A multilevel examination of school and student characteristics associated with moderate and high levels of physical activity among elementary school students (Ontario, Canada).
Hobin, Erin P. ; Leatherdale, Scott T. ; Manske, Steve R. 等
Physical inactivity is associated with an increased risk for
chronic illnesses. (1) Current data indicate that 87% of Canadian youth
do not meet the Canadian physical activity (PA) guidelines. (2) Lower
levels of PA have contributed to the rising prevalence of youth obesity
in Canada in the last two decades. (3,4) Increasing the PA levels of
Canadian youth is an important public health challenge.
Most existing PA interventions assume that PA is self-determined by
an individual, (5) despite ecological models suggesting there are
reciprocal influences between the individual and the environment in
which they are situated. (6,7) To increase PA at the population level, a
better understanding of environmental influences on PA is required.
Schools are an important environment for PA promotion since school-based
PA accounts for up to 40% of total activity among youth populations. (8)
Numerous school-based PA interventions have been targeted to individual
students, yet such programs tend to produce only modest effects. (9,10)
More recently, interventions have targeted characteristics of the school
environment. Significant between-school variability in PA exists (11-16)
and studies have begun to identify school characteristics that explain
that variability (e.g., having PA facilities on school grounds
(13,15,16)). Additional research is needed to examine how school
policies and programs are associated with PA among students.
In an effort to promote youth PA, many health and education experts
have called upon schools to offer daily physical education (PE). (17,18)
Despite this call to action for daily PE, most provincial and
territorial governments have instead opted for a policy requiring
schools to provide elementary students with a specified amount of daily
PA. The Ontario Ministry of Education (OME), for example, recently
mandated Memorandum 138 [Daily Physical Activity (DPA)] whereby
"school boards must ensure ... a minimum of 20 minutes of sustained
moderate to vigorous PA each school day during instructional time."
(19) However, this policy does not mandate how schools or teachers must
implement DPA into their schedules. Moreover, the frequency of PE
classes and the provision of other PA programs (e.g., intramurals) are
not specified by government policy, creating differences in
implementation across schools. Our understanding of how these school
PA-related policies and programs are associated with student PA levels
is largely unknown. Developing this understanding is critical for
appropriately tailoring and targeting PA interventions; therefore, the
aim of this study was to identify the school and student characteristics
associated with moderate and high levels of PA in a sample of Ontario
elementary schools.
METHODS
Design
This cross-sectional study used self-reported data collected in
20072008 among students in grades 5 to 8 students attending 30
elementary schools in Ontario as part of the PLAY-ON study. (20) Schools
were purposively selected within public and separate school boards
located in major geographic regions of Ontario. Self-reported
student-level data were collected from consenting students using the
SHAPES Physical Activity Module (PAM). (21,22) The PAM asks students
about PA, sedentary behaviours, correlates for PA, and participation in
teams and sporting activities at school. Validity testing has
demonstrated significant criterion validity based on Spearman
correlations for our self-reported measure of PA (r=0.44, p<0.01).
(23)
School-level data were collected from administrators using the PA
categories of the elementary school version of the School Health
Environment Survey (SHES). (24) Aligned with the Government of
Ontario's Foundations for a Healthy School framework, (25) the
school-level module assesses PA-related policies, programs and
facilities in the school environment. Additional details about SHES are
available in print24 and online (www.shapes.uwaterloo.ca/SHES or
http://www.healthyschoolplanner.uwaterloo.ca/
jcshsite_app/controller/index.cfm).
Procedure
All students at the participating schools were eligible to
participate. Prior to participating in the study, active consent from
parents was required and at any time students were able to decline
participation. Eligible students completed the PAM during class time
without compensation for participation. At each participating school,
the administrator(s) most knowledgeable about the school's
programs, policies and facilities was asked to complete the SHES survey.
The University of Waterloo Office of Research Ethics and appropriate
School Board Ethics committees approved the study procedures.
Participants
Of the 4,838 students enrolled in grades 5 to 8 at the 30
participating schools, 50.6% (n=2,449) participated in the survey;
missing respondents resulted from parent/student refusal and absenteeism
on the day of the survey. This distribution is consistent with a
previous active consent study examining PA among Canadian elementary
students.26 Since 2.8% (n=70) of participating students did not provide
PA data, the final sample was 2,379. The SHES survey was completed by
all 30 elementary schools.
