You are on page 1of 10

Preventive Medicine 56 (2013) 152–161

Contents lists available at SciVerse ScienceDirect

Preventive Medicine
journal homepage: www.elsevier.com/locate/ypmed

Review

A systematic review and meta-analysis of interventions designed to increase
moderate-to-vigorous physical activity in school physical education lessons
Chris Lonsdale a,⁎, Richard R. Rosenkranz a, b, Louisa R. Peralta a, Andrew Bennie a,
Paul Fahey a, David R. Lubans c
a
School of Science and Health, University of Western Sydney Locked Bag 1797, Penrith 2751, Australia
b
Department of Human Nutrition, Kansas State University, 201 Justin Hall, Manhattan, KS 66506, USA
c
School of Education, University of Newcastle, University Drive, Callaghan NSW 2308, Australia

a r t i c l e i n f o a b s t r a c t

Available online 14 December 2012 Objectives. Physical education (PE) that allows students to engage in moderate-to-vigorous physical activ-
ity (MVPA) can play an important role in health promotion. Unfortunately, MVPA levels in PE lessons are
Keywords: often very low. In this review, we aimed to determine the effectiveness of interventions designed to increase
Motor activity the proportion of PE lesson time that students spend in MVPA.
Adolescent
Methods. In March 2012, we searched electronic databases for intervention studies that were conducted
Child
Students
in primary or secondary schools and measured the proportion of lesson time students spent in MVPA. We
Physical education and training assessed risk of bias, extracted data, and conducted meta-analyses to determine intervention effectiveness.
Results. From an initial pool of 12,124 non-duplicate records, 14 studies met the inclusion criteria. Students in
intervention conditions spent 24% more lesson time in MVPA compared with students in usual practice condi-
tions (standardized mean difference =0.62).
Conclusions. Given the small number of studies, moderate-to-high risk of bias, and the heterogeneity of
results, caution is warranted regarding the strength of available evidence. However, this review indicates
that interventions can increase the proportion of time students spend in MVPA during PE lessons. As
most children and adolescents participate in PE, these interventions could lead to substantial public health
benefits.
© 2012 Elsevier Inc. All rights reserved.

Contents

Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Eligibility criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Information sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Data collection process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Risk of bias of individual studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Summary measures and synthesis of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Risk of bias across studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Risk of bias within individual studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Outcome measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

⁎ Corresponding author. Fax: +61 2 9852 5416.
E-mail address: c.lonsdale@uws.edu.au (C. Lonsdale).

0091-7435/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ypmed.2012.12.004

. . . . 157 Risk of bias across studies . . . . . . . . . . . (2006) Cluster RCT (USA) 1 68 18 Students 68 18 5a Both (54% boys) McKenzie et al. . . 2011. . . . . . 2011). . . However. . .. .. . . . the proportion of lesson time during which students are engaged been determined. . . . . . Department of lessons may provide insufficient MVPA for students to benefit. . . 2012). . . . . . . . . . . . . Hardy et al. . . . . . . . . . . . 156 Quantitative synthesis . . . . . . . daily MVPA (Dobbins et al. . . . . . . . . . . . . Compared with their inactive counterparts. . . (1996) Cluster RCT (USA) 96 648a 400a Lessons 5352 3743 3 Both (52% boys) McKenzie et al. . . . . . . . (2008) Cluster RCT (USA) 36 215a 215a Lessons 1080a 1080a 6 to Girls 8 Young et al. . . .. . . or (b) a mixture of interventions sized. . . . . . . . . . . . . . . . . . . . . . . . . . . 2000. . . . Timperio et al. . . . . . . . . . . . 2010. 2000). . . . . 158 Summary of evidence . . . to our many children and adolescents are currently insufficient to promote knowledge. . . . . . . . . . . 2007. . . Sallis. . . . .S. quality PE could have far-reaching health implications for increase MVPA during PE lessons (Dudley et al. . . .e. . . . . . . the proportion of PE lesson time students spend in MVPA) has (Singh et al. . these are sufficiently active enjoy better physical health (U. . . 2006. . . . . . . . . (1997) Cluster RCT (USA) 7 70 33 Classes 1045a 493a 4 Both (53% boys) Scantling and Dugdale (1998) Cluster RCT (USA) 1 21 22 Students 21 22 9 Girls c Simons-Morton et al. . . . . . . . . . . . . . . 2006). . . . . no systematic review has been completed to determine the these benefits (Crawford. . . (2008) Cross-over design 1 45 Students 45 45 5 Both (56% girls) (UK) Sallis et al. . . . . . . . .. . . . . . . . . . . designed to increase total daily MVPA and those specifically designed to therefore. Unfortunately. . . the topic. . . . . . . Salmon et ventions designed to increase MVPA during PE. . Unfortunately. (1991) Cluster RCT (USA) 4 96 73 96c 73c 3 to Not reported 4 Strand and Anderson (1996) Cluster RCT (USA) 1 30 30 Students 30 30 7 Boys van Beurden et al. . 158 Strengths and limitations . . . . . . . . . than the 50% target that was physical activity (MVPA) can lead to a variety of benefits for children proposed by the U. . . . .. . and the central have examined: (a) the effect of school-based interventions on total role of physical education (PE) programs in this effort has been empha. .500a 12. . . . . . . . . no meta-analysis of PE-focused interventions has been conducted. . . . 2008). . . . . . .S. . . . to our knowl- nearly all youth (Crawford. . . . .. . . . . . .. . . . Table 1 Study design and sample characteristics. . . . . . . a Estimated value. . . 160 Funding . . . . . . . . . . these reviews al. . . . . . . Fairclough and Stratton (2005b) completed a narrative review on In response to this evidence.. . . and achieve higher academic results (i. the importance of schools in PA promo. . . . . . . . . . . . . . . . . . . . . . . (2007) Cluster RCT 16 190 190 Students 399 365 4 to Both (50% boys) (Belgium) (Accelerometer data) (Accelerometer data) 5 Verstraete et al.. . . . . . . . . . . . . and often far less. . . . . . . . . . . . . . . . . . . . . . . . C. . . . PE classes encompass virtually all members of an age cohort.S. . . . As a result. . . . . . . . . . . . . . 2011) and increasing active learning time in PE quality of life (Shoup et al.. . report more positive physical self. . . . . . . . . . . 2006). . . . 2000). . . . . . . . . . Health and Human Services. . youth who and the UK's Association for Physical Education (2008). . 160 Conclusions . . . . 2009. . students' MVPA levels edge. . . . . . . . . 158 Discussion . . . . the effectiveness of interventions on MVPA in PE has not Indeed. . . . Instead. . . . . . . . / Preventive Medicine 56 (2013) 152–161 153 Qualitative synthesis . . . . . . . . in PE lessons are often very low (Fairclough and Stratton. . .. Citation Study design Schools Intervention Control sample Data unit Total Total Year Gender (country) (n) sample size size represents students in students level (n) (n) intervention in control Fairclough and Stratton (2005a) Cluster RCT (UK) 1 12 (Heart rate 14 (Heart rate Students 15 18 7 Girls monitor data) monitor data) Fairclough and Stratton (2005a) Cluster RCT (UK) 1 5 (SOFIT data) 5 (SOFIT data) Lessons 15 18 7 Girls Ignico et al. . . the physical activity (PA) levels of been a component of many of these interventions. . . . van Sluijs et al. . . . 2009). 2004. . . 2005b. . 160 Conflict of interest statement . . . . . Kriemler et al. b Participants served as their own controls. . (2004) Cluster RCT (USA) 24 351 360 Lessons 12. . . . . . . . . . . . . . . perceive a better al. . Pate et al. . . . . . . effect of interventions aimed at increasing active learning during PE les- Department of Health and Human Services. . . . . . . . . . . . . . . . . . . 2010. U. . . . . . .. (2003) Quasi-experiment 18 117a 117a Lessons 523a 523a 3 to Both (53% boys) (Australia) 4 Verstraete et al. . . . . . 160 There is ample evidence that participating in moderate-to-vigorous in MVPA is typically less. . . . . . . . . . . Lonsdale et al. . . . .. . . . . School-based interventions can promote MVPA in youth (Dudley et concept and global self-esteem (Dishman et al. . . . . . 160 References . . . . . . . . . . . . . . . . . . . . (2007) Cluster RCT 16 19 (SOFIT data) 19 (SOFIT data) Lessons 399 365 4 to Both (50% boys) (Belgium) 5 Webber et al. . . . . . 2007). .. . . . As a result. . Department of Health and Human Services (2000) and adolescents. . . . . . . . . sons. . Other systematic reviews were not focused specifically on inter- tion has been highlighted (Cox et al. . c Unclear value. . . (2006) RCT (USA) 1 40 41 Lessons 116 105 9 Girls Note: RCT = Randomized Controlled Trial and SOFIT = System for Observing Fitness Instruction Time. . . . . . . . . . . . . . Also. . .500a 6 to Both (% not indicated) 8 Quinn and Strand (1995) Cluster RCT (USA) 1 29 31 Students 29 31 7 Boys b Rowlands et al. . . . . . . . 2009). .

