back against the wall, and then the nurse used a square placed on the top of their head to measure their height to the nearest centimeter against a standard metric tape measure fastened to the wall. Weight was measured to the nearest one decimal place in kilograms using a Bathscapes® LED digital bath scale (model 1053WWBA). At each school, the administrator(s) most knowledgeable about the school’s PA policies completed the HSP survey. Participants Of the 3926 students enrolled in grades 1 to 4 at the 30 participating elementary schools, 59.4% (N = 2331) completed the survey. Missing respondents resulted from parent refusal and absenteeism on the day of the survey. This distribution is consistent with an active consent study examining overweight among Canadian elementary students.23
http://dx.doi.org/10.5993/AJHB.38.2.5
201
School Characteristics and Physical Activity among Grade 1-4 Students
Table 1 Descriptive Statistics for Youth in Grades 1 to 4 by Sex in PLAY-On (Ontario, Canada, 2007-2008)
Student-level characteristics
Male (N = 1187)
Female (N = 1139)
%
%
(n)
a
(n)
Chi-square
a
Physical Activity Level b
Low active Moderately active Very active
10.0 (115) 63.3 (726) 26.7 (306)
14.3 (158) 62.7 (695) 23.0 (255)
χ2=11.41, df=2, p < .01
Grade
1 2 3 4
22.7 (270) 21.5 (255) 27.5 (326) 28.3 (336)
22.4 (256) 21.1 (241) 27.1 (309) 29.4 (335)
χ2=0.32, df=3, p = .95
Normal weight Overweight Obese Missing
75.0 (890) 13.7 (162) 6.1 (144) 5.2 (62)
74.5 (848) 14.0 (159) 6.2 (71) 5.4 (61)
χ2=0.09, df=3 , p = .99
Sedentary behavior b
≤2 hour per day 2 to 4 hours per day >4 hours per day
70.6 (810) 21.8 (250) 7.6 (87)
75.2 (834) 18.9 (210) 5.9 (65)
χ2=6.37, df=2, p < .05
Physical activity level compared to other kids e
More active About the same Less active
7.4 (86) 28.5 (333) 64.2 (750)
9.3 (105) 18.2 (206) 72.5 (822)
χ2=34.56, df=2, p < .001
Prefers to play alone or with other children e
Alone Both Other children
2.4 (28) 57.1 (667) 40.5 (474)
1.0 (11) 54.8 (621) 44.2 (501)
χ2=9.24, df=2, p < .01
Prefers to play active games or quiet games e
Active Both Quiet
4.5 (52) 41.4 (484) 54.1 (633)
7.6 (86) 35.1 (398) 57.3 (649)
χ2=16.40, df=2, p < .001
Preference for playing sports e
Likes sports Dislikes sports
96.5 (1,127) 3.5 (41)
92.7 (1,049) 7.3 (83)
χ2=16.46, df=1, p < .001
Preference for reading e
Likes reading Dislikes reading
79.6 (931) 20.4 (238)
89.5 (1,014) 10.5 (119)
χ2=42.66, df=1, p < .001
Prefers to play outside or inside e
Outside Both Inside
4.1 (48) 20.6 (241) 75.3 (880)
4.2 (48) 17.2 (195) 78.6 (890)
χ2=4.35, df=2, p = .11
c
Weight status
c,d
Note. a Numbers may not add to total because of missing values b Reported by parent or guardian c Measured by PLAY-On nurse d Body mass index (BMI) values used to determine weight status have been adjusted for age and sex e Reported by student and recorded by PLAY-On nurse
social environment that predispose, reinforce and enable enjoyable, lifelong participation in PA or that hinder such activities); and Community Partnerships (the accessibility and availability of support services for PA which may include partnerships with public health units and community based services and resources). Each indicator was assigned a classification based on the corresponding phase of implementation as outlined by the HSP developers: Initiation (falls short or exhibits extensive room for improvement in meet-
202
ing the recommendations related to school capacity for PA); Action (meets the recommendations in several, but not all areas related to school capacity for PA, exhibits some room for improvement); Maintenance (consistently meets or exceeds the recommendations related to school capacity for PA, encouraged to maintain the current level of commitment to supporting PA at school). Each of the 4 FHS components was also assigned an ‘overall’ phase classification based on the combined responses to component indicators.
Leatherdale
Table 2 Results of the Univariate Multi-level Logistic Regression Analyses Examining School-level Characteristics Associated with being Moderately or Highly Active among Youth in Grades 1 to 4 in PLAY-On (Ontario, Canada, 2007-2008) Model Estimates (Standard Error) Phase
Model 1 Moderately Active vs. Low Active
Model 2 Highly Active vs. Low Active
Student access to a variety of facilities on and off school grounds during school hours †
Action Maintenance
0.35 (0.52) 0.28 (0.51)
0.07 (0.54) 0.18 (0.54)
Availability of physical activities during inclement weather †
Action Maintenance
-0.19 (0.25) 0.17 (0.86)
0.19 (0.25) -0.69 (0.98)
Student access to facilities and equipment outside of school hours †
Action Maintenance
-0.01 (0.24) 1.44 (0.62) *
0.08 (0.24) 1.79 (0.80) *
Support for active transportation to and from school †
Action Maintenance
-0.02 (0.30) -0.34 (0.33)
-0.05 (0.30) -0.54 (0.33)
Action
0.37 (0.81)
0.13 (0.27)
Implementation of daily PA ‡
Maintenance
0.29 (0.26)
0.27 (0.27)
Time spent per week engaged in PA during physical education classes †
Action Maintenance
-0.26 (0.78) -0.61 (0.81)
-0.50 (0.79) -0.87 (0.82)
Classes taught by a qualified physical education specialist †
Action
0.03 (0.40)
0.22 (0.41)
Availability and use of intramural/club activities †
Action Maintenance
0.24 (0.35) 0.13 (0.50)
0.44 (0.35) 0.26 (0.50)
Consistency of intramural programming across grade divisions and seasons †
Action Maintenance
0.21 (0.27) 0.03 (0.35)
0.22 (0.28) 0.16 (0.37)
Availability and use of interschool programs †
Action Maintenance
0.29 (0.25) 0.22 (0.52)
0.09 (0.27) -0.07 (0.54)
Maintenance
0.39 (0.85)
0.42 (0.86)
Action
0.29 (0.26)
0.27 (0.27)
Emphasis placed on maximizing participation in PA through school programs †
Action Maintenance
0.69 (0.39) 0.75 (0.31) *
0.92 (0.41) * 0.84 (0.36) *
Incorporation of PA into other school subjects †
Action Maintenance
0.15 (0.30) 0.14 (0.41)
0.28 (0.31) 0.09 (0.42)
Special recognition of students who participate in school physical activities †
Action Maintenance
0.11 (0.46) 0.04 (0.40)
-0.05 (0.47) -0.14 (0.42)
Formal collection of suggestions from the school community about PA at school †
Action Maintenance
0.28 (0.28) 0.15 (0.37)
0.05 (030) 0.05 (0.40)
Promotion of PA programs and events for students, families and school staff †
Action Maintenance
0.59 (0.25) * -0.20 (0.25)
0.63 (0.27) * -0.10 (0.27)
Use of PA as a reward, not as discipline †
Action Maintenance
-0.25 (0.27) -0.04 (0.34)
-0.14 (0.28) -0.22 (0.36)
Presence of written policies or practices that support PA †
Action Maintenance
0.05 (0.31) -0.32 (0.35)
0.14 (0.33) -0.20 (0.36)
Action
0.28 (0.80)
0.60 (0.82)
Support available for school staff involved with PA ‡
Maintenance
-0.24 (0.25)
-0.11 (0.27)