Measures
Outcomes--Physical Activity
PA level was based on kilocalories per kilogram of body weight per
day (KKD). Using validated measures, (21) students were asked how many
minutes of vigorous PA (VPA) and moderate PA (MPA) they engaged in on
each of the last seven days. The average KKD expended in VPA and MPA
were calculated as: KKD= [Hours of VPA*6MET) + (Hours of MPA*3MET)]/7
days.
Since youth over-report time spent doing PA in self-report (23,27)
and categorical dependent variables make the results more meaningful to
stakeholders, (28) our data have been coded to differentiate groups of
active and inactive students: students one standard deviation or more
(<16th percentile) below the sample mean for KKD were classified as
low active, students one standard deviation or more (>84th
percentile) above the sample mean for KKD were classified as highly
active; all others were classified as moderately active.
Student-level Characteristics
Consistent with previous research, (26) students were asked to
report their sex, grade, and the number of PE classes in the previous 7
days (<2 PE classes per week, >2 PE classes per week). Frequency
of PE participation was categorized as a student-level variable since
not all students within a given school reported the same frequency of PE
participation (e.g., scheduling differences between classes, student
absence).
School-level Characteristics
Offering intramural PA programs was measured by asking
administrators, "Does your school offer intramural programs/club
activities that involve PA?" (Yes/No). Offering interschool PA
programs was measured by asking administrators, "Does your school
offer interschool programs that involve PA? (Yes/No). To measure the
chosen implementation model for DPA in schools, administrators were
asked to report how the DPA policy is implemented at their school (i.e.,
in addition to daily PE class, as part of daily PE class, or only on
days with no PE class).
Data analysis
A basic 2-level nested structure was created where individual
students (level 1) were nested within schools (level 2). As such, multi
level logistic regression analyses were then used to contrast 1)
moderately active and low active students, and 2) highly active and low
active students. Each analysis used a 3-step modeling procedure. Step 1
tested whether the odds of being either moderately active or highly
active were random or fixed across schools. Step 2 included a series of
univariate analyses examining whether each of the school-level
indicators was associated with being moderately active or highly active.
Only significant school-level variables (p<0.05) were retained for
further analyses. In Step 3, multivariate models were developed to
examine how the significant school-level characteristics identified in
Step 2 and student-level characteristics were associated with being
moderately active (Model 1) and highly active (Model 2). Analyses were
conducted using SAS version 9.1.3 (Cary, NC).
RESULTS
Student characteristics
The sample was 47.3% (n=1,126) male and 52.7% (n=1,253) female. As
presented in Table 1, 16.4% of students were classified as low active,
67.2% were classified as moderately active, and 16.4% were classified as
highly active. More males reported being highly active than females (X2
= 15.05, df =2, p<0.001). The PA levels of children excluded due to
missing data were not significantly different compared to those students
retained for regression analysis.
School characteristics
The mean prevalence of moderately active students at a school was
66.6% (range, 53.6% to 82.8%) and the mean prevalence of highly active
students at a school was 16.6% (range, 4.2% to 30.3%). As presented in
Table 2, the majority of schools offered intramural (83.3%) and
interschool (90%) sports programs. Chosen DPA implementation models
varied between schools, with 70% offering DPA only on days without PE
class, 20% offering DPA in addition to daily PE class, and 10% offering
DPA as part of daily PE class.
Characteristics associated with being moderately active or highly
active
Significant between-school variation was identified for being
moderately active [[[sigma].sup.2])[sub.[mu]0]0.166(0.043), p<0.001]
and highly active [[[sigma].sup.2] [sub.[mu]0]=0.259(0.067),
p<0.001]. Using the null models, we found school-level differences
accounted for 4.8% of the variability in the odds of being moderately
active and 7.3% of the variability in the odds of being highly active
when controlling for individual-level variance. After completing the
final models, we found the proportion of between-school variability
explained by the school-level factors included in the models was 72.5%
for the odds of being moderately active and 81.0% for the odds of being
highly active.
Moderately active (Model 1)
A student participating in >2 PE classes per week was more
likely to be moderately active than a student participating in <2 PE
classes per week (OR=1.86, 95% CI: 1.43-2.40). A student attending a
school that offered interschool programs was significantly less likely
to be moderately active than a student attending a school that did not
offer interschool programs (OR=0.61, 95% CI: 0.48-0.77). There were no
significant associations with sex, grade, implementing DPA as part of
daily PE class or offering intramural PA programs at school.