(35 weeks) Strand and Anderson (1996) Fitness infusion Fitness Skill Play Integration Model: Time None stated Skill development and game (American football) for skill development and game play was play continued for entire reduced to allow for an aerobic fitness lesson. (various) sessions for teachers. Schools were recommended to schedule three 30 minute PE lessons each week. followed by remainder of class. (2007) Teaching strategies Portions of SPARK intervention: Guidelines Based on SPARK intervention Usual practice. (104 weeks) Quinn and Strand (1995) Fitness infusion Fitness Skill Play Integration Model: Time None stated Skill development and game (American football) for skill development and game play was play continued for entire reduced to allow for an aerobic fitness lesson. (104 weeks) Scantling and Dugdale (1998) Fitness infusion Fitness Skill Play Integration Model: None stated Attendance followed by skill (badminton) Fitness activities during attendance development game play for taking place at start off class. (4) teacher positioning. component in the last 10 min of lesson. and (c) on-site spread over 3 sessions per consultation with teachers. Citation Intervention type Intervention description (duration of Theoretical basis for intervention Control description (activities involved) intervention) Fairclough and Stratton Teaching strategies The teacher was made aware of study None stated Same lesson content. (various) encourage MVPA. (4 weeks) Rowlands et al. teach movement equipment only skills and be enjoyable. (various) to provide health-related physical education and increase MVPA during lessons. teacher training. (2008) Teaching strategies TAAG: PE-specific portions of the Operant Learning Theory. Organizational Change Theory. Intervention included: agreed to provide a (a) CATCH PE curriculum and materials. skill development and game play. emphasis on increasing Suggested strategies were: (1) MVPA. (5) active learning. Social Cognitive Usual practice. (78 weeks) Webber et al. (various) intervention included class management Theory.154 C. and (6) having fun. management and instruction tips. (5 weeks) Ignico et al. (4 weeks) van Beurden et al. (b) minimum of 90 min of PE. equipment and resources. attempts. (3) lesson pace. (1997) Teaching strategies SPARK: PE classes designed to promote high Social Learning Theory Schools provided with (various) levels of physical activity. (1991) Teaching strategies CAPE: 5 × 6-to-8-week units designed to Social Cognitive Theory Usual practice. jump rope and obstacle courses. organization of groups and use of space. and the provision of appropriate equipment and choices of physical activity. (4 weeks) Simons-Morton et al. (104 weeks) . (1996) Teaching strategies CATCH: Goal to provide MVPA during Social Learning Theory Usual practice. emphasis and the Diffusion of Innovation Model in a on the importance of engaging girls in Social–Ecologic Framework MVPA during class. (various) on supporting teachers and creating supportive environments and healthy school policies. (2008) Other (dance and Motiv8: Instructors from an external None stated Lessons taught by specialist soccer) agency taught PE lessons with goals of PE teacher employed by 20 min moderate and 10 min vigorous school. strategies. / Preventive Medicine 56 (2013) 152–161 Table 2 Intervention and control condition descriptions. (24 weeks) McKenzie et al. 10-minute aerobic fitness component in the last 10 min of lesson. assist teachers to promote MVPA through modified curricula. Schools (various) enjoyable PE lessons. activity during lesson. Sessions designed to create awareness. and enhanced man- agement and instructional skills. improved curricula. skill-building activities. including organization. Buddy system used to improve PE teaching and increase PA levels and fundamental movement skills mastery by matching pre-service teachers with generalist classroom teachers. (2) teaching approaches. (2003) Teaching strategies Move It Groove It: Intervention focused None stated Usual practice. Each unit included cardiovascular fitness activities such as dancing. component in the last 10 min of lesson. but no (2005a) (gymnastics) aim (to improve students' MVPA). (1 week) Sallis et al. Ecological Theorya Usual practice. Lonsdale et al. aerobic games. (2006) Fitness infusion Fitness infusion: Skill development with None stated Traditional skill (various) short bouts of MVPA between practice development. McKenzie et al. (2004) Teaching strategies M-SPAN: Professional development Social Learning Theory. (130 weeks) week. running. (22 weeks) Verstraete et al.