Connection to community resources †
Action Maintenance
-0.36 (0.38) -0.44 (0.30)
-0.34 (0.40) -0.35 (0.32)
Action Maintenance
-0.09 (0.36) -0.33 (0.33)
-0.14 (0.38) -0.19 (0.35)
-0.30 (0.09) *
-0.29 (0.10) *
Healthy Physical Environment
Overall score for this indicator † Instruction and Programs
Consistency of interschool programming across seasons † Overall score for this indicator † Supportive Social Environment
Overall score for this indicator † Community Partnerships
Overall score for this indicator † Number of opportunity structures within a 1 kilometer (km) radius of a school Recreation facilities
* p < .05 Note. Model 1: 1 = Moderately Active (N = 1,406), 0 = Low Active (N = 268) Model 2: 1 = Highly Active (N = 553), 0 = Low Active (N = 268) † Reference group is Initiation, ‡ Reference group is Action
Am J Health Behav.™ 2014;38(2):200-207
DOI:
http://dx.doi.org/10.5993/AJHB.38.2.5
203
School Characteristics and Physical Activity among Grade 1-4 Students Analyses Using student-level data, the prevalence of PA, parent measures and student measures were examined by sex. Consistent with previous research,6,9,16,17 2 series of multi-level logistic regression analyses examined characteristics associated with being (1) moderately active versus low active, and (2) highly active versus low active; each analysis used a stepped modelling procedure. Step 1 examined if the differences in being moderately active or highly active were random or fixed across schools. The school-level variance term from Step 1 (σ2μ0) was used to calculate the intraclass correlation (ICC). In Step 2, a series of univariate analyses were used as exploratory analyses to examine if the implementation phase score for each of the FHS indicators were associated with being moderately active or highly active. Only significant school-level variables (p < .05) were retained for further analyses. In Step 3, multivariate models were developed to examine how the student characteristics and the significant school characteristics identified in Step 2 were associated with being moderately active (Model 1) and highly active (Model 2). Since most students (94.6%) reported that they like playing sports, it was excluded from the multivariate predictive models. The multi-level statistical analyses were conducted on MLwiN Version 2.0230 and the descriptive statistics were conducted on SAS 9.2.31 RESULTS Demographic characteristics of the students are presented in Table 1. In brief, the sample was 51.0% (N = 1187) male and 49.0% (N = 1139) female. A total of 273 (12.1%) students were considered low active, 1421 (63.0%) were considered moderately active, and 561 (24.9%) were considered very active by their parents. Males were more likely to be considered very active compared to females (c2=11.41, df=2, p < .001). The mean number of recreation facilities within a 1-km buffer of the schools was 0.8 (range, 0 to 4). As presented previously,6,16 there was substantial variability across schools in relation to the PA policy indicators measured. Factors Associated with being Moderately Active Among students in grades 1 to 4, there was significant between-school random variation in being moderately active [σ2μ0=0.258(0.105), p < .05]; school-level differences accounted for 7.3% of the variability in the odds of a student being moderately active versus low active. The univariate analyses (Table 2) revealed that the number of recreation centers and 3 school-level policy characteristics (student access to facilities and equipment outside of school hour, emphasis placed on maximizing participation in PA through school programs, promotion of PA programs and events for students, families and school staff) were significant, and thus, in-
204
cluded in the final model. In the final multivariate model (Table 3: Model 1), only one school characteristic was significantly associated with being moderately active when controlling for student characteristics. If a student attended a school that was in the maintenance phase for the indicator student access to facilities and equipment outside of school hours, he/she was over twice as likely to be moderately active than a similar student attending a school that was in the initiation phase for this indicator (OR=2.82). This model also revealed that students considered overweight (OR=0.64) or obese (OR=0.59) were less likely to be moderately active. Conversely, students spending 2 to 4 hours per day in sedentary behaviors (OR=1.56) or students who reported they prefer to play active games (OR=2.18) were more likely to be moderately active. Factors Associated with being Highly Active Among students in grades 1 to 4, there was significant between-school random variation in being highly active [σ2μ0=0.257(0.113), p < .05]; schoollevel differences accounted for 7.3% of the variability in the odds of a student being highly active versus low active. The univariate analyses (Table 2) revealed that the number of recreation centres and 3 of the school-level policy characteristics (student access to facilities and equipment outside of school hours, emphasis placed on maximizing participation in PA through school programs, promotion of PA programs and events for students, families and school staff) were significant, and thus, included in the final model. In the final multivariate model (Table 3: Model 2), only one school characteristic was significantly associated with being highly active when controlling for student characteristics. If a student attended a school that was in the maintenance phase for the indicator student access to facilities and equipment outside of school hours, he/ she was over 5 times as likely to be highly active than a similar student attending a school that was in the initiation phase for this indicator (OR=5.24). This model also revealed that students considered overweight (OR=0.51) or obese (OR=0.37) were less likely to be highly active. Conversely, male students (OR=1.67) and students spending 2 to 4 hours per day in sedentary behaviors (OR=1.91) were more likely to be highly active. DISCUSSION Considering that few researchers have data linking school PA policies to the PA levels of youth in grades 1 to 4, this study provides a unique new contribution to the literature by being the first to examine how school characteristics (both the school PA policy environment and the built environment surrounding a school) were associated with PA levels of young children in grades 1 to 4. Specifically, we identified that even when controlling for individual student characteristics, the PA levels of grade 1 to 4 students can vary as a function of the characteristics of the school they attend.