Highly active (Model 2)
Female students were less likely to be highly active than male
students (OR=0.59, 95% CI: 0.43-0.81). Conversely, a student
participating in [greater than or equal to]2 PE classes per week was
more likely to be highly active than a student participating in <2 PE
classes per week (OR=2.12, 95% CI: 1.47-3.05). A student attending a
school that offered interschool programs was significantly less likely
to be highly active than a student attending a school that did not offer
interschool programs (OR=0.26, 95% CI: 0.18-0.36). There were no
significant associations with grade or offering intramural PA programs
at school. Since DPA implementation was not significant in the
univariate analysis, it was excluded from the final model.
DISCUSSION
The school environment appears to be associated with the PA levels
of youth. Consistent with previous research, (11-16) we identified
significant variation in PA across schools. Although the amount of
school-level variability identified was modest, from a population
perspective it is meaningful as even small shifts in PA at the school
level could result in a substantial population-level impact when applied
across a large number of schools. (29)
As anticipated, PA levels were associated with the frequency of
participating in PE classes. This is positive considering that there is
an emergence of education policies designed to increase the frequency of
PE classes or extend the number of PE credits required for graduation
(e.g., the policy recently implemented in Manitoba (30)) for the
purposes of increasing youth PA. This finding is also consistent with
the advice of stakeholders who have been advocating for schools to
provide daily PE classes to students. (17,18) DPA policies present
another curricular-based opportunity to increase student PA levels, yet
our results suggest that the DPA implementation model chosen by a school
was not associated with the PA levels of its students. More research
examining the association between student PA and DPA implementation
models and the association between the fidelity of DPA implementation
and students' overall PA is required.
We also identified that interschool PA programming is associated
with student PA. It was surprising that interschool programs were
associated with being low active since such programs provide students
with additional opportunities to be active. However, considering that
interschool programs are traditionally sports-based competitive
programs, they may in fact discourage less skilled students from
participating. Further research examining the role of interschool PA
programs on student PA is necessary.
Limitations
Causal relationships cannot be inferred from cross-sectional data.
Data were based on self-report and thus susceptible to several sources
of bias, however the tools used demonstrated reliability and validity,
(23) and honest reporting was encouraged by ensuring confidentiality
during data collection. Data were not available for examining other
relevant measures associated with implementation of PE curriculum (e.g.,
teacher training) or student overweight status.
CONCLUSIONS
There is substantial variation between schools in student PA.
Participating in at least two PE classes per week is associated with
increased PA levels, whereas attending a school with interschool PA
programs is associated with decreased PA levels. Additional research
evaluating different DPA implementation models and reconsidering how
interschool PA programs can best contribute to PA of youth is required.
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 5. Leatherdale). The project was conducted by the
SHAPES team at the University of Waterloo. Dr. Leatherdale is a Cancer
Care Ontario Research Chair in Population Studies. The Canadian Cancer
Society provided funding to develop SHAPES, the system used to collect
the PLAY-ON data. Erin P. Hobin is supported by a Canadian Institutes of
Health Research Canada Graduate Scholarship Doctoral Research Award.
Received: July 23, 2009 Accepted: July 6, 2010
REFERENCES
(1.) Booth FW. Fundamental questions about genes, inactivity, and
chronic diseases. Physiol Genomics 2007;28(2):146-57.
(2.) Canadian Fitness and Lifestyle Research Institute. Kids CAN
PLAY--2008 series, May 2009. Bulletin #1. Available at:
http://www.cflri.ca/eng/statistics/surveys/documents/Canada.pdf
(Accessed May 10, 2009).
(3.) Shields M, Tjepkema M. Regional differences in obesity. Health
Rep 2006;17(3):61-67.
(4.) Tremblay MS, Willms JD. Secular trends in the body mass index
of Canadian children. CMAJ 2000;163(11):1429-33.
(5.) Hill JO, Peters JC. Environmental contributions to the obesity
epidemic. Science 2000;280:1371-74.
(6.) Sallis JF, Baumann A, Pratt M. Environmental and policy
interventions to promote physical activity. Am JPrev Med 1998;15:379-97.
(7.) Sallis JF, Conway TL, Prochaska JJ, McKenzie TL, Marshall SJ,
Brown M. The association of school environments with youth physical
activity. Am J Public Health 2001;91(4):618-20.
(8.) Craig CL, Cameron C, Russell SJ, Beaulieu A. Increasing
physical activity: Supporting children's participation. Ottawa, ON:
Canadian Fitness and Lifestyle Research Institute, 2001.
(9.) Dobbins M, De Corby K, Robeson P, Husson H, Tirilis D.
School-based physical activity programs for promoting physical activity
and fitness in children and adolescents aged 6-18. Cochrane Database
Syst Rev 2009; Art. No.:CD007651.