the theories. 2011). intervention delivery (Table 2). System for discussed and points of difference resolved. 2. in the report. and was the study adequately school*))) AND (LIMIT-TO(LANGUAGE. These criteria aligned with standards outlined in the Con- We also inspected the reference lists of all included primary source articles.. namely a deliberate attempt to 7.None stated for this aspect of Usual practice. badminton) would not have been included.e. or ap. where possible. Following extraction. / Preventive Medicine 56 (2013) 152–161 155 Table 2 (continued) Citation Intervention type Intervention description (duration of Theoretical basis for intervention Control description (activities involved) intervention) Young et al. such as age or gender)? exercise OR fitness OR “motor activity” OR “activity level*”)) AND (TITLE.g. a study between the two researchers. For example. When a criterion was not fulfilled. This measure could be objective (e. (35 weeks) Note: MVPA = Moderate-to-Vigorous Physical Activity and SOFIT = System for Observing Fitness Instruction Time. We included studies employing experimental (e. soccer vs. Finally.g. Cross-sectional each article. (2006) Teaching strategies MVPA during PE was promoted by teach. usual ementary) or secondary schools (i. (various) ing topics that were active in nature (e. comparing active learning time during ball games (e. and PsycINFO. A ‘risk of bias’ score for each study was calculated on an eight-point scale after two researchers (CL and DL) independently assigned a value of Search 0 (absent or inadequately described) or 1 (present and explicitly described) to each of the following questions: Our Scopus search terms and strategy are detailed below. outcome measure used to estimate tion of the intervention or the follow-up period. SOFIT.)) AND (TITLE-ABS-KEY(Intervention 3.. When data required for meta-analysis were not presented (e. and sion of the remaining articles and recommended each for inclusion or exclu- (b) conduct meta-analyses to determine the effectiveness of these sion. We employed 1.g. peer-reviewed articles (not abstracts) that reported primary data deviations not reported).. Were baseline values accounted for in the analyses? ABS-KEY(“Physical education” OR PE OR P. Were groups comparable at baseline on key characteristics (positive if baseline character- the same key words in our searches of the other three databases.e. OR experiment OR Training OR compar* OR Contrast* OR Condition)) AND random number generating algorithm)? (TITLE-ABS-KEY(Student* OR pupil OR learner OR child* OR adolesc* OR 4. Other implement a change to usual teaching practice in order to increase the pro- portion of PE lesson time spent in MVPA. Skills training was limited to that needed for competency.. istics were presented for the proportion of MVPA during PE class. plus one other demo- ((TITLE-ABS-KEY(“Physical activit*” OR “active learning” OR movement OR graphic detail.“j”)). We set no limits calculations and/or assumptions possible from the information provided on the earliest publication date. based on and had been published online or in print by March 24. criteria. 6. these were estimated. Not a full-text peer-reviewed article containing primary data Studies must have tested an intervention. From these searches. We included English.e. 2010). because both types Data collection process of activities are considered part of usual practice in PE. Our objectives were to: (a) describe the nature of the interven. consistent with procedures outlined by the Cochrane Collaboration (Higgins and Green. Observing Fitness Instruction Time) (McKenzie.. No intervention designed to increase MVPA compared with a control (e. soccer) versus net sports (e. Sluijs et al. for exclusion. Incorrect age (non-primary/secondary school sample) Included studies sampled students from PE classes in primary schools (i. Pre-school or post-secondary practice condition) institution samples were excluded. two researchers (CL and DL) independently read the full ver- proaches researchers have used to design their interventions).E. (Table 1). solidated Standards of Reporting Trials (CONSORT) statement (Schulz et al. powered to detect MVPA changes during PE lessons? ..g. A biostatistician (PF) made all estimations. rather than proficiency. duration of follow-up. PubMed. The order in which criteria were considered was as follows: 1. and the Studies needed to include a measure of the proportion of PE lesson time mean (and SD) proportion of lesson time spent in MVPA in the intervention spent in MVPA.g. 2012. a Personal communication from T.. No objective or observational measure of MVPA 3. including: Scopus. Objectives Study selection After exporting records into Endnote reference manager software and de- Our aim was to systematically review the evidence related to inter- leting duplicate records. McKenzie on June 30th. We assessed risk of bias for each study using criteria adapted from van ined their reference lists for primary sources that met our inclusion criteria. but not an intervention.. Data items included characteristics of study design and sampling and cohort designs were excluded. Next. remaining studies and excluded those that did not meet our inclusion tions that have been undertaken (i.. standard language. Duplicate data from another article Methods 2.g.g. recording the specific criterion not fulfilled for studies recommended interventions. we identified review articles and exam. the proportion of lesson time spent in MVPA. high school). the article was not consid- ered any further. (2007). No proportion of time in MVPA reported (or data from which proportions could be Eligibility criteria calculated) 4. any non-identical data were systematic direct observation by an independent rater (e. No limitations were set regarding the dura. strategies. and playing games in small groups. we read abstracts from the lessons..“English”)) AND (LIMIT-TO(SRCTYPE.g. breaking skills training into small-group activities. SPORTDiscus. Were randomization procedures clearly described and adequately carried out (e. Studies focusing on a comparison Disagreements regarding criteria fulfillment were resolved by discussion of conditions. 2012.. Lonsdale et al. softball). el- 5.g. intervention.. Information sources Risk of bias of individual studies We searched four databases. accelerometer) or involve and control conditions. we removed studies whose title clearly indicated the ventions designed to increase active learning time during school PE study did not meet the inclusion criteria. were excluded. 2009). randomized controlled Two members of the research team independently extracted data from trials and cross-over designs) and quasi-experimental methods. C.. Did the authors report a power calculation.

