Thisistheauthors’finalpeeredreviewed(postprint)versionoftheitempublishedas: Ridgers,NicolaD.,Timperio,Anna,Crawford,DavidandSalmon,Jo2012,Five‐yearchangesinschoolrecessandlunchtimeandthecontributiontochildren’sdailyphysicalactivity,Britishjournalofsportsmedicine,vol.46,no.10,pp.741‐746.

Available from Deakin Research Online:  http://hdl.handle.net/10536/DRO/DU:30047133Reproducedwiththekindpermissionofthecopyrightowner.Copyright:2012,BMJGroup

1

Five-year changes in school recess and lunchtime and the contribution to children’s daily

physical activity

Nicola D. Ridgers, Anna Timperio, David Crawford, Jo Salmon

Centre for Physical Activity and Nutrition Research, Deakin University, Melbourne, Australia

Corresponding Author:

Dr Nicola Ridgers

Centre for Physical Activity and Nutrition Research

Deakin University, 221 Burwood Hwy

Burwood, 3125, Australia

Fax: 61-3-9244-6017

Email: nicky.ridgers@deakin.edu.au

2

ABSTRACT

Objective: To investigate longitudinal changes in children’s recess and lunchtime physical

activity levels and in the contribution of recess and lunchtime to daily physical activity levels

over 5-years amongst 5-6 and 10-12 year olds.

Methods: Data were drawn from two longitudinal studies that were conducted in

metropolitan Melbourne, Australia. Boys and girls (n=2782) aged 5-6 years and 11-12 years

participated in baseline (T0) measures. Physical activity (n=2490) was measured every 60-

secconds for 8 consecutive days using hip mounted accelerometry. Subsequent

measurements were taken at 3-years (T1; n=773) and 5-years (T2; n=634) follow-up.

Physical activity intensities were derived using age-adjusted cut-points. Sedentary time was

defined as 100 counts/min. Longitudinal data were analysed using 3-level (time, child,

school) multilevel analyses, stratified by sex and cohort, and adjusted for potential

confounding variables.

Results: Significant decreases in recess and lunchtime moderate and vigorous physical

activity were observed (p<0.001), with larger decreases occurring in the older cohort.

Associated increases were observed in sedentary time over time (p<0.01). While the

contribution of recess to daily moderate intensity physical activity increased in the younger

cohort over time (p<0.001), significant decreases were observed in the older cohort

(p<0.001).

Conclusion: Physical activity levels during recess and lunchtime decreased in both cohorts

over time. Decreases in the contribution of recess and lunchtime to older children’s daily

physical activity were also observed. Interventions are needed in both primary and

secondary schools to promote physical activity levels during recess and lunchtime,

particularly during the early years of secondary school.

3

INTRODUCTION

Physical activity in childhood is associated with physical and mental health.[1-2] There is

increasing concern that children are not engaging in sufficient physical activity to benefit

health[1] and that increased sedentary recreation outside of school may be a contributing

factor to obesity risk profiles, independent of physical activity levels.[3] Moreover, while

light intensity physical activity (LPA) has negative associations with metabolic risk factors,[4]

little is known how physical activity accrual changes over time. Understanding physical

activity and patterns over time is important for the design of interventions in children that

aim to benefit overall physical activity and health.

Children spend a significant proportion of their waking time in the school environment,

which explains a significant amount of variance in youth physical activity.[5] At school,

opportunities to be active are afforded by physical education classes and recess and

lunchtime breaks.[6] Of these, recess and lunchtime provide opportunities for discretionary

physical activity,[7] where children have some choice about their activities.[6, 8] Boys often

engage in more moderate-to-vigorous (MVPA) and vigorous (VPA) physical activity than girls

during recess and lunchtime,[6, 9-11] although differences between age groups and grades

are inconsistent.[9, 12] However, no studies have examined longitudinal changes in

children’s physical activity during recess and lunchtime, or whether these changes vary by

sex and age of the child. Furthermore, while recess and lunchtime can contribute up to 40%

towards daily physical activity recommendations,[12-13] little research has investigated the

overall contribution of recess and lunchtime to daily physical activity or how this may

change as children age. This is of interest, especially in light of age-related declines in daily

physical activity engagement.[14-15]