Leatherdale
Table 3 Multi-level Logistic Regression Analyses of School- and Student-level Characteristics Associated with Physical Activity among Youth in Grades 1 to 4 in PLAY-On (Ontario, Canada, 2007-2008) Adjusted Odds Ratio a (95% CI) Model 1 Model 2 Moderately Active Highly Active vs. Low Active vs. Low Active Student-Level Characteristics Sex
Female Male
1.00 1.36 (0.77, 1.96)
1.00 1.67 (1.20, 2.32)**
Grade
1 2 3 4
1.00 1.07 (0.71, 1.62) 1.02 (0.69, 1.52) 1.07 (0.72, 1.59)
1.00 0.78 (0.49, 1.27) 0.75 (0.48, 1.19) 0.73 (0.46, 1.16)
Weight status b
Normal weight Overweight Obese Missing BMI
1.00 0.64 (0.43, 0.94)* 0.59 (0.36, 0.97)* 0.60 (0.33, 1.07)
1.00 0.51 (0.32, 0.81)** 0.37 (0.19, 0.70)** 0.54 (0.26, 1.10)
Sedentary behavior
≤2 hour per day 1.00 2 to 4 hours per day 1.56 (1.11, 2.20)** >4 hours per day 2.69 (0.79, 9.20)
1.00 1.91 (1.30, 2.81)** 2.84 (0.95, 8.47)
Physical activity level compared to other kids
About the same More active Less active
1.00 0.78 (0.46, 1.32) 1.23 (0.88, 1.71)
1.00 0.76 (0.40, 1.41) 1.11 (0.75, 1.65)
Prefers to play alone or with other children
Alone Other children Both
1.00 0.84 (0.27, 2.62) 0.98 (0.32, 3.03)
1.00 0.73 (0.19, 2.84) 0.78 (0.20, 3.05)
Prefers to play active games or quiet games
Quiet Active Both
1.00 2.18 (1.08, 4.40)* 1.21 (0.90, 1.63)
1.00 1.16 (0.49, 2.74) 1.24 (0.88, 1.75)
Preference for reading
Dislikes reading Likes reading
1.00 0.67 (0.43, 1.05)
1.00 0.66 (0.40, 1.08)
Prefers to play outside or inside
Inside Outside Both
1.00 0.73 (0.39, 1.38) 1.34 (0.92, 1.95)
1.00 0.66 (0.30, 1.47) 1.04 (0.67, 1.61)
Initiation Action Maintenance
1.00 1.23 (0.77, 1.96) 2.82 (1.14, 6.97)*
1.00 1.50 (0.89, 2.53) 5.24 (1.61, 17.06)**
Emphasis placed on maximizing participation in PA through school programs
Initiation Action Maintenance
1.00 1.50 (0.71, 3.18) 1.53 (0.78, 2.97)
1.00 1.96 (0.83, 4.66) 1.78 (0.82, 3.84)
Promotion of PA programs and events for students, families and school staff
Initiation Action Maintenance
1.00 1.45 (0.82, 2.56) 0.84 (0.49, 1.45)
1.00 1.39 (0.74, 2.62) 0.86 (0.47, 1.58)
1 unit increase
1.18 (0.95, 1.48)
1.17 (0.91, 1.51)
School-Level Characteristics Healthy Physical Environment
Student access to facilities and equipment outside of school hours Supportive Social Environment
Opportunity Structures in the Built Environment Number of recreation centres within 1-km radius * p < .05
**p < .01
Note. a Odds ratios adjusted for all other variables in the table and controlling for random variation across schools. b BMI values used to determine weight status have been adjusted for age and sex Model 1: 1 = Moderately Active (N = 1,406), 0 = Low Active (N = 268) Model 2: 1 = Highly Active (N = 553), 0 = Low Active (N = 268)
Am J Health Behav.™ 2014;38(2):200-207
DOI:
http://dx.doi.org/10.5993/AJHB.38.2.5
205
School Characteristics and Physical Activity among Grade 1-4 Students This is valuable insight for informing school-based PA prevention programming as it provides new evidence that the PA levels of young children are associated with their schools PA policy environment, and evidence justifying the need for school-based interventions early in elementary school because this study identified significant between-school variation which could be amenable to school-level intervention; if a school-level program or policy has a small effect on either shifting or normalizing the distribution of PA across schools, the impact at the student-level would be substantial.5 Consistent with PA research among students in grades 5 to 8,17 the results highlight that schools should work to ensure that they maintain or implement policies to provide students with access to facilities and equipment outside of school hours. However, there was no association between the number of recreation centers surrounding a school and PA among grade 1 to 4 students. This is consistent with evidence that suggests the built environment may not have an impact on younger students.32,33 Future research may not need to consider the built environment for this age group. Important student-level characteristics were also identified. First, boys were more likely than girls to be very active. This is consistent with previous research12,14 and may be related to differences in activity preferences among young boys and girls. Parents and educators should consider providing children with sex specific PA opportunities that incorporate the different types of student activity preferences identified here (eg, active games for boys). Second, consistent with evidence from older students,11,14,34 overweight and obese students were less likely to be moderately or highly active. Although we cannot determine the temporal nature of this relationship, this finding suggests that it may be wise to target interventions to promote PA to overweight children because they appear to be at increased risk for inactivity relative to normal weight youth. Such a targeted approached to PA promotion would require evaluation. Third, students were more apt to be active if they engaged in 2-4 hours per day of sedentary behavior per day. Although this appears counterintuitive, research with older youth has found that youth who spend substantial time engaging in sedentary behaviors can also spend substantial time engaging in PA.10 However, because sedentary behaviors are associated with youth obesity,9 strategies to reduce sedentary behavior among young children may still remain warranted. This study is subject to limitations. The crosssectional nature of the study design does not allow for inferences regarding causality. The self-report nature of the PA data is a limitation, but given that this is a secondary analysis of a host study designed to measure obesity in youth and evidence that parental proxy reports of young children’s PA levels are valid and reliable measure of youth PA,24,25 this was our only option within the Play-
206
On study. The generalizability of these findings is also unknown as limited school- and student-level activity data on these grades exists. Conclusions Characteristics of the school that a grade 1 to 4 student attends are associated with their physical activity levels. Future research should evaluate if implementing and promoting a school policy that provides student access to facilities and equipment outside of school hours is effective in promoting higher levels of PA among students in grades 1 to 4. At the student-level, it appears that activities promoting PA should be targeted to overweight or obese youth as they are the least likely to be moderately or highly active. Developing a better understanding of the school- and student-level characteristics associated with PA among children is critical for informing the targeting and tailoring of future intervention programs and policies. Human Subjects Statement The University of Waterloo Office of Research Ethics and appropriate school board ethics committees approved the study procedures. Conflict of Interest Statement The author declares that he have no conflict of interest or competing interests in this research. Acknowledgements Data used in this analysis were drawn from the PLAY-On project, funded by the Heart and Stroke Foundation of Ontario (grant awarded to S. Leatherdale). Dr Leatherdale is a Cancer Care Ontario Research Chair in Population Studies. REFERENCES
1. Public Health Agency of Canada. Physical Activity Benefits, 2011. Available at: http://www.phac-aspc.gc.ca. Accessed November 3, 2012. 2. Janssen I, LeBlanc AG. Systematic review of the health benefits of physical activity and fitness in school-aged children and youth. Int J Behav Nutr Phys Act. 2010;7:116. 3. Shields M. Overweight and obesity among children and youth. Health Rep. 2006;17:27-42. 4. Leatherdale ST, Ahmed R. Screen-based sedentary behaviours among a nationally representative sample of youth: are Canadian kids couch potatoes? Chronic Dis Inj Can. 2011;31:141-146. 5. Leatherdale ST. The association between overweight and school policies on physical activity: a multi-level analysis among elementary school youth in the PLAY-On study. Health Educ Res. 2010;5;1061-1073. 6. Leatherdale ST, Faulkner G, Arbour-Nicitopoulos K. School and student characteristics associated with screen time sedentary behavior among elementary school youth in the PLAY-On study: a multi-level analysis. Prev Chronic Dis. 2010;7:A128 7. Kelder SH, Chetyl L, Lytle L. Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. Am J Pub Health. 1994;84:1121-1126. 8. Kurc A, Leatherdale ST. The effect of social support and school- and community-based sports on youth physical activity. Can J Pub Health. 2009;100:60-64.