This statistic provides Complete results of risk of bias assessments can be viewed in an indication of the extent to which results are consistent across trials. (2007) simultaneously employed SOFIT and accelerometry. 2007).g. with eight studies involving assessments completed Study characteristics at least 35 weeks after baseline. Strand and Anderson. practice teaching. while in Fairclough and Stratton's (2005a) study. This included an overall analysis to determine the ef- the intervention. Lonsdale et al..50.124 non-duplicate records.. In some studies. Two interventions were tested in grade 9 students intervention condition. a significant experimental design (Van Beurden et al.2% agreement on risk of bias ratings (110 of values of . If smaller studies were found to have larger effects. fect sizes are not directly comparable. as well as sub-group analyses to de. while Study selection results are detailed in Fig.. Ten studies were conducted in the USA. Verstraete et al. respectively. items after discussion. Five studies included analyses that accounted Eleven studies employed a cluster randomized controlled design. 1998). 2008. We examined a funnel plot to assess risk of publication bias (Egger et al.e. In 10 of from n=18 (Fairclough and Stratton. Ignico et al.001). 2008). I 2 Table 3..g. 2004). 2008). while four conditions? studies focused on girls only (Fairclough and Stratton. 1995. studies. 2007).g. with a median of n=106. Were participant dropout rates described.. Among studies not all participants provided MVPA data.. n=15 (Fairclough and Stratton.008) and SOFIT (p b . In other (Young et al. for baseline MVPA levels (see Table 3 for details).... Did the study include measures of MVPA known to produce reliable and valid (i. Full details regarding sample that employed two measures of MVPA. Study characteristics can be viewed in Table 1.05)... screening of titles and abstracts lead to rate monitors. basketball) with vigorous Difference of means (proportion of lesson time spent in MVPA) be- fitness activities (e. and Summary measures and synthesis of results (b) ‘fitness infusion’ (n= 4 studies). MVPA blinded to allocation)? Complete details regarding intervention and control conditions can Studies with a score of zero to two were considered to have a high risk of be viewed in Table 2. Other volved comparisons at follow-up. 2003) and a randomized finding indicated a between-group difference in the change in the pro- controlled trial with treatment group allocation at the student level portion of lesson spent in MVPA (e. 2008. All studies found a higher proportion of active learning time in 2004).500 (McKenzie et al. Verstraete et al. and not more than 20% for studies with follow-up of six months or shorter. Outcome measures As shown in Table 4. / Preventive Medicine 56 (2013) 152–161 5. We also conducted sub-group analyses to investigate the effects of age. these investigations. 2005a. Scantling 8. In all studies. Verstraete et al. and intervention duration. 2005a) to n=12. Risk of bias within individual studies group differences in the proportion of lesson time spent in MVPA was gen- erated. significance levels and ef- studies included a cross-over design (Rowlands et al. Two studies were conducted exclusively with 7. 2011). .. 2006). In criteria. teachers learned strategies to encourage MVPA through effective activ- ity selection. there would be evidence that publication bias was present in the meta-analysis.. The number of schools involved in each study ranged from n = 1 (in seven studies) All data related to the proportion of lesson time spent in MVPA can to n = 96 (McKenzie et al. Were outcome assessments blinded (positive if those responsible for assessing and Dugdale. and high. We identified two main types of interventions: bias. running. a forest plot of between..500 (McKenzie et al... 2005a. while others only in- with randomization occurring at the school or class level. From these. the article)? Half of the studies were conducted with a mixture of boys and girls. both (i. Nine studies included direct observation methods. 1998. From an initial pool of Fairclough and Stratton (2005a) measured MVPA using SOFIT and heart 12. Publication dates ranged from 1991 to 2008. There was 98. gender. the difference was statistically significant (p b . with a median of n=89.. For each analysis.g. Control condition samples ranged the intervention group compared with the control group. tween the intervention and control conditions at follow-up was the sum- 2008) did not provide enough information to determine the nature of mary measure. two studies used accelerometers to measure the Results proportion of lesson time spent in MVPA (Rowlands et al. while five were implemented heart rate monitor (p = .05) scores favored the in grades 6–8. in which teachers supplemented students' participation in sport activities (e.. 2006) to n= the 12 studies that employed a single measurement instrument. One study (Rowlands et al. 1. follow-up assessments were conducted one to 156 weeks later. Webber et al. moderate.156 C. termine the effectiveness of different types of interventions.25. 1997). while one study had low risk Risk of bias across studies of bias (Webber et al. We conducted the meta-analyses in Stata Version 12 (StataCorp LP.e. 14 articles met the inclusion Ten of the 14 studies included baseline MVPA data collection. Heart rate monitors were employed in five Study selection studies. the control condition involved usual fectiveness of interventions generally. Five studies were rated as having high risk of bias. secondary/high school) (Scantling and Dugdale. jumping). 6.. Was timing of measurements comparable between intervention and control boys (Quinn and Strand. 2003). 2006). one in Belgium and one in Australia.. 112 items) and raters reached agreement on these two discrepant tively (Higgins et al. Young scores (positive if reliability and validity evidence was reported or referred to in et al. elementary school grades 3–5). Eight studies had moderate risk. be viewed in Table 4. and . respec. 109 full-text articles read. Verstraete et al. in which have moderate and low risk of bias. 2006). Studies with scores of three to five and six to eight were considered to (a) ‘teaching strategies’ with an MVPA focus (n= 9 studies). In some studies. (2007) size can be viewed in Table 1. and an I 2 consistency statistic was calculated. 1998. reported a significant effect only on the SOFIT measure of MVPA Seven interventions were conducted during primary school years (p b .. 1996). and instruction.75 were considered low. .. Qualitative synthesis two in the UK. Scantling and Dugdale.. As a result. a significant finding indicated a between-group difference at The number of participants in the intervention conditions ranged from follow-up only (e. a quasi. 12. 1996). and 30% for studies with follow-up of more than six and the gender distribution in these studies was generally equal months? (range=46%–56% girls). class organization and management. Young et al.

(2006) 1 1 0 0 1 1 1 0 5 Note: 1. We conducted meta-analyses on data extracted from 13 of the 14 For studies in which two measures of MVPA were obtained (SOFIT studies (see Fig. (2003) 0 1 0 0 1 0 0 0 2 Verstraete et al. Table 3 Risk of bias results for included studies. plus an objective measurement device) (Fairclough and Stratton. Did the study include measures of physical activity known to produce reliable and valid scores (positive if reliability and validity evidence was reported or referred to in the study)? 6. 8.124 records after duplicates removed Screening 912 abstracts screened 803 excluded 109 full-text articles assessed Eligibility 95 full-text articles excluded 6 . (2006) 0 0 0 0 1 0 1 0 2 McKenzie et al. 9. Were groups comparable at baseline on key characteristics (positive if baseline characteristics were presented for the proportion of MVPA during PE class. Data from one study (Simons-Morton et al.. such as age or gender)? 2. 6–8=low risk of bias). Citation 1. (1996) 1 1 0 0 1 1 1 0 5 McKenzie et al. Lonsdale et al. (2007) 1 1 0 0 1 1 1 0 5 Webber et al.g. 2012. C. Was timing of measurements comparable between intervention and control conditions? 8. plus one other demo- graphic detail. Were outcome assessments blinded (positive if those responsible for assessing MVPA were blinded to allocation)? 9. . Were participant dropout rates described. 2). 5.non-primary/secondary school participants 41 .. Total Risk of Bias Score (0–2=high risk of bias. (2004) 1 0 0 0 1 1 1 0 4 Quinn and Strand (1995) 0 0 0 0 1 0 1 0 2 Rowlands et al. Random 4. Valid 6. (2008) 0 0 0 0 1 0 1 0 2 Sallis et al.no report of % lesson time in MVPA 2 . and was the study adequately powered to detect MVPA changes during PE lessons? 5. Baseline in 3. 1.773 records identified 10 records identified through other Identification through database searching sources 12.duplicate data 30 . comparable analysis assignment Power assessment Dropout Timing Blind Total Fairclough and Stratton (2005a) 1 1 0 0 1 0 0 0 3 Ignico et al. Did the authors report a power calculation. 7. Were baseline values accounted for in the analyses? 3. Fig. / Preventive Medicine 56 (2013) 152–161 157 12.no objective or observational MVPA measure 12 . 3–5=moderate risk of bias. Quantitative synthesis 1991) could not be included because most results were presented graphically and accurate estimates of results could not be made. and not more than 20% for studies with follow-up of six months or shorter.no peer review or primary data Included 14 articles in qualitative synthesis 13 studies included in quantitative synthesis (meta-analysis) Note: The final electronic database search was conducted on March 24. (1991) 0 0 0 0 1 1 1 0 3 Strand and Anderson (1996) 0 0 0 0 1 0 1 0 2 van Beurden et al. Baseline 2. and 30% for studies with follow-up of more than six months? 7. (1997) 1 0 0 0 1 1 1 0 4 Scantling and Dugdale (1998) 1 0 0 0 1 0 1 0 3 Simons-Morton et al.no intervention 4 . Were randomization procedures clearly described and adequately carried out (e. random number generating algorithm)? 4. Flow diagram. (2008) 1 0 0 1 1 1 1 1 6 Young et al.