4

No research has investigated how recess and lunchtime physical activity levels and the

contribution of these periods to overall daily physical activity change during adolescence or

over the transition from primary (elementary) to secondary (high) school. Moreover, few

studies have investigated recess and lunchtime activity intensities lower than moderate

intensity (MPA).[16] Examining patterns of change over time will help understand the

interplay between the different physical activity intensities, identify potential intervention

points and inform public policy and the debate concerning the importance of recess and

lunchtime in schools.[8]

The purpose of this study is to investigate longitudinal changes in children’s recess and

lunchtime physical activity levels and the contribution of recess and lunchtime to daily

physical activity levels over 5-years amongst 5-6 and 10-12 year olds, and how these vary by

age and sex.

METHOD

Procedure

Data were drawn from two longitudinal studies of Australian school children; the Children

Living in Active Neighbourhoods Study (CLAN)[17] and the Health, Eating and Play Study

(HEAPS).[18] Ethical approval for both studies was provided by the Deakin University Human

Research Ethics Committee, the Department of Education and Training Victoria, and the

Victorian Catholic Education Office. Informed written consent was obtained from parents

and secondary school children at each wave of data collection.

5

Sample

Stratified random sampling proportionate to school size (enrolment greater than 200 pupils)

was employed to recruit schools from metropolitan Melbourne, Australia, into both studies.

Nineteen (10 low socioeconomic status [SES], 9 high SES) and 24 (9 high SES, 7 middle SES, 8

low SES) elementary schools selected agreed to participate in the CLAN and HEAPS studies,

respectively. In both studies, all children in Grade Prep (5-6 years) and Grades 5-6 (10-12

years) were invited to participate. In CLAN, 1220 children (38.3% response rate) provided

active consent to participate at baseline (2001; T0). In HEAPS, 1562 children (42% response

rate) provided consent to participate in baseline measures (2002/2003; T0). Families who

were willing to be contacted about future research were invited to participate in follow-up

measures in 2004 and 2006 for CLAN (T1 and T2), and in 2006 and 2008 for HEAPS (T1 and

T2).

Measures

In both studies, questions about demographic characteristics were included in a survey sent

home with the child and were completed by 2689 parents. In addition, 2490 children had

their physical activity measured by accelerometry. At baseline, complete accelerometer

data (based on the inclusion criteria) were collected from 2075 children. At T1 and T2

complete accelerometer were collected from 773 children (88.9% children monitored at T1)

and 634 children (90.2% children monitored at T2), respectively.

6

Physical activity

Children’s physical activity levels were measured for eight consecutive days at each time

point using hip-mounted uni-axial accelerometers (Actigraph model 7164, Fort Walton

Beach, Florida, USA). The epoch length was 60 seconds. Children were instructed to wear

the accelerometer during all waking hours except during water-based activities (e.g.

swimming, bathing). Accelerometers have been validated for use in paediatric populations

and are accepted as an effective objective field-based measurement tool for physical

activity.[19-20]

Break period duration

School bell times were recorded by the schools and used to determine recess and lunchtime

breaks.

Demographic data

At baseline, a questionnaire was completed by the children’s parents that collected

demographic information about the child (e.g. sex, age, date of birth) and the family as a

whole (e.g. maternal education, employment status, marital status). The self-reported

highest level of maternal education was used as a proxy-measure of SES, consistent with

previous studies,[21-22] and was classed as low (some high school attendance or less),

medium (high school or trade certificate completed) and high (tertiary education).