Leatherdale 9. Leatherdale ST, Papadakis S. A multi-level examination of the association between older social models in the school environment and overweight and obesity among younger students. J Youth Adolesc. 2011;40:361-372. 10. Wong SL, Leatherdale ST. Association between sedentary behaviour, physical activity, and obesity: inactivity among active kids. Prev Chronic Dis. 2009;6:A26. 11. Leatherdale ST, Wong SL. Modifiable characteristics associated with sedentary behaviours among youth. Int J Pediatr Obes. 2008;3:93-101. 12. Van der Horst K, Paw MJCA, Twisk JWR, van Mechelen W. A brief review on correlates of physical activity and sedentariness in youth. Med Sci Sports Exerc. 2007;39:1241-1250. 13. Young DR, Felton GM, Grieser M, et al. Policies and opportunities for physical activity in middle school environments. J Sch Health. 2007;77:41-47. 14. Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of physical activity of children and adolescents. Med Sci Sports Exerc. 2000;32:963-975. 15. Craig CL, Cameron C, Russell SJ, Beaulieu A. Increasing physical activity: Supporting children’s participation. Ottawa: Canadian Fitness and Lifestyle Research Institute; 2001. 16. Hobin EP, Leatherdale ST, Manske SR, Robertson-Wilson JE. A multilevel examination of school and student characteristics associated with moderate and high levels of physical activity among elementary school students (Ontario, Canada). Can J Pub Health. 2010;101:495-499. 17. Leatherdale ST, Manske S, Faulkner G, et al. A multilevel examination of school programs, policies and resources associated with physical activity among elementary school youth in the PLAY-ON study. Int J Behav Nutr Phys Act. 2010;7:6. 18. Naylor PJ, Macdonald HM, Warburton DER, et al. An active school model to promote physical activity in elementary schools: Action Schools! BC. Br J Sports Med. 2008;42:338-343. 19. Sallis JF, Conway TL, Prochaska JJ, et al. The association of school environments with youth physical activity. Am J Pub Health. 2001;91:618-620. 20. Zhu W. A multilevel analysis of school factors associated with health-related fitness. Res Q Exerc Sport. 1997;68:125-135. 21. The Play-On Study. Available at: http://www.shapes.uwater-
loo.ca/index.cfm?section=1002&page=1041. Accessed November 13, 2012. 22. Leatherdale ST, Pouliou T, Church D, Hobin E. The association between overweight and opportunity structures in the built environment: a multi-level analysis among elementary school youth in the PLAY-On study. Int J Public Health, 2011;56:237-246. 23. Veugelers PJ, Fitzgerald AL. Effectiveness of school programs in preventing childhood obesity: a multilevel comparison. Am J Pub Health. 2005;95:432-435. 24. Burdette HL, Whitaker RC, Daniels SR. Parental reports of outdoor playtime as a measure of physical activity in preschool-aged children. Arch Pediatr Adolesc Med. 2006;158:353-357. 25. Manios Y, Kafatos A, Markakis G. Physical activity in 6-year-old children: validation of two proxy reports. Pediatr Exerc Sci. 1998;10:176-188. 26. Canadian Society of Exercise Physiology. Canadian Physical Activity Guidelines for Children 5-11 Years. Available at: http://www.csep.ca. Accessed November 1, 2012. 27. Cole TJ, Bellizzi MC, Flegal KM, Dietz HW. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320:1-6. 28. Canadian Society of Exercise Physiology. Canadian Sedentary Activity Guidelines for Children 5-11. Available at: http://www.csep.ca. Accessed November 1, 2012. 29. Pouliou T, Elliott SJ. Individual and socio-environmental determinants of overweight and obesity in urban Canada. Health Place. 2010;16:389-398. 30. Rasbash J, Browne W, Healy M, et al. MLwiN version 2.02. Bristol, UK. Institute of Education, Multilevel Models Project; 2005. 31. SAS Institute Inc. The SAS System for Windows: Version 9.2. Cary, NC: SAS Institute Inc.; 2002. 32. Doak CM, Visscher TLS, Renders CM, Seidell JC. The prevention of overweight and obesity in children and adolescents: a review of interventions and programmes. Obes Rev. 2006;7:111-136. 33. Sallis JF, Glanz K. The role of the built environment in physical activity, eating and obesity in childhood. Fut Child. 2006;16:89-108. 34. Janssen I, Katzmarzyk PT, Boyce WF, et al. Comparison of overweight and obesity prevalence in school-aged youth from 34 countries and their relationships with physical activity and dietary patterns. Obes Rev. 2005;6:123-132.
Am J Health Behav.™ 2014;38(2):200-207
DOI:
http://dx.doi.org/10.5993/AJHB.38.2.5
207
Copyright of American Journal of Health Behavior is the property of PNG Publications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
R
egular participation in physical activity (PA) is critical for healthy child development. PA is associated with a decreased risk for chronic illness and obesity, and improved physical, mental, and social well-being among children.1,2 Unfortunately, PA levels are declining among children3 as time spent in sedentary behaviors is becoming more common.4-6 Considering that health behavior patterns of children may become resistant to change around the 6th grade,7 understanding the factors that impact PA among young children should be a priority. Existing research suggests that student-level characteristics (eg, sex, body mass index (BMI), sedentary behavior, and activity preferences) are associated with PA levels in older children.5,8-14 However, because school-based PA can account for up to 40% of the total activity among youth populations in Canada,15 it suggests that there is a need to move beyond only examining individual student characteristics associated with youth PA if we want to understand how best to target school-based PA interventions so that they are most likely to have impact.16,17 It may be just as important to identify the school-level correlates of children’s PA. For instance, whereas there is evidence demonstrating that characteristics of the school PA policy environment or the built environment surrounding a school are important predictors of PA for students
Scott T. Leatherdale, University of Waterloo, School of Public Health and Health Systems, Waterloo, ON, Canada. Correspondence Dr Leatherdale; [email protected]
200
and overweight or obese students were less likely to be moderately or highly active. Conclusions: Research should evaluate if implementing a school policy that provides student access to facilities and equipment outside of school hours is effective in promoting physical activity in grade 1 to 4 students. Key words: physical activity, children, school policy Am J Health Behav. 2014;38(2):200-207 DOI: http://dx.doi.org/10.5993/AJHB.38.2.5
in higher grades (5 to 8) even when controlling for student characteristics,16-20 similar evidence does not yet exist for children in grades 1 to 4. Using a similar approach as research published with elementary school students in grades 5 to 8,16,17 this exploratory study is the first to examine how school-based PA policies and the number of recreation centers surrounding a school were associated with the PA levels of a sample of elementary school students in grades 1 to 4. METHODS Design This cross-sectional study used data collected in 2007-08 from a convenience sample of grade 1 to 4 students attending 30 elementary schools (Ontario, Canada) as part of the PLAY-On study.21 The purpose of the PLAY-On study was to examine the school-level factors associated with PA and overweight among a sample of elementary school students in grades 1 to 8. Given the different data collection protocols required for respondents of this age range, there were 2 separate study arms of the PLAY-On host study: the first arm of the study was among respondents in grades 5 to 8 (aged 10 to 13 years) and the second arm of the study was among respondents in grades 1 to 4 (aged 6 to 9 years). Research has previously examined the school-level factors associated with overweight5,22 and PA16,17 among the grade 5 to 8 respondents in PLAY-On. The following study examines the PA data collected from the sample of grade 1 to 4 respondents in PLAY-On. In this component of PLAY-On, studentlevel data were collected from eligible students
Leatherdale in grades 1 to 4 using 2 methods: data provided by parents, and data provided by students to the PLAY-On study staff. School-level built environment data were provided by the Enhanced Points of Interest (EPOI) data resource. The EPOI data file is a database of the type and location of different opportunity structures within the built environment (eg, recreation facilities grocery stores, fast-food restaurants). Additional details about the DMTI-EPOI resources are available online (http://www.dmtispatial.com). The school-level PA policy data were provided by the elementary school version of the Physical Activity Module (PAM) of the Healthy School Planner (HSP). The HSP-PAM is a tool designed to assess policies, activities, committees, facilities and guidelines surrounding PA in the school environment. Additional details about the HSP-PAM measures and assessment categories are available online (http:// www.healthyschoolplanner.uwaterloo.ca).