none of these factors was 43.9) strategies (n = 40) (n = 41) Note: MVPA = Moderate-to-Vigorous Physical Activity and SOFIT = System for Observing Fitness Instruction Time.2 (29.7) 35.54% of lesson time spent in lessons. Effective intervention strategies included teacher fusion intervention condition spent more time in MVPA compared with professional learning focusing on class organization.001). in the intervention groups compared with the control condition (standardized mean difference = 0. p b .40%. 95% CI=0.6a (9.8 (17. 95% CI=4. 2007).1) strategies (n = 70) (n = 33) Scantling and Fitness Heart rate monitor ≥155 bpm No baseline t-Test (b0.33%–14. we included data from the objective results of these four studies was high (I2 =92. (2003) Teaching SOFIT (lesson) ≥Walking 78 weeks Hierarchical logistic 39. Examination of the funnel plot (Fig.001) (n = 19) (n = 19) Webber et al. (2007) Teaching SOFIT (lesson) ≥Walking 85 weeks Linear mixed models 56.e.001) 14.2) 13.1 (18. (2) ‘fitness in- fusion strategies’ (k=4).7%. However. Heterogeneity in the activity (i.0016) (n = 648) (n = 400) McKenzie et al. 2005a. 2008).9) 34. Lonsdale et al. (1991) Teaching Children's physical activity Unclear 104 weeks Confidence intervals as 36. Summary of evidence trols (absolute difference=6. of interventions based on: (1) ‘teaching strategies’ (k=8). However.5 (2.008) 40. mean MVPA in the control groups was 26.6) infusion (student) measure (n = 29) (n = 31) Rowlands et al. CI = 6.5 (8.5 (19.2) 28. The weighted mean across the control groups age. 95% CI=0.05) 49. The lated to interventions designed to increase students' MVPA within PE weighted mean of the control groups was 44.9) 30.62.84).158 C.35.2) 48. Overall.. management and controls (absolute difference=16.025) Young et al.8 (35.. interventions were associated with 24% more active MVPA and.2 (9. p=.2 (25.3) 46.2a (. pb . and intervention duration.05) 42. Verstraete et al.37% (95% spent in MVPA in the intervention condition relative to controls.0 (19.7) 24. 95% CI=5. 95% CI = 0.7) 42.39–0.0) strategies (0.05) 47..8 (12.9 (21.6 (0.25–2. (2008) Teaching SOFIT (lesson) ≥Walking 104 weeks Mixed-model 42.4) 34.39) (n = 190) (n = 190) Verstraete et al. contact the first author. and (3) one ‘other’ study whose intervention Discussion content was not clearly described (Rowlands et al. (2004) Teaching SOFIT (lesson) ≥Walking 52 weeks Randomized regression 53.8%.20–0.1) (student) (n = 19) (n = 19) Sallis et al. The estimated difference of appeared to moderate intervention effects.001) 46.7 (25.3%.25%–27. fitness infusion).5 (15.047) 18.2) 0. gender.15%. 3) revealed no evidence of We also conducted subgroup analyses to determine the effectiveness publication bias in the meta-analysis. (2007) Teaching Accelerometer ≥3 METs 85 weeks Linear mixed models 67.9 (23. For complete details of these 10. (1996) Teaching SOFIT (lesson) ≥Walking 52 weeks Mixed-model ANOVA 51.5) strategies (0.04)a 7.5 (23. a Estimated value.09) (n = 117) (n = 117) Verstraete et al. (2008) Other Accelerometer ≥192 cpm 1 week ANOVA (b0.001). The weighted measurement device only. time spent in MVPA).55%.35.1) 38.15%–8.58) 60. the students in intervention condition spent 14% learning time compared with usual practice (10% more of total lesson more lesson time in MVPA relative to controls. Heterogeneity in the results The aim of our study was to systematically review the evidence re- was low to moderate across these eight studies (I2 =40.27%.2 (18.5 (4.0)a strategies observation form (student) graphs (n = 96) (n = 73) Strand and Anderson (1996) Fitness Heart rate monitor (student) ≥156 bpm No baseline t-Test (>0. (2006) Teaching SOFIT (lesson) ≥Walking 35 weeks ANCOVA (b0. standard. on average. Risk of bias across studies it must be noted that there was a high degree of heterogeneity in the results across the 13 studies (I 2 = 86. Finally.2) (2005a) (n = 5) (n = 5) Ignico et al.9) 39.37% of lesson time corresponds to 24% more active learning time analyses (including forest plots).11). / Preventive Medicine 56 (2013) 152–161 Table 4 Intervention results relating to the proportion of lesson time spent in MVPA.3 (30.02) (n = 351) (n = 360) Quinn and Strand (1995) Fitness Heart rate monitor ≥156 bpm No baseline t-Test (b0.001) 54.3 (10.0 (12.8) (2005a) strategies (student) heart rate (n = 12) (n = 14) reserve Fairclough and Stratton SOFIT (lesson) ≥Walking 5 weeks t-Test (0.2) 41. Students in the fitness in. There was considerable heterogeneity in .3 (29.7) infusion (student) measure (n = 68) (n = 18) McKenzie et al.4 (12. (2006) Fitness Heart rate monitor ≥150 bpm No baseline ANOVA (b0.45% of lesson time in MVPA.6 (17.9) Dugdale (1998) infusion (student) measure (n = 21) (n = 22) Simons-Morton et al.9) infusion measure (n = 30) (n = 30) van Beurden et al. Citation Type of Outcome measure (data Criterion for Follow-up Type of analysis Proportion of Lesson time in intervention level) MVPA period reported (p-value) MVPA at follow-up after baseline Intervention Control mean % (SD) (n) mean % (SD) (n) Fairclough and Stratton Teaching Heart rate monitor ≥50% of estimated 5 weeks ANCOVA (0.05%.41%) of lesson time spent in MVPA in favor of the inter.6 (6) strategies models (0.9 (32.50).27) strategies (student) (0. Students in the teaching strategies intervention condition were more active than con.1) strategies regression (n = 215) (n = 215) (0. instruction.001) 50. and supplementing usual PE lessons with high-intensity ized mean difference=1. (1997) Teaching SOFIT (lesson) ≥Walking 104 weeks ANOVA (b0.3 (9. we conducted sub-group analyses to investigate the effects of ventions over controls. suggesting 61% more time The meta-analysis indicated an absolute difference of 10.45).4 (10.0 (14. standardized mean difference=0.9) strategies regression (0.