7

Data management

Accelerometer data were downloaded and data recorded on the first day discarded. Only

weekday data were examined here. Non-wear time was defined as sustained 20-minutes

periods of zero counts, and total duration of these periods represented the duration the

total non-wear duration.[23] For a day to be considered valid, children were required to

have produced counts for 610 minutes (T0), 646 minutes (T1) and 633 minutes (T2),

representing non-missing counts for at least 80% of a standard measurement day, defined

as the length of time that at least 70% of the sample wore the monitor,[23] and to have

produced counts for at least 50% of both recess and lunchtime. Children who provided valid

recess, lunchtime and daily physical activity data on at least three weekdays were retained

for further analysis.

Age-specific cut-points [24] determined durations of light activity (LPA; 1.5 -3.99 METS),

MPA (4-5.99 METS),[25] and VPA (>6.0 METS). Sedentary time was defined as <100

counts/minute.[26] The average duration (minutes) of sedentary, LPA, MPA and VPA per

valid day was determined. The average accelerometer wear time per valid day was also

calculated. The percentage of time children engaged in sedentary, LPA, MPA and VPA during

recess and lunchtime were determined for days that met inclusion criteria. The relative

contribution of recess and lunchtime to daily weekday physical activity was calculated as a

proportion using ((time in activity intensity/total time in activity intensity during the school

day)*100), and averaged over valid days.

Statistical analysis

8

Descriptive analyses were initially calculated for all measured variables. Initial analyses were

conducted to investigate whether differences occurred between children who provided 3, 4

and 5 days of data for use in the analyses. As children with five valid days engaged in more

sedentary activity and less MPA and VPA than children with 3 or 4 days of data at each time

point, analyses adjusted for number of valid days.

Multilevel analyses were used to examine changes in children’s recess and lunchtime

physical activity over time, and the contribution of recess and lunchtime to daily physical

activity. Multilevel models are the most appropriate technique for analysing nested data

that are not independent of each other (e.g. time points within children) and violate the

assumption of independent observations.[27-28] Multilevel modelling is robust against

missing data points, and can estimate effects over time using incomplete data sets.[29] As

such, all data collected were included in the analyses.

A three-level model was used in the analyses, namely the measurement time point (T0, T1,

T2; Level 1), children (Level 2) and baseline school (Level 3). To estimate changes in activity,

and the contribution of recess and lunchtime to daily physical activity, two dummy predictor

variables were generated. These were for physical activity levels at T1 and T2 compared to

T0. The random structure considered random intercepts and random slopes on T1 and T2.

Potential confounding variables were also added to the model as these may influence the

magnitude of any differences observed.[28] These variables included sex, maternal

education, study, number of valid days of accelerometry, daily accelerometer wear time and

the break period duration (recess or lunchtime). Analyses were stratified by cohort,

9

therefore the younger cohort analyses indicate changes across primary school, while the

older cohort analyses document the transition from primary school into secondary school.

Analyses were conducted with the percentage of recess and lunchtime spent in sedentary,

LPA, MPA and VPA, and the contribution of break periods to daily MPA and VPA as the

dependent variables. Separate analyses were conducted for recess and lunchtime.

To examine differences between boys and girls on the dependent variables, potential effect

modification was assessed by constructing interaction terms between sex and the T1 and T2

dummy variables. Subgroup analyses were conducted in the event of significant

interactions. The Wald Statistic was used to assess the significance of the regression

coefficients.[28] Statistical significance was set at p<0.05 and p<0.1 for interaction

terms.[28] Data were analysed using MLwiN 1.10 software (Institute of Education,

University of London, UK).

RESULTS

At T0, most families reported speaking English at home (89.3%). Approximately a third

(38.9%) of mothers reported full-time employment, while 19.7% reported being employed

part-time. Based on maternal education, 37.1% were of medium SES and 34.6% were of high

SES. Average recess and lunchtime durations were 27.2 ± 5.5 and 55.9 ± 6.7 minutes (T0),

25.1 ± 6.3 and 53.4 ± 8.2 minutes (T1), and 24.8 ± 6.2 and 56.4 ± 13.9 minutes (T2),

respectively. There were no differences in physical activity between those with follow-up

data and those without, although those followed up had significantly higher maternal

10

education (43% vs. 32%) and higher proportions were employed part-time employment

(26% vs. 17%), and spoke English at home (96 % vs. 4%).