Measures Outcome – physical activity level. Consistent with evidence that parental proxy report of young children’s PA levels is a valid and reliable measure of youth PA,24,25 parents were asked to report: “On
average, how many hours per day is your child physically active? Please include both moderate activity (eg, walking, biking to school, recreational swimming) and vigorous activity (eg, jogging, active team sports, fast dancing).” Based on Canadian PA guidelines for children,26 PA was coded as follows: low active (3 hours per day). Student-level predictor variables. Body Mass Index (BMI) was calculated for each student using the objective measures of weight (kg) and height (m) (BMI=kg/m2). Weight status was then determined using the BMI classification system of the International Obesity Task Force27 based on age and sex adjusted BMI cut-points for individuals at 18 years of age. Parents were asked to report: “On an average day, how much time does your child spend watching TV/movies, playing video/ computer games, surfing the internet, instant messaging or talking on the phone?” Based on Canadian sedentary behavior guidelines for children,28 sedentary behavior was coded as follows: low sedentary (≤2 hours), moderately sedentary (2 to 4 hours), and very sedentary (>4 hours). Students were asked 6 questions by the nurse collecting the Play-On data. Compared to other kids in your class, when you are at school (during recess, lunch, nutrition breaks) are you…(more active, less active, about the same)? Do you prefer to play alone or do you prefer to play with other children (alone, other children, both)? Do you prefer active games (eg, tag, kickball) or do you prefer quiet games (eg, board games) (active games, quiet games, both)? Do you like playing sports (eg, soccer, basketball) or do you dislike playing sports (likes sports, dislikes sports)? Do you like to read or do you dislike reading (likes reading, dislikes reading)? Do you like to play outside or do you like to play inside (outside, inside, both)? School-level predictor variables. The 2008 EPOI data were used to identify recreational facilities (includes dance studios, fitness/gym facilities, and sport and recreation clubs) within a 1-km buffer surrounding each of the 30 schools using the same method as Pouliou and Elliott.29 As described and presented elsewhere,16 the HSP tool measured PA policies in the 30 schools. In brief, the HSP PA tool measures school PA policies aligned with the 4 components that form the basis of the Foundations for a Healthy School (FHS): Healthy Physical Environment (availability of, access to, and adequacy in meeting student needs for, indoor and outdoor facilities, equipment and resources for safe, quality PA on or near school grounds, both during and outside of school hours); Instruction and Programs (availability, delivery and characteristics of curricular physical education, extracurricular PA programs, and active transportation to school, including barriers to implementing such programs); Supportive Social Environment (characteristics of the school’s
Am J Health Behav.™ 2014;38(2):200-207
DOI:
Data Collection All students in grades 1 to 4 at the participating schools were eligible to participate. Prior to participating, active consent from parents (or guardians) was required and at any time students were able to decline participation. During the active consent process, parents were asked to report their child’s PA. During the data collection from students, the registered nurse working for PLAY-On asked each eligible student 6 questions pertaining to their physical activity routines and preferences. After their responses were recorded by the nurse, students were then asked to remove their footwear, and height and weight were objectively measured and recorded by the nurse. Students were asked to stand upright with their back against the wall, and then the nurse used a square placed on the top of their head to measure their height to the nearest centimeter against a standard metric tape measure fastened to the wall. Weight was measured to the nearest one decimal place in kilograms using a Bathscapes® LED digital bath scale (model 1053WWBA). At each school, the administrator(s) most knowledgeable about the school’s PA policies completed the HSP survey. Participants Of the 3926 students enrolled in grades 1 to 4 at the 30 participating elementary schools, 59.4% (N = 2331) completed the survey. Missing respondents resulted from parent refusal and absenteeism on the day of the survey. This distribution is consistent with an active consent study examining overweight among Canadian elementary students.23
http://dx.doi.org/10.5993/AJHB.38.2.5
201
School Characteristics and Physical Activity among Grade 1-4 Students
Table 1 Descriptive Statistics for Youth in Grades 1 to 4 by Sex in PLAY-On (Ontario, Canada, 2007-2008)
Student-level characteristics
Male (N = 1187)
Female (N = 1139)
%
%
(n)
a
(n)
Chi-square
a
Physical Activity Level b
Low active Moderately active Very active
10.0 (115) 63.3 (726) 26.7 (306)
14.3 (158) 62.7 (695) 23.0 (255)
χ2=11.41, df=2, p < .01
Grade
1 2 3 4
22.7 (270) 21.5 (255) 27.5 (326) 28.3 (336)
22.4 (256) 21.1 (241) 27.1 (309) 29.4 (335)
χ2=0.32, df=3, p = .95
Normal weight Overweight Obese Missing
75.0 (890) 13.7 (162) 6.1 (144) 5.2 (62)
74.5 (848) 14.0 (159) 6.2 (71) 5.4 (61)
χ2=0.09, df=3 , p = .99
Sedentary behavior b
≤2 hour per day 2 to 4 hours per day >4 hours per day
70.6 (810) 21.8 (250) 7.6 (87)
75.2 (834) 18.9 (210) 5.9 (65)
χ2=6.37, df=2, p < .05
Physical activity level compared to other kids e
More active About the same Less active
7.4 (86) 28.5 (333) 64.2 (750)
9.3 (105) 18.2 (206) 72.5 (822)
χ2=34.56, df=2, p < .001
Prefers to play alone or with other children e
Alone Both Other children
2.4 (28) 57.1 (667) 40.5 (474)
1.0 (11) 54.8 (621) 44.2 (501)
χ2=9.24, df=2, p < .01
Prefers to play active games or quiet games e
Active Both Quiet
4.5 (52) 41.4 (484) 54.1 (633)
7.6 (86) 35.1 (398) 57.3 (649)
χ2=16.40, df=2, p < .001
Preference for playing sports e
Likes sports Dislikes sports
96.5 (1,127) 3.5 (41)
92.7 (1,049) 7.3 (83)
χ2=16.46, df=1, p < .001
Preference for reading e
Likes reading Dislikes reading
79.6 (931) 20.4 (238)
89.5 (1,014) 10.5 (119)
χ2=42.66, df=1, p < .001
Prefers to play outside or inside e
Outside Both Inside
4.1 (48) 20.6 (241) 75.3 (880)
4.2 (48) 17.2 (195) 78.6 (890)
χ2=4.35, df=2, p = .11
c
Weight status
c,d
Note. a Numbers may not add to total because of missing values b Reported by parent or guardian c Measured by PLAY-On nurse d Body mass index (BMI) values used to determine weight status have been adjusted for age and sex e Reported by student and recorded by PLAY-On nurse
social environment that predispose, reinforce and enable enjoyable, lifelong participation in PA or that hinder such activities); and Community Partnerships (the accessibility and availability of support services for PA which may include partnerships with public health units and community based services and resources). Each indicator was assigned a classification based on the corresponding phase of implementation as outlined by the HSP developers: Initiation (falls short or exhibits extensive room for improvement in meet-
202
ing the recommendations related to school capacity for PA); Action (meets the recommendations in several, but not all areas related to school capacity for PA, exhibits some room for improvement); Maintenance (consistently meets or exceeds the recommendations related to school capacity for PA, encouraged to maintain the current level of commitment to supporting PA at school). Each of the 4 FHS components was also assigned an ‘overall’ phase classification based on the combined responses to component indicators.