lack of access to facilities.g. a recent high-quality primary the spot.40 (3.g..40 (3.38 Subtotal (I-squared = 89. Considering the physical activity decline typically observed should be interpreted with caution. Therefore.49 (3.10 (-6. Since there are so few studies.. 9.05) 30. 27.110) 6. 2004). the bene.01.. p = 0.58 (5. For learning time in PE should be a public health priority. our findings increased. Overall (I-squared = 88. Social Cognitive Theory. and attractive sedentary alternatives). / Preventive Medicine 56 (2013) 152–161 159 Study % ID ES (95% CI) Weight Teaching Strategy Interventions Young et al (2006) 16. 1998.58 Quinn and Strand (1995) 15. there is a need and opportunity to ents and teachers attempting to increase children's physical activity out- evaluate the effects of interventions to increase active learning time in side of the school setting (e.69 (-6. Forest plot. explaining students' behavior.. 8. 2. Fitness infusion typically involves year. This is an appealing strategy school obesity prevention intervention involving 45 min of daily PE for increasing active learning time because it requires minimal organi- significantly increased children's within-school physical activity by zation and creativity from teachers. These benefits would be further amplified in schools that devote teachers incorporating high-intensity activity (e.54) 9.05.12.. 1995.. duration. spending 20 min 2006. Theory of Planned Be- havior.58.87. the intervention did not tainability of this type of intervention is unknown. 2011) is surprising.g.. the fitness infusion interventions (Ignico et al. 1996) were more effective in increasing MVPA than would represent an additional 6.72) 8.8%. con- sidering the importance placed on professional learning in the general education literature (Avalos.15 (5.44 Subtotal (I-squared = 40.40) 62.46 . the lack of focus on teacher professional learning in many school-based PA interventions (Kriemler et al. For example.15. Fitness Infusion Interventions Ignico et al (2006) 30.33 Verstraete et al (2007) 6.04 . and sample size.72.33.41) 8.12) 8. 1997) interventions. Quinn and Strand. intervention studies included in our review only measured MVPA dur- By including strategies to maximize MVPA within PE lessons.35. 2008.90 (-3. In addition to the above-mentioned theories.04 Subtotal (I-squared = .) 12.72 Scantling and Dugdale (1998) 14.000) 10..61 Sallis et al (1997) 3. 35.16 McKenzie et al (1996) 9. interventions that include teacher professional learning to improve lesson preparation and management appear to have potential long-term benefits for teachers and students (McKenzie et al. yet during adolescence (Nader et al. 9.000) 16.15 (9. 2004).10 Fairclough and Stratton (2005) 11. 14.25) 8. 2011)..g. the long-term sus- 11 min/day (Kriemler et al. which were guided by Social Learning Theory (Bandura.93) 7.50 . p = 0. Notable exceptions were the CATCH (McKenzie et al. 11. Strand extra of lesson time in MVPA every 10 days (10% of total lesson time) and Anderson.25. fits of these types of school-based interventions may be further In contrast.37. As such. running on greater lesson time to PE.22. ing lessons that took place during the intervention (i.61 McKenzie et al (2004) 4.92) 6. example. Transtheoretical Model) and that target the hypothesized mediators of behavior change are more successful in changing behavior than atheoretical ones (Lubans et al. 13.. increasing active The public health implications of these findings are substantial. where state law mandates that students in Not surprisingly. unsafe high-quality cluster randomized controlled trials. Less than half of the studies included in our review cited a relevant theoretical framework to explain students' behavior.15) 4.61. as fitness infusion specifically target improvements in active learning time in PE lessons. 1977) and M-SPAN. 5. jumping. p = 0. in California.e. 20. 26. 12..99. neighborhoods..37 (6. 24. 13. Scantling and Dugdale. which was also underpinned by Ecological Theory (Bronfenbrenner. and affect (e.17 van Beurden et al (2003) 4. Lonsdale et al. terms of study design. 27.15 (24.76) 9. 20.93) 7. 1979). Funnel plot. Grades 1–6 receive 200 min of PE every 10 days. p = .62. Other Interventions Rowlands et al (2008) 12. However. 2010).50 (6. C..39) 6.81 Strand and Anderson (1996) 3. cognition. Evidence suggests that interventions developed in reference to a theory of behavior change (e.%.90 (-1.04 Webber et al (2006) 3.3%. star jumps) into usual PE lessons. 3. 2008) and the challenges faced by par- their results are generally positive.40 (5.41) 100.7 h of MVPA across a 40-week school the teaching strategies interventions. no follow-up). researchers seeking to increase active learning time in PE may also wish to consider motivational theories that have proved useful in Fig.87.63) 6. 1996) and SPARK (Sallis et al.00 -10 0 10 20 30 40 Absolute difference in % MVPA Fig.51. emotion) in .2%.81) 6.65 (3. Unfortunately. Michie and Abraham.79) 10.50 (-2.27 (4. 19.

cern with many studies in this review was the lack of intervention fidel- Deci. we offer the following recommendations. and teacher perspectives. 2011.. 2008. The design and reporting Grants Scheme (Science.. R. 2010. Educ. 1977. Despite these The authors declare no conflict of interest associated with this study. Conclusions pact on PE students' outcomes. 2009) and studies were assessed for risk of bias using criteria adapted Conflict of interest statement from the CONSORT statement (Schulz et al. Health 24. Bronfenbrenner. tactics and objectives of various quality trials are needed to determine how best to promote MVPA in games. ized controlled trials. few have reported their effects on the proportion of PE lessons spent in MVPA. Morb. Sanchez-Vaznaugh et al. data. relevant feedback. as well as those aspects that might be pression symptoms among adolescent girls. While this approach may tions should target and assess effects on hypothesized psycho-social assist students in meeting their daily PA requirements. 2006). 1985. 2011. Sport 13.. U. of all confounding factors.S. 1991. K.M. The Future of Sport in Australia. Med. Second. abili. including self-reported effort during lessons (Taylor and Lonsdale. 2012. ters. Indeed. SOFIT. Teach. Exerc.160 C. Sallis et al. Wkly. Profes- Physical education is a valuable source of physical activity for chil. M. 34. Experimentally based. 2012). Interventions that enhance motivation to. Our review suggests that PE-based interventions can increase stu- wards PE and increase student MVPA during lessons may also increase dents' MVPA during lessons by about 24% compared with usual practice. tional studies emerge. 46–52.. Canberra.. 