The physical activity levels of the children at baseline (mean, SD) are displayed in Table 1.

The proportion of time spent sedentary or engaged in physical activity were similar across

recess and lunchtime for all groups.

Table 1: Children's physical activity levels at baseline by age group and sex (T0)

Younger Girls(a)

(n = 300)

Younger Boys(a)

(n = 315)

Older Girls(b)

(n = 796)

Older Boys(b)

(n = 664)

Recess

%Sedentary 16.7 (12.7) 13.8 (13.5) 25.0 (14.4) 19.8 (15.0)

%LPA 47.1 (11.3) 40.3 (14.3) 51.2 (12.5) 49.8 (14.1)

%MPA 23.8 (10.6) 26.0 (10.9) 14.0 (8.5) 19.7 (10.2)

%VPA 12.3 (10.3) 19.9 (14.6) 9.8 (6.9) 10.7 (10.9)

Lunchtime

%Sedentary 14.9 (11.1) 13.3 (10.2) 23.2 (13.3) 18.4 (13.9)

%LPA 46.5 (10.5) 41.2 (11.4) 52.2 (10.4) 48.9 (11.9)

%MPA 26.3 (9.9) 27.5 (9.3) 15.4 (7.9) 21.5 (10.3)

%VPA 12.3 (8.6) 18.0 (11.5) 9.2 (6.1) 11.2 (9.8)

Daily MPA (min) 97.7 (22.3) 106.6 (22.1) 47.9 (16.1) 62.6 (17.7)

Daily VPA (min) 39.7 (16.9) 52.9 (22.9) 18.8 (13.6) 28.9 (17.1)

Daily MVPA (min) 137.4 (34.7) 159.5 (37.9) 66.7 (26.2) 91.5 (30.7)

11

Daily wear time (min) 751.9 (77.5) 754.0 (65.5) 806.9 (79.7) 813.2 (84.0)

a5-6 years at baseline; b10-12 years at baseline

12

Changes in recess and lunchtime physical activity over time

Potential effect modification analyses revealed significant interaction terms between sex

and the time variables for all physical activity intensities for recess and lunchtime, and for

the relative contribution of recess and lunchtime to daily physical activity levels. The results

presented in Tables 2 and 3, therefore, are stratified by gender.

Statistically significant T1 effects were found for boys and girls in both the younger and

older cohorts for all recess and lunchtime physical activity intensities, with the exception of

girls’ LPA (younger and older cohort) during recess and lunchtime and MPA (younger cohort)

during recess and lunchtime (Table 2; p>0.05). For example, the proportion of time spent

sedentary during recess between T0 and T1 increased between 5.6-7.1% for younger

children and 22.6-25.2% for older children, with similar increases observed during

lunchtime. In contrast, recess VPA decreased 5.4-5.9% for younger children and 4.2-8% for

older children between T0 and T1, with similar decreases observed during lunchtime.

Significant decreases in LPA, MPA and VPA between T0 and T1 were observed for both boys

and girls in older cohort.

A similar pattern was observed for the T2 analyses. In the younger cohort, significant

increases in sedentary time and LPA and significant decreases in MPA and VPA were found

for boys and girls during recess and lunchtime, with the exception of boys’ LPA during

recess. For the older cohort, significant increases in sedentary time and significant decreases

in LPA, MPA and VPA were found between T0 and T2 (Table 2).

13

Tabl

e 2:

Ave

rage

cha

nges

in re

cess

and

lunc

htim

e ph

ysic

al a

ctiv

ity o

ver 3

yea

rs (T

0 to

T1)

and

5 y

ears

(T0

to T

2) b

y se

x an

d ag

e gr

oup.