(10.) Thomas H. Obesity prevention programs for children and youth:
Why are the results so modest? Health Educ Res 2006;21(6):783-95.
(11.) Ma X. Health outcomes of elementary school students in New
Brunswick. The education perspective. Eval Rev 2000;24(5):435-56.
(12.) Maes L, Lievens J. Can the school make a difference? A
multilevel analysis of adolescent risk and health behaviour. Soc Sci Med
2003;56(3):517-29.
(13.) Wong SL. Lessons learned from tobacco control: A multi-level
analysis of school characteristics and adolescent physical activity
[thesis]. Waterloo, ON: University of Waterloo, 2007.
(14.) Barnett T, O'Loughlin J, Gauvin L, Paradis G, Hanley J,
McGrath J, Lambert M. School opportunities and physical activity
frequency in nine year old children. Int J Public Health 2009;54:150-57.
(15.) Haug E, Torsheim T, Samdal O. Physical environmental
characteristics and individual interests as correlates of physical
activity in Norwegian secondary schools: The health behaviour in
school-aged children study. Int J Behav Nutr Phys Act 2008;5(47).
(16.) Nichol M, Pickett W, Janssen I. Associations between school
recreational environments and physical activity. J Sch Health
2009;79:247-54.
(17.) Physical and Health Education Canada (formerly known as the
Canadian Association for Health and Physical Education, Recreation, and
Dance). About quality daily physical education. Available at:
http://www.cahperd.ca/eng/ physicaleducation/about_qdpe.cfm (Accessed
April 1, 2009).
(18.) World Health Organization. A guide for population-based
approaches to increasing levels of physical activity. Implementation of
the WHO global strategy on diet, physical activity, and health. Geneva,
Switzerland: World Health Organization, 2007. Available at:
http://www.who.int/dietphysicalactivity/PA-promotionguide-2007.pdf
(Accessed February 12, 2010).
(19.) Ontario Ministry of Education. Policy/Program Memorandum No.
138, 2005. Available at: http://www.edu.gov.on.ca/eng/teachers/dpa.html
(Accessed March 10, 2009).
(20.) Leatherdale ST, Manske S, Faulkner G, Arbour-Nicitopoulos K,
Bredin C. A multi-level examination of school programs, policies and
resources associated with physical activity among elementary school
youth. Int J Behav Nutr Phys Act 2010;7:6. doi:10.1186/1479-5868-7-6.
(21.) Leatherdale ST, Manske S, Wong SL, Cameron R. Integrating
research, policy and practice in school-based physical activity
prevention programming: The School Health Action, Planning and
Evaluation System (SHAPES) Physical Activity Module. Health Promot Pract
2009;10(2):254-61.
(22.) Cameron R, Manske S, Brown KS, Jolin MA, Murnaghan D, Lovato
CY. Integrating public health policy, practice, evaluation,
surveillance, and research using local data collection and feedback
systems: The example of the School Health Action Planning and Evaluation
System (SHAPES). Am J Public Health 2007;97(4):648-54.
(23.) Wong SL, Leatherdale ST, Manske S. Reliability and validity
of a school-based physical activity questionnaire. Med Sci Sports Exerc
2006;38(9):1593-600.
(24.) Kroeker C, Manske S, Beyers J, Vaillancourt J, Murkin E,
Etches V, et al. Development of the School Health Environment Survey
(SHES). Centre for Behavioural Research and Program Evaluation (CBRPE),
August 2006.
(25.) Ontario Ministry of Education. Foundations for a Healthy
School, 2009. Available at: http://www.edu.gov.on.
ca/eng/healthyschools/foundations.html (Accessed July 23, 2009).
(26.) Veugelers PJ, Fitzgerald AL. Prevalence of and risk factors
for childhood overweight and obesity. CMAJ 2005;173(6):607-13.
(27.) McMurray RG, Ring KB, Treuth MS, Welk GJ, Pate RR, Schmitz
KH, et al. Comparison of two approaches to structured physical activity
surveys for adolescents. Med Sci Sports Exerc 2004;36:2135-43.
(28.) Shoemaker PJ. How to Build Social Science Theories. Thousand
Oaks, CA: Sage Publications, 2004;20-22.
(29.) Rose G. The Strategy of Preventive Medicine. New York, NY:
Oxford University Press, 1992.
(30.) Implementation of Grades 11 and 12 Physical Education/Health
Education: A Policy Document. Manitoba Education, Citizenship and Youth,
April 2007. Available at: http://www.edu.gov.mb.
ca/k12/docs/policy/imp_pehe/document.pdf (Accessed April 1, 2009).