260–267. important role is to provide students with the knowledge. Archer. as well as the rules. Ryan. the risk of bias was moderate or high in all but one of the Cox. Sci.. 2012). Nicholls. Harvard University Press. skills. While these measures produce MVPA es. For these reasons. C. high levels of active learning time need to PE lessons. Hagger.L. 2011). over ten years. R.. G. Psychol. motivational climates that tions are often required to assess the intervention effects with adequate emphasize effort and personal improvement and provide students with statistical power.. 2010). self-assessment. 2008). to our knowledge. type of accelerometer and cut-points used) will allow for more ity instruction. As addi. Psychol. evidence suggests that different measures will produce dif. have been validated against criterion measures (Rowe et al. 2006. and opportunity to practice (Gallahue meaningful comparisons of intervention effects across studies.. N. Rep. observed in school-based studies. these interven- around a field for the duration of the lesson. healthy eating and physical activity.K. PE is an appropriate setting for learn... 1988.. Also. 2008. teacher-focused intervention to help physical education teachers be more autono- count for potential confounding factors in meta-analyses. First. my supportive toward their students. Furthermore. Effects of an intervention based on self-deter- ferent estimates (McClain et al. Educ. and the composition of groups) have a positive im.A... Intrinsic Motivation and Self-determination in Human Be- ity assessment. 80. timing. sional learning focused on teacher pedagogy and behavior offers consid- dren and adolescents (Tudor-Locke et al.P. 10–20. Teacher professional development in teaching and teacher education a variety of measurement tools to assess MVPA (e. and reflection. J. the internal validity of these studies is havior. Importantly. Teach. Crawford. 2011. Ames. 2008). a PE teacher controlled trials to determine the most effective strategies and theoret- could achieve 100% MVPA in their PE lessons by making students run ical frameworks for increasing MVPA in PE.S. a large number of schools and observa- ences in PE. 2009.. Importantly. Reeve. there are some limitations that should be noted. I. 2010). and heart rate monitors). Due to the considerable intra-class correlation coefficients typically tional climate’ created by the PE teacher can influence students' experi. select task difficulty. measures could be more or less sensitive to change in MVPA. Vallerand. Plenum Press. Educ. New York. School health guidelines to promote et al. it was not possible to meaningfully test the potential effects egies and motivation processes. J.g... J. Educ. 2008) and it is possible that different mination theory on self-reported leisure-time physical activity participation. accelerome. most of the studies were cluster random- al. time of physical activity. Cheon. First. among youth. et al. 2006). 1994) and and Design.S.. Bandura. mary.. 2008) and (Cheon and this increase could have a substantial positive influence on the total et al. Third. Cambridge.. J. Physical self-concept and self-esteem of behavior change.. Health position paper. but perhaps its most erable potential for increasing physical activity in youth.. Deci et improved in the future. rental. Association for Physical Education. D.. Hales. Pfeiffer. 29–48. 2012.. vention strategies..g. including details of professional learning for teachers. longitudinally designed.g. studies in this review included Avalos.M. Responsibility for children's physical activity: pa- studies included in our review (Webber et al. Despite the multitude of school- References based physical activity interventions (Kriemler et al.D. In sum- and Ozmun. Chatzisarantis.. Motivation and education: the self-determination the extent to which the intervention impacted on theoretical mediators perspective. 325–346. 1979. the existing evidence indicates that interventions can increase ing self-management strategies (e. Matters 3. D. Ryan. ties. Achievement goals in the classroom: students' learning strat- As a result.. 1988. Dishman. Sport. J. 1989) have been employed to explain how the ‘motiva. but higher monitoring). Social Learning Theory.. Duda. 1985. only 14 studies met our inclusion criteria... / Preventive Medicine 56 (2013) 152–161 PE. Psychol.. M. child. . According to these theories. NJ.. it will be important to update this review and ac.L. 1–76. Commonwealth of Australia... and the proportion of PE lesson time students spend in MVPA. S. teacher practice. 365–396. Mortal. and to determine the most effective and sustainable inter- be balanced with opportunities for instruction. In particular. 25. 26. strengths. G. E. 8–12.. R. there was a lack of clarity concerning Deci. 27. As a result.. Based on our review. PE provides children with oppor. Skill mastery requires qual. The Ecology of Human Development: Experiments by Nature timates that are correlated with each other (McKenzie et al. somewhat difficult to discern. and confidence to be active both now and throughout their lifetime there is a need for additional adequately powered cluster randomized (Sallis et al. Schofield. Finally. E.J. Sun Centers for Disease Control & Prevention. For example.L. Self-Determination Theory (Deci and Ryan. Engineering and Medicine). Englewood Cliffs. feedback. 2010). Moon. gies. Prentice Hall. more transparent reporting of intervention strate- that this type of lesson will engage students and prepare them for a life. Technology. 60. it is unlikely mediators. no Schools are ideal settings for the promotion of physical activity study has investigated the impact of an intervention based on Self. (e. 396–407. opportunities to demonstrate leadership and make decisions (e. Kolt. Strengths and limitations Funding This is the first systematic review and meta-analysis of interventions This study was funded by the University of Western Sydney Research designed to increase MVPA within PE lessons. Psychol. R. Importantly. amount of physical activity children and adolescents accumulate. of this review were guided by the PRISMA statement (Liberati et al. Lonsdale et al. is needed. MA. Phys. 1991) and Achievement Goal Theory (Ames and Archer. 1997. B. However. J. An additional con.H... consistency in physical activity assessment ious games and sports (Lubans et al.. Student activity levels in PE are likely dependent upon teacher behavior tunities to develop competence in a range of fundamental and and there is a need to improve our understanding of how to improve specialized movement skills that are necessary for participation in var.. This omission limits identification of the effective mediate cross-sectional relations of physical activity and sport participation with de- components of the intervention. A.. and PE is the primary vehicle associated with this objective Determination Theory or Achievement Goal Theory on students' active in schools (Centers for Disease Control and Prevention.g. their leisure time activity (Chatzisarantis and Hagger. goal setting. learning time during lessons. Health Psychol.. 2012). but few adjusted for the clustered nature of the 1996.