Yo

unge

r Coh

ort

Old

er C

ohor

t

Boys

Rece

ss

%Se

dent

ary

7.05

(2.9

2 to

12.

09)*

* 18

.60

(13.

25 to

23.

68)*

**

25.2

3 (1

9.83

to 3

0.63

)***

27

.58

(24.

17 to

30.

99)*

**

%LP

A 5.

57 (2

.57

to 8

.57)

***

3.07

(-0.

35 to

6.5

0)

-11.

22 (-

15.0

4 to

-7.4

0)**

* -1

2.65

(-15

.42

to -9

.88)

***

%M

PA

-6.1

1 (-8

.97

to -3

.24)

***

-12.

47 (-

15.6

8 to

-9.2

8)**

* -8

.49

(-10.

47 to

-6.5

2)**

* -1

0.35

(-12

.04

to -8

.65)

***

%VP

A -5

.93

(-8.8

8 to

-2.9

8)**

* -9

.08

(-11.

84 to

-6.3

2)**

* -4

.19

(-5.0

8 to

-3.2

9)**

* -4

.49

(-5.4

9 to

-3.4

8)**

*

Lunc

htim

e

%Se

dent

ary

7.64

(3.9

3 to

11.

34)*

**

16.0

5 (1

2.21

to 1

9.88

)***

23

.14

(19.

09 to

27.

14)*

**

32.1

6 (2

8.98

to 3

5.33

)***

%LP

A 5.

29 (2

.65

to 7

.94)

***

5.75

(2.6

6 to

8.8

3)**

* -9

.24

(-11.

72 to

-6.7

6)**

* -1

5.88

(-18

.08

to -1

3.69

)***

%M

PA

-7.2

9 (-9

.64

to -4

.94)

***

-12.

74 (-

15.5

8 to

-9.9

0)**

* -8

.50

(-9.6

3 to

-7.3

6)**

* -1

0.96

(-12

.21

to -9

.70)

***

%VP

A -5

.35

(-7.7

9 to

-2.9

1)**

* -9

.04

(-11.

19 to

-6.8

9)**

* -4

.53

(-5.4

1 to

-3.6

4)**

* -5

.24

(-6.2

2 to

-4.2

6)**

*

Girl

s

14

Rece

ss

%Se

dent

ary

5.64

(1.1

2 to

10.

15)*

8.

30 (3

.22

to 1

3.37

)**

22.6

1 (1

7.14

to 2

8.08

)***

26

.61

(22.

98 to

30.

23)*

**

%LP

A 1.

35 (-

1.56

to 4

.26)

9.

86 (6

.44

to 1

3.29

)***

-1

.85

(-5.7

7 to

2.0

7)

-4.6

5 (-7

.67

to -1

.64)

**

%M

PA

-1.4

1 (-4

.22

to 1

.41)

-5

.16

(-8.3

5 to

-1.9

8)**

-1

0.63

(-12

.68

to -8

.58)

***

-12.

44 (-

14.2

8 to

-10.

59)*

**

%VP

A -5

.36

(-8.2

6 to

-2.4

6)**

* -1

2.20

(-15

.73

to -1

0.21

)***

-7

.96

(-8.9

7 to

-6.9

6)**

* -9

.31

(-10.

47 to

-8.1

5)**

*

Lunc

htim

e

%Se

dent

ary

4.09

(0.4

4 to

7.3

3)*

8.27

(4.5

9 to

12.

24)*

**

17.6

3 (1

3.52

to 2

1.72

)***

23

.97

(20.

62 to

27.

32)*

**

%LP

A 0.

93 (-

1.64

to 3

.51)

7.

58 (4

.50

to 1

0.66

)***

-0

.19

(-2.7

7 to

2.3

9)

-2.6

3 (-5

.03

to -0

.23)

*

%M

PA

-1.7

8 (-4

.08

to 0

.52)

-4

.88

(-7.7

1 to

-2.0

5)**

* -9

.66

(-10.