Leatherdale
Table 2 Results of the Univariate Multi-level Logistic Regression Analyses Examining School-level Characteristics Associated with being Moderately or Highly Active among Youth in Grades 1 to 4 in PLAY-On (Ontario, Canada, 2007-2008) Model Estimates (Standard Error) Phase
Model 1 Moderately Active vs. Low Active
Model 2 Highly Active vs. Low Active
Student access to a variety of facilities on and off school grounds during school hours †
Action Maintenance
0.35 (0.52) 0.28 (0.51)
0.07 (0.54) 0.18 (0.54)
Availability of physical activities during inclement weather †
Action Maintenance
-0.19 (0.25) 0.17 (0.86)
0.19 (0.25) -0.69 (0.98)
Student access to facilities and equipment outside of school hours †
Action Maintenance
-0.01 (0.24) 1.44 (0.62) *
0.08 (0.24) 1.79 (0.80) *
Support for active transportation to and from school †
Action Maintenance
-0.02 (0.30) -0.34 (0.33)
-0.05 (0.30) -0.54 (0.33)
Action
0.37 (0.81)
0.13 (0.27)
Implementation of daily PA ‡
Maintenance
0.29 (0.26)
0.27 (0.27)
Time spent per week engaged in PA during physical education classes †
Action Maintenance
-0.26 (0.78) -0.61 (0.81)
-0.50 (0.79) -0.87 (0.82)
Classes taught by a qualified physical education specialist †
Action
0.03 (0.40)
0.22 (0.41)
Availability and use of intramural/club activities †
Action Maintenance
0.24 (0.35) 0.13 (0.50)
0.44 (0.35) 0.26 (0.50)
Consistency of intramural programming across grade divisions and seasons †
Action Maintenance
0.21 (0.27) 0.03 (0.35)
0.22 (0.28) 0.16 (0.37)
Availability and use of interschool programs †
Action Maintenance
0.29 (0.25) 0.22 (0.52)
0.09 (0.27) -0.07 (0.54)
Maintenance
0.39 (0.85)
0.42 (0.86)
Action
0.29 (0.26)
0.27 (0.27)
Emphasis placed on maximizing participation in PA through school programs †
Action Maintenance
0.69 (0.39) 0.75 (0.31) *
0.92 (0.41) * 0.84 (0.36) *
Incorporation of PA into other school subjects †
Action Maintenance
0.15 (0.30) 0.14 (0.41)
0.28 (0.31) 0.09 (0.42)
Special recognition of students who participate in school physical activities †
Action Maintenance
0.11 (0.46) 0.04 (0.40)
-0.05 (0.47) -0.14 (0.42)
Formal collection of suggestions from the school community about PA at school †
Action Maintenance
0.28 (0.28) 0.15 (0.37)
0.05 (030) 0.05 (0.40)
Promotion of PA programs and events for students, families and school staff †
Action Maintenance
0.59 (0.25) * -0.20 (0.25)
0.63 (0.27) * -0.10 (0.27)
Use of PA as a reward, not as discipline †
Action Maintenance
-0.25 (0.27) -0.04 (0.34)
-0.14 (0.28) -0.22 (0.36)
Presence of written policies or practices that support PA †
Action Maintenance
0.05 (0.31) -0.32 (0.35)
0.14 (0.33) -0.20 (0.36)
Action
0.28 (0.80)
0.60 (0.82)
Support available for school staff involved with PA ‡
Maintenance
-0.24 (0.25)
-0.11 (0.27)
Connection to community resources †
Action Maintenance
-0.36 (0.38) -0.44 (0.30)
-0.34 (0.40) -0.35 (0.32)
Action Maintenance
-0.09 (0.36) -0.33 (0.33)
-0.14 (0.38) -0.19 (0.35)
-0.30 (0.09) *
-0.29 (0.10) *
Healthy Physical Environment
Overall score for this indicator † Instruction and Programs
Consistency of interschool programming across seasons † Overall score for this indicator † Supportive Social Environment
Overall score for this indicator † Community Partnerships
Overall score for this indicator † Number of opportunity structures within a 1 kilometer (km) radius of a school Recreation facilities
* p < .05 Note. Model 1: 1 = Moderately Active (N = 1,406), 0 = Low Active (N = 268) Model 2: 1 = Highly Active (N = 553), 0 = Low Active (N = 268) † Reference group is Initiation, ‡ Reference group is Action
Am J Health Behav.™ 2014;38(2):200-207
DOI:
http://dx.doi.org/10.5993/AJHB.38.2.5
203
School Characteristics and Physical Activity among Grade 1-4 Students Analyses Using student-level data, the prevalence of PA, parent measures and student measures were examined by sex. Consistent with previous research,6,9,16,17 2 series of multi-level logistic regression analyses examined characteristics associated with being (1) moderately active versus low active, and (2) highly active versus low active; each analysis used a stepped modelling procedure. Step 1 examined if the differences in being moderately active or highly active were random or fixed across schools. The school-level variance term from Step 1 (σ2μ0) was used to calculate the intraclass correlation (ICC). In Step 2, a series of univariate analyses were used as exploratory analyses to examine if the implementation phase score for each of the FHS indicators were associated with being moderately active or highly active. Only significant school-level variables (p < .05) were retained for further analyses. In Step 3, multivariate models were developed to examine how the student characteristics and the significant school characteristics identified in Step 2 were associated with being moderately active (Model 1) and highly active (Model 2). Since most students (94.6%) reported that they like playing sports, it was excluded from the multivariate predictive models. The multi-level statistical analyses were conducted on MLwiN Version 2.0230 and the descriptive statistics were conducted on SAS 9.2.31 RESULTS Demographic characteristics of the students are presented in Table 1. In brief, the sample was 51.0% (N = 1187) male and 49.0% (N = 1139) female. A total of 273 (12.1%) students were considered low active, 1421 (63.0%) were considered moderately active, and 561 (24.9%) were considered very active by their parents. Males were more likely to be considered very active compared to females (c2=11.41, df=2, p < .001). The mean number of recreation facilities within a 1-km buffer of the schools was 0.8 (range, 0 to 4). As presented previously,6,16 there was substantial variability across schools in relation to the PA policy indicators measured. Factors Associated with being Moderately Active Among students in grades 1 to 4, there was significant between-school random variation in being moderately active [σ2μ0=0.258(0.105), p < .05]; school-level differences accounted for 7.3% of the variability in the odds of a student being moderately active versus low active. The univariate analyses (Table 2) revealed that the number of recreation centers and 3 school-level policy characteristics (student access to facilities and equipment outside of school hour, emphasis placed on maximizing participation in PA through school programs, promotion of PA programs and events for students, families and school staff) were significant, and thus, in-
204
cluded in the final model. In the final multivariate model (Table 3: Model 1), only one school characteristic was significantly associated with being moderately active when controlling for student characteristics. If a student attended a school that was in the maintenance phase for the indicator student access to facilities and equipment outside of school hours, he/she was over twice as likely to be moderately active than a similar student attending a school that was in the initiation phase for this indicator (OR=2.82). This model also revealed that students considered overweight (OR=0.64) or obese (OR=0.59) were less likely to be moderately active. Conversely, students spending 2 to 4 hours per day in sedentary behaviors (OR=1.56) or students who reported they prefer to play active games (OR=2.18) were more likely to be moderately active. Factors Associated with being Highly Active Among students in grades 1 to 4, there was significant between-school random variation in being highly active [σ2μ0=0.257(0.113), p < .05]; schoollevel differences accounted for 7.3% of the variability in the odds of a student being highly active versus low active. The univariate analyses (Table 2) revealed that the number of recreation centres and 3 of the school-level policy characteristics (student access to facilities and equipment outside of school hours, emphasis placed on maximizing participation in PA through school programs, promotion of PA programs and events for students, families and school staff) were significant, and thus, included in the final model. In the final multivariate model (Table 3: Model 2), only one school characteristic was significantly associated with being highly active when controlling for student characteristics. If a student attended a school that was in the maintenance phase for the indicator student access to facilities and equipment outside of school hours, he/ she was over 5 times as likely to be highly active than a similar student attending a school that was in the initiation phase for this indicator (OR=5.24). This model also revealed that students considered overweight (OR=0.51) or obese (OR=0.37) were less likely to be highly active. Conversely, male students (OR=1.67) and students spending 2 to 4 hours per day in sedentary behaviors (OR=1.91) were more likely to be highly active. DISCUSSION Considering that few researchers have data linking school PA policies to the PA levels of youth in grades 1 to 4, this study provides a unique new contribution to the literature by being the first to examine how school characteristics (both the school PA policy environment and the built environment surrounding a school) were associated with PA levels of young children in grades 1 to 4. Specifically, we identified that even when controlling for individual student characteristics, the PA levels of grade 1 to 4 students can vary as a function of the characteristics of the school they attend.