J. Effects of a life skills inter- Quinn.. Nader.G.F. 239–257.P. 703....A. C.. Improving health-enhancing physical activity in girls' of a 2-year physical education program (SPARK) on physical activity and fitness in physical education.. L.. J. P. 52.J. J. Rev.L.. 166. J. A.E. Robinson. K. Taylor. J. 38. H... Conway. A review of mediators of behavior in interventions StataCorp LP. Rosengard. Tudor-Locke. Biddle.O.P. Educ. Van der Mars. ucation and recess..F. L.M. S. Sanchez..T.. vigorous physical activity from ages 9 to 15 years. ventions: explanation and elaboration. elementary school students.R.. J. Can Michie. Rev.G.. 2011.F.C. Van Beurden.. Department of Health and Human Services. lines for reporting parallel group randomised trials.. Life Res. J. 315.J. 629–634. et al. G. 655–673. Early Child Dev..G. Egerter. D. Corson.. Salmon. Foster. Nursing. De Bourdeaudhuij. 2007. 20. Sports Exerc. Lonsdale. Med. 173–184.L. I. U.. L. J. Prev.. Morgan. Ignico. Educ. Young.. 63.A. Prev.. The effects of two instructional formats on the heart Higgins. 2008. 407–412. 2008. Baek. Res... Duda...F. S. Med. A. 2007. Prochaska. T. San Diego State University. S... Sun. B.F. Effectiveness of interventions to pro- Nicholls. T. Minder.. Beets. 295–305. J. L. Middlebrooke. S.P. Smith... C. 1382–1388. King. N. D. trials. Schindler.. Bradley. M. J. Erwin. S. tiveness of a two-year health-related physical education intervention in elementa- entific statement from the American Heart Association Council on Nutrition. 1991.. ry school physical education. Rosas. Cambridge. 2010. psychological need satisfaction. Eston.G. Beighle..0.J. Cosgrove. randomised controlled trial.L. R. J.. School physical education: ef. Care 176. Sci. Tirilis. 2008... 2010. 1996.L. 2000. The PRISMA statement for reporting Physical activity and performance at school: a systematic review of the literature systematic reviews and meta-analyses of studies that evaluate health care inter. Uijtdewilligen.. J.N.C.M. education's role in public health: steps forward and backward over 20 years and Fairclough.S. 1328–1334.. D. ical activity participation among children and adolescents. S. I..A. pedometer-determined physical activity during the segmented school day. L. 181. 834–840. 2011. A.. Estabrooks.. Med. 1997. A. M. Sport Exerc. R. Med.1. D. Public Health 81. L.G. G. 20–29. Forthofer.L. Sport 7. Am. Schneider. skill achievement and student attitude. M. Phys.R.J. R. Cochrane Database Syst.W. 17.. 32. C. Promoting physical activity in middle with the Councils on Cardiovascular Disease in the Young and Cardiovascular school girls.. S.. Lee. 290–302. McKenzie. S... McRitchie.. Promoting phys- Gallahue. to promote physical activity among children and adolescents.L.A. Yang. Effect of school based physical activity Shoup. U. E.. Br. A. A.X. Thompson. McGraw-Hill... TX. Pro. McKenzie. J.. S. 217. Med. De Clercq.... 62.J. J. Health 19. / Preventive Medicine 56 (2013) 152–161 161 Dobbins.. P. Strikmiller. 138.. Strand. Children. and Metabolism (Physical Activity Committee) in collaboration Webber..L. tion and accelerometer measures of children's physical activity during physical ed. A. E. S. P.G. CONSORT 2010 statement: updated guide- children's heart rates and skill performance.. Harvard Univer. Altman. Arch. Houts. Maximizing motivation in sport and physical education among chil.. Healthy People 2010: Understand- Evaluation of a two-year middle-school physical education intervention: M-SPAN. adolescents and young adults: review and update. J. skills and enjoyment of physical activity. we skill and activate children through primary school physical education lessons? based or evidence-inspired? Psychol. J. 2009. J. Age-related decline in physical activity: a synthesis of human and an- Egger. J.. education policy compliance and children's physical fitness. A. Fairclough.L. D. L. 25. Public Health 87. Epidemiol. Rowe.. Phys. Baranowski. Schmidt. Prev. 2011. E. and effort in British Description and Procedures Manual.V. 36... 2004..T. Phys. Health Educ. M. J.A.. Educ. Green. Tetzlaff. 753–761.M. Sports Exerc.. J. Physical activity content of Motive8 PE compared to primary school pe lessons Dudley. Am. graphical test. Adolesc. Abraham... Epidemiol. 125–135. 340. 1998. Physical physical education: a review. M.L.. Exerc. Pangrazi. Qual.. Stratton. 144–159. Washington. A review of physical activity levels during elementa.P. Br. 2010.. Sci. School-based physi.. J..F. R... G. Med.G.. . Husson. Okely. Sallis. 2009. Chinapaw. M. 557–560.. J. 2007... U. D. D. D.. T. tion study. Measuring inconsistency rate intensity and skill development of physical education students. tions Version 5. Simons-Morton. S.H. Rev. 2010.F. Barnett. D.R. J. Salmon. Pediatr. Pediatr. Murphy. Catellier.. Dugdale.. Lubans.R. Sallis. Liberati. to eighth grade students.W. Med. Cochrane Handbook for Systematic Reviews of Interven. Mass. H.M. quality programme (KISS) on fitness and adiposity in primary schoolchildren: cluster of life. DC. Sci.. Vidoni. McKenzie. Schuldheisz. Espinel. Neb. School of Exercise and Nutritional Sciences and Chinese physical education. 1995. Stata Version 12. Lytle. Sports Exerc. P.R. Sanchez-Vaznaugh. J..J. and weight status in overweight children.. Eur. 55. M. 335. Dietrich.G.L. 17. T.. Lee. 2080. Sci. U.M. Stratton. 2006. S.. 36.... J. 2006. 62. J. B. R. C. Med. c785.. Prev. McMinn. Marshall.. Educ. 160. Brusseau Jr. M. T. Cultural differences in the relationships among autono- McKenzie. 1255–1261. Phillips. A.. J.. T. J. C. Altman. J. Altman. Association between direct observa. NSW Schools Physical Activ. 2003. Pediatr. A...L. Health Phys. Phongsavan. C. 2004.. C.. 2000. S.. Armstrong. S.. Sport 83. Higgins. S.. Gattshall. movement 26–33. P.. E.. T.. B. van Mechelen. Strand. Boston. K. movement skills in children and adolescents: review of associated health benefits. 3. Teo-Koh. van Sluijs. D. physical education and school sport interventions targeting physical activity. 29. E. L.. Griffin. Zask. Kriemler. DeCorby. K. Dandamudi. fect of the child and adolescent trial for cardiovascular health. Bauman. Moher.. Exerc. R. P. 1989. G. 2012. Med. Sallis.A. Ball. Epoch length and measuring physical activity of children in simulated free-living conditions. hope for the future. 21–24..B. J.L. Teach. Kolody. J. 493–501. Parcel. Haythornthwaite. Q. Med. 1214–1224...R. Andersen. Sci. D. including a methodological quality assessment. T. S. 2008. G. N. 25. Esliger.. Res.G. Arch.... Adults. StatCorp LP. Effect of cal activity and a healthful diet among children: results of a school-based interven- school-based interventions on physical activity and fitness in children and adoles.M.. Barnett. J.. McKenzie. Clin. Martin. J. Verstraete.. Davis. B. Clin... Brooks. 9.. P.. Epidemiol. Educ.F. Exerc. H. 40. Pate.. E. A comparison of two instructional formats on heart rate Lubans. 26. Prev. P. 2008.. L.J. 17. ing and Improving Health2nd ed. 2006.. S143.J..J. 1994. J. Rowlands.D. T. C. Young. 1997. A. P. Deeks. activity among children. by a simple.. 923–930.. Dance 27. 986–991. The effects of an intervention strategy on Schulz. 20. D. A systematic review of the effectiveness of in the context of children's overall daily activity levels. T.R. Evidence-based strategies to promote physical McKenzie...R.K... 1732–1738. 2006. Effec- moting physical activity in children and youth a leadership role for schools: a sci. Circulation 114. W. Med. 2009. 6th ed.. 136–149. “Move it groove it” — a collaborative health promotion intervention.. Lonsdale et al. Am. Moderate-to. O'Brien. 34. Cardon. Beighle.. 20–34. Bias in meta-analysis detected imal studies. Educ. 29–49. J. Exerc. San Diego. Physical activity.. 2005a. in meta-analyses.. Sci. Cliff.H.. T. 423–431. Fundamental intensity. 2012.M. M.... Okely. Psychol.. D. Phys. Br. G. Alcaraz... Pediatr. Adolescents. B. H. Sports Exerc.. J. Assoc.. D. L.. Physical ity and Nutrition Survey (SPANS): Full Report..N.. Pearson. accelerometer outputs in children: comparison to direct observation.J. 2010. Beard. Sci. Br. J. Physical activity levels in middle and high school Sallis. Sci.S. Timperio. 448–457. M. 32. Pediatr. Sydney. 1–34. A comparison of two instructional formats on heart rate vention for increasing physical activity in adolescent girls. 26. P. Fit. 1996. Interventions to change health behaviours: evidence. Validation of the RT3 accelerometer for McClain.. Med. E. C. D.M. CA. intensity and skill development. Yu. Exerc. Promoting physi- Kriemler. 2004. mote physical activity in children and adolescents: systematic review of controlled sity Press... Sallis.R. Zahner. Br.. 2008. The Competitive Ethos and Democratic Education. 47. Med. Med. C.. A. C. 1996. Government Printing Office. L. Trial of Activity for Adolescent Girls.J... J.M. P. 2008. E. The effects Fairclough....S. B.. Measuring physical activity in physical cal activity programs for promoting physical activity and fitness in children and education: validation of the SOFIT direct observation instrument for use with first adolescents aged 6–18.. Morgan. Am. Med.M. J. 2012.. 300. cents: a review of reviews and systematic update. Pilgrim. Med.. Cotton. T.S. 40. 353–378. Teach. 2006... A. The Cochrane Collaboration.. Booth. J. M. 452–459.J. 2006. 2011. van Sluijs. ry schools. Am. Faucette. Med. C. Sports Med.L. Recreat.. 463–470. 1598–1600. S. subjective vitality. Martin. Robeson.. G..D.. Adolesc. C.. Anderson. SOFIT (System for Observing Fitness Instruction Time): Generic my support. Timperio. Sports Exerc. 49–55. NSW Ministry of Health.... Meyer. Singh. Ozmun. Scantling. 1997. 6. Children's McKenzie. Stratton. 2005b. A. J. J.V. A. Hardy. et al. W.. Tudor-Locke. 1019–1035. Med. 2003.M. Abraham. Physical Activity.F.L.. 45.M. 42. Sci.M. 327. Sci. et al. P. Stone. P. Hovell. Nader.. T. Phys.L. College Station. dren and adolescents: the case for greater task involvement. Understanding Motor Development: Infants. Quest 48. Sports Med. J..W. Twisk.