Erin P. Hobin, MA, [1] Scott T. Leatherdale, PhD, [2] Steve R.
Manske, EdD, [3] Jennifer Robertson-Wilson, PhD [4]
Author Affiliations
[1.] Department of Health Studies and Gerontology, University of
Waterloo, Waterloo, ON
[2.] Scientist & Research Chair, Department of Population
Studies and Surveillance, Cancer Care Ontario, Toronto, ON
[3.] Senior Scientist, Propel Centre for Population Health Impact,
Lyle Hallman Institute, University of Waterloo, Waterloo, ON
[4.] Assistant Professor, Department of Kinesiology and Physical
Education, Wilfrid Laurier University, Waterloo, ON
Correspondence: Erin P. Hobin, Department of Health Studies and
Gerontology, University of Waterloo, 200 University Avenue West,
Waterloo, ON N2L 3G1, Tel: 519-888-4567, ext. 36317, Fax: 519-886-6424,
E-mail: ephobin@uwaterloo.ca
Conflict of Interest: None to declare.
Table 1. Descriptive Statistics for the Sample of Elementary School
Students Who Are Inactive, Moderately Active, and Highly Active
Moderately
Inactive Active Highly Active
Student Characteristic % (n) % (n) % (n)
Overall 16.4% (389) 67.2% (1599) 16.4% (391)
Sex
Male 15.5% (175) 64.9% (731) 19.6% (220)
Female 17.1% (214) 69.3% (868) 13.6% (171)
Grade
5 17.7% (105) 65.3% (388) 14.8% (88)
6 16.0% (102) 63.9% (407) 18.1% (115)
7 14.4% (93) 66.5% (428) 16.5% (106)
8 16.1% (89) 67.9% (376) 14.8% (82)
# of PE Classes per
School Week
<2 23.8% (133) 61.7% (345) 12.9% (72)
[greater than or 13.7% (256) 67.1% (1254) 17.1% (319)
equal to] 2
Table 2. Descriptive Statistics for the Sample
of Elementary Schools (N=30)
Total
School Characteristic % (N)
Offer Intramural PA Programs
Yes 83.3% (25)
No 16.7% (5)
Offer Interschool PA Programs
Yes 90.0% (27)
No 10.0% (3)
DPA Implementation Model
In addition to daily PE class 20.0% (6)
As part of daily PE class 10.0% (3)
Only on days with no PE class 70.0% (21)
Table 3. Odds Ratios for School- and Student-level Factors
Associated With Being Moderately Active or Highly
Active Among Youth in Grades 5 to 8 (Ontario,
Canada)
Adjusted Odds Ratio
Model 1
Moderately Active
vs. Low Active
n=1988
moderately
active
Student-level Characteristics
Sex
Male 1.00
Female 0.97 (0.76-1.22)
Grade
5 and 6 1.00
7 and 8 1.11 (0.88-1.41)
# of PE classes per week
<2 PE classes per school week 1.00
[greater than or equal to] >2 1.86 (1.43-2.40) ([double dagger])
PE classes per school week
School-level Characteristics
Offer Intramural PA Programs
No 1.00
Yes 0.85 (0.54-1.32)
Offer Interschool PA Programs
No 1.00
Yes 0.61 (0.48-0.77) ([double dagger])
DPA implementation model
Only on days with no PE class 1.00
As part of daily PE class 0.63 (0.39-1.00)
In addition to daily PE class 1.00 (0.39-1.02)
Random Variance 0.046
Adjusted Odds Ratio
Model 2
Highly Active
vs. Low Active
n=780
highly active
Student-level Characteristics
Sex
Male 1.00
Female 0.59 (0.43-0.81) **
Grade
5 and 6 1.00
7 and 8 0.99 (0.72-1.36)
# of PE classes per week
<2 PE classes per school week 1.00
[greater than or equal to] >2 2.12 (1.47-3.05) ([double dagger])
PE classes per school week
School-level Characteristics
Offer Intramural PA Programs
No 1.00
Yes 1.14 (0.66-1.96)
Offer Interschool PA Programs
No 1.00
Yes 0.26 (0.18-0.36) ([double dagger])
DPA implementation model
Only on days with no PE class Not included in model
As part of daily PE class as not significant in
In addition to daily PE class univariate analysis
Random Variance 0.053
CI = confidence interval
Controlling for school nesting and all other variables in the model
* p<0.05, ([dagger]) p<0.01, ([double dagger]) p<0.001