Leatherdale
Table 3 Multi-level Logistic Regression Analyses of School- and Student-level Characteristics Associated with Physical Activity among Youth in Grades 1 to 4 in PLAY-On (Ontario, Canada, 2007-2008) Adjusted Odds Ratio a (95% CI) Model 1 Model 2 Moderately Active Highly Active vs. Low Active vs. Low Active Student-Level Characteristics Sex
Female Male
1.00 1.36 (0.77, 1.96)
1.00 1.67 (1.20, 2.32)**
Grade
1 2 3 4
1.00 1.07 (0.71, 1.62) 1.02 (0.69, 1.52) 1.07 (0.72, 1.59)
1.00 0.78 (0.49, 1.27) 0.75 (0.48, 1.19) 0.73 (0.46, 1.16)
Weight status b
Normal weight Overweight Obese Missing BMI
1.00 0.64 (0.43, 0.94)* 0.59 (0.36, 0.97)* 0.60 (0.33, 1.07)
1.00 0.51 (0.32, 0.81)** 0.37 (0.19, 0.70)** 0.54 (0.26, 1.10)
Sedentary behavior
≤2 hour per day 1.00 2 to 4 hours per day 1.56 (1.11, 2.20)** >4 hours per day 2.69 (0.79, 9.20)
1.00 1.91 (1.30, 2.81)** 2.84 (0.95, 8.47)
Physical activity level compared to other kids
About the same More active Less active
1.00 0.78 (0.46, 1.32) 1.23 (0.88, 1.71)
1.00 0.76 (0.40, 1.41) 1.11 (0.75, 1.65)
Prefers to play alone or with other children
Alone Other children Both
1.00 0.84 (0.27, 2.62) 0.98 (0.32, 3.03)
1.00 0.73 (0.19, 2.84) 0.78 (0.20, 3.05)
Prefers to play active games or quiet games
Quiet Active Both
1.00 2.18 (1.08, 4.40)* 1.21 (0.90, 1.63)
1.00 1.16 (0.49, 2.74) 1.24 (0.88, 1.75)
Preference for reading
Dislikes reading Likes reading
1.00 0.67 (0.43, 1.05)
1.00 0.66 (0.40, 1.08)
Prefers to play outside or inside
Inside Outside Both
1.00 0.73 (0.39, 1.38) 1.34 (0.92, 1.95)
1.00 0.66 (0.30, 1.47) 1.04 (0.67, 1.61)
Initiation Action Maintenance
1.00 1.23 (0.77, 1.96) 2.82 (1.14, 6.97)*
1.00 1.50 (0.89, 2.53) 5.24 (1.61, 17.06)**
Emphasis placed on maximizing participation in PA through school programs
Initiation Action Maintenance
1.00 1.50 (0.71, 3.18) 1.53 (0.78, 2.97)
1.00 1.96 (0.83, 4.66) 1.78 (0.82, 3.84)
Promotion of PA programs and events for students, families and school staff
Initiation Action Maintenance
1.00 1.45 (0.82, 2.56) 0.84 (0.49, 1.45)
1.00 1.39 (0.74, 2.62) 0.86 (0.47, 1.58)
1 unit increase
1.18 (0.95, 1.48)
1.17 (0.91, 1.51)
School-Level Characteristics Healthy Physical Environment
Student access to facilities and equipment outside of school hours Supportive Social Environment
Opportunity Structures in the Built Environment Number of recreation centres within 1-km radius * p < .05
**p < .01
Note. a Odds ratios adjusted for all other variables in the table and controlling for random variation across schools. b BMI values used to determine weight status have been adjusted for age and sex Model 1: 1 = Moderately Active (N = 1,406), 0 = Low Active (N = 268) Model 2: 1 = Highly Active (N = 553), 0 = Low Active (N = 268)
Am J Health Behav.™ 2014;38(2):200-207
DOI:
http://dx.doi.org/10.5993/AJHB.38.2.5
205
School Characteristics and Physical Activity among Grade 1-4 Students This is valuable insight for informing school-based PA prevention programming as it provides new evidence that the PA levels of young children are associated with their schools PA policy environment, and evidence justifying the need for school-based interventions early in elementary school because this study identified significant between-school variation which could be amenable to school-level intervention; if a school-level program or policy has a small effect on either shifting or normalizing the distribution of PA across schools, the impact at the student-level would be substantial.5 Consistent with PA research among students in grades 5 to 8,17 the results highlight that schools should work to ensure that they maintain or implement policies to provide students with access to facilities and equipment outside of school hours. However, there was no association between the number of recreation centers surrounding a school and PA among grade 1 to 4 students. This is consistent with evidence that suggests the built environment may not have an impact on younger students.32,33 Future research may not need to consider the built environment for this age group. Important student-level characteristics were also identified. First, boys were more likely than girls to be very active. This is consistent with previous research12,14 and may be related to differences in activity preferences among young boys and girls. Parents and educators should consider providing children with sex specific PA opportunities that incorporate the different types of student activity preferences identified here (eg, active games for boys). Second, consistent with evidence from older students,11,14,34 overweight and obese students were less likely to be moderately or highly active. Although we cannot determine the temporal nature of this relationship, this finding suggests that it may be wise to target interventions to promote PA to overweight children because they appear to be at increased risk for inactivity relative to normal weight youth. Such a targeted approached to PA promotion would require evaluation. Third, students were more apt to be active if they engaged in 2-4 hours per day of sedentary behavior per day. Although this appears counterintuitive, research with older youth has found that youth who spend substantial time engaging in sedentary behaviors can also spend substantial time engaging in PA.10 However, because sedentary behaviors are associated with youth obesity,9 strategies to reduce sedentary behavior among young children may still remain warranted. This study is subject to limitations. The crosssectional nature of the study design does not allow for inferences regarding causality. The self-report nature of the PA data is a limitation, but given that this is a secondary analysis of a host study designed to measure obesity in youth and evidence that parental proxy reports of young children’s PA levels are valid and reliable measure of youth PA,24,25 this was our only option within the Play-
206
On study. The generalizability of these findings is also unknown as limited school- and student-level activity data on these grades exists. Conclusions Characteristics of the school that a grade 1 to 4 student attends are associated with their physical activity levels. Future research should evaluate if implementing and promoting a school policy that provides student access to facilities and equipment outside of school hours is effective in promoting higher levels of PA among students in grades 1 to 4. At the student-level, it appears that activities promoting PA should be targeted to overweight or obese youth as they are the least likely to be moderately or highly active. Developing a better understanding of the school- and student-level characteristics associated with PA among children is critical for informing the targeting and tailoring of future intervention programs and policies. Human Subjects Statement The University of Waterloo Office of Research Ethics and appropriate school board ethics committees approved the study procedures. Conflict of Interest Statement The author declares that he have no conflict of interest or competing interests in this research. Acknowledgements Data used in this analysis were drawn from the PLAY-On project, funded by the Heart and Stroke Foundation of Ontario (grant awarded to S. Leatherdale). Dr Leatherdale is a Cancer Care Ontario Research Chair in Population Studies. REFERENCES
1. Public Health Agency of Canada. Physical Activity Benefits, 2011. Available at: http://www.phac-aspc.gc.ca. Accessed November 3, 2012. 2. Janssen I, LeBlanc AG. Systematic review of the health benefits of physical activity and fitness in school-aged children and youth. Int J Behav Nutr Phys Act. 2010;7:116. 3. Shields M. Overweight and obesity among children and youth. Health Rep. 2006;17:27-42. 4. Leatherdale ST, Ahmed R. Screen-based sedentary behaviours among a nationally representative sample of youth: are Canadian kids couch potatoes? Chronic Dis Inj Can. 2011;31:141-146. 5. Leatherdale ST. The association between overweight and school policies on physical activity: a multi-level analysis among elementary school youth in the PLAY-On study. Health Educ Res. 2010;5;1061-1073. 6. Leatherdale ST, Faulkner G, Arbour-Nicitopoulos K. School and student characteristics associated with screen time sedentary behavior among elementary school youth in the PLAY-On study: a multi-level analysis. Prev Chronic Dis. 2010;7:A128 7. Kelder SH, Chetyl L, Lytle L. Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. Am J Pub Health. 1994;84:1121-1126. 8. Kurc A, Leatherdale ST. The effect of social support and school- and community-based sports on youth physical activity. Can J Pub Health. 2009;100:60-64.
Leatherdale 9. Leatherdale ST, Papadakis S. A multi-level examination of the association between older social models in the school environment and overweight and obesity among younger students. J Youth Adolesc. 2011;40:361-372. 10. Wong SL, Leatherdale ST. Association between sedentary behaviour, physical activity, and obesity: inactivity among active kids. Prev Chronic Dis. 2009;6:A26. 11. Leatherdale ST, Wong SL. Modifiable characteristics associated with sedentary behaviours among youth. Int J Pediatr Obes. 2008;3:93-101. 12. Van der Horst K, Paw MJCA, Twisk JWR, van Mechelen W. A brief review on correlates of physical activity and sedentariness in youth. Med Sci Sports Exerc. 2007;39:1241-1250. 13. Young DR, Felton GM, Grieser M, et al. Policies and opportunities for physical activity in middle school environments. J Sch Health. 2007;77:41-47. 14. Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of physical activity of children and adolescents. Med Sci Sports Exerc. 2000;32:963-975. 15. Craig CL, Cameron C, Russell SJ, Beaulieu A. Increasing physical activity: Supporting children’s participation. Ottawa: Canadian Fitness and Lifestyle Research Institute; 2001. 16. Hobin EP, Leatherdale ST, Manske SR, Robertson-Wilson JE. A multilevel examination of school and student characteristics associated with moderate and high levels of physical activity among elementary school students (Ontario, Canada). Can J Pub Health. 2010;101:495-499. 17. Leatherdale ST, Manske S, Faulkner G, et al. A multilevel examination of school programs, policies and resources associated with physical activity among elementary school youth in the PLAY-ON study. Int J Behav Nutr Phys Act. 2010;7:6. 18. Naylor PJ, Macdonald HM, Warburton DER, et al. An active school model to promote physical activity in elementary schools: Action Schools! BC. Br J Sports Med. 2008;42:338-343. 19. Sallis JF, Conway TL, Prochaska JJ, et al. The association of school environments with youth physical activity. Am J Pub Health. 2001;91:618-620. 20. Zhu W. A multilevel analysis of school factors associated with health-related fitness. Res Q Exerc Sport. 1997;68:125-135. 21. The Play-On Study. Available at: http://www.shapes.uwater-
loo.ca/index.cfm?section=1002&page=1041. Accessed November 13, 2012. 22. Leatherdale ST, Pouliou T, Church D, Hobin E. The association between overweight and opportunity structures in the built environment: a multi-level analysis among elementary school youth in the PLAY-On study. Int J Public Health, 2011;56:237-246. 23. Veugelers PJ, Fitzgerald AL. Effectiveness of school programs in preventing childhood obesity: a multilevel comparison. Am J Pub Health. 2005;95:432-435. 24. Burdette HL, Whitaker RC, Daniels SR. Parental reports of outdoor playtime as a measure of physical activity in preschool-aged children. Arch Pediatr Adolesc Med. 2006;158:353-357. 25. Manios Y, Kafatos A, Markakis G. Physical activity in 6-year-old children: validation of two proxy reports. Pediatr Exerc Sci. 1998;10:176-188. 26. Canadian Society of Exercise Physiology. Canadian Physical Activity Guidelines for Children 5-11 Years. Available at: http://www.csep.ca. Accessed November 1, 2012. 27. Cole TJ, Bellizzi MC, Flegal KM, Dietz HW. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320:1-6. 28. Canadian Society of Exercise Physiology. Canadian Sedentary Activity Guidelines for Children 5-11. Available at: http://www.csep.ca. Accessed November 1, 2012. 29. Pouliou T, Elliott SJ. Individual and socio-environmental determinants of overweight and obesity in urban Canada. Health Place. 2010;16:389-398. 30. Rasbash J, Browne W, Healy M, et al. MLwiN version 2.02. Bristol, UK. Institute of Education, Multilevel Models Project; 2005. 31. SAS Institute Inc. The SAS System for Windows: Version 9.2. Cary, NC: SAS Institute Inc.; 2002. 32. Doak CM, Visscher TLS, Renders CM, Seidell JC. The prevention of overweight and obesity in children and adolescents: a review of interventions and programmes. Obes Rev. 2006;7:111-136. 33. Sallis JF, Glanz K. The role of the built environment in physical activity, eating and obesity in childhood. Fut Child. 2006;16:89-108. 34. Janssen I, Katzmarzyk PT, Boyce WF, et al. Comparison of overweight and obesity prevalence in school-aged youth from 34 countries and their relationships with physical activity and dietary patterns. Obes Rev. 2005;6:123-132.
Am J Health Behav.™ 2014;38(2):200-207
DOI:
http://dx.doi.org/10.5993/AJHB.38.2.5
207
Copyright of American Journal of Health Behavior is the property of PNG Publications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