91 to

-8.4

0)**

* -1

1.76

(-13

.20

to -1

0.31

)**

%VP

A -3

.02

(-5.4

1 to

-0.6

3)*

-11.

10 (-

13.2

3 to

-8.9

8)**

* -6

.36

(-7.3

5 to

-5.3

8)**

* -9

.03

(-10.

16 to

-7.9

6)**

*

***p

< 0

.001

, **p

< 0

.01,

*p<

0.05

. CI =

Con

fiden

ce in

terv

als.

The

val

ue re

flect

s the

per

cent

age

chan

ge in

chi

ldre

n’s a

ctiv

ity le

vels

durin

g re

cess

and

lunc

h be

twee

n Ye

ar 1

(T0)

and

Yea

r 3 (T

1) a

nd Y

ear 1

(T0)

and

Yea

r

5 (T

2). A

pos

itive

v

alue

refle

cts a

n in

crea

se in

chi

ldre

n’s p

hysic

al a

ctiv

ity le

vels

durin

g re

cess

or l

unch

time

at e

ither

T1

or T

2 co

mpa

red

to T

0, w

hilst

a

nega

tive

val

ue re

flect

s a d

ecre

ase

rece

ss o

r lun

chtim

e ph

ysic

al a

ctiv

ity. A

ll m

odel

s are

adj

uste

d fo

r sex

, stu

dy, m

ater

nal e

duca

tion,

num

ber o

f val

id d

ays,

daily

wea

r tim

e an

d br

eak

perio

d du

ratio

n.

15

Tabl

e 3:

Ave

rage

cha

nge

in th

e co

ntrib

utio

n of

rece

ss a

nd lu

ncht

ime

to d

aily

phy

sica

l act

ivity

ove

r 3 y

ears

(T0

to T

1) a

nd 5

yea

rs (T

0 to

T2)

by

sex

and

age

grou

p.

Yo

unge

r Coh

ort

Old

er C

ohor

t

Rece

ss

Ove

rall

%M

PA

2.36

(1.3

8 to

3.3

3)**

* 2.

54 (1

.24

to 3

.84)

***

-3.5

3 (-4

.57

to -2

.45)

***

-2.8

2 (-3

.86

to -1

.78)

***

%VP

A 0.

83 (-

0.92

to 2

.59)

-0

.05

(-2.4

1 to

2.3

2)

-5.4

6 (-6

.79

to -4

.13)

***

-6.6

8 (-8

.03

to -5

.32)

***

Boys

%M

PA

1.97

(0.8

8 to

3.0

7)**

* 1.

57 (0

.20

to 2

.94)

* -3

.47

(-4.5

9 to

-2.3

4)**

* -3

.15

(-4.2

8 to

-2.0

2)**

*

%VP

A -0

.07

(-2.0

5 to

1.9

1)

-0.9

3 (-3

.13

to 1

.27)

-4

.68

(-6.1

5 to

-3.2

2)**

* -5

.93

(-7.4

4 to

-4.4

2)**

*

Girl

s

%M

PA

2.70

(1.6

3 to

3.7

8)**

* 3.

50 (2

.13

to 4

.87)

***

-3.6

1 (-4

.76

to -2

.45)

***

-2.4

3 (-3

.61

to -1

.25)

***

%VP

A 1.

53 (-

0.42

to 3

.46)

0.

80 (-

1.40

to 3

.01)

-6

.42

(-7.9

6 to

-4.8

7)**

* -7

.70

(-9.3

4 to

-6.0

5)**

*

Lunc

htim

e

16

Ove

rall

%M

PA

3.97

(2.5

6 to

5.3

8)**

* 3.

82 (1

.86

to 5

.78)

***

-6.7

6 (-8

.68

to -4

.84)

***

-7.5

8 (-9

.57

to -5

.58)

***

%VP

A 2.

19 (-

0.05

to 4

.43)

0.

14 (-

2.38

to 2

.66)

-9

.13

(-11.

71 to

-6.5

4)**

* -1

3.00

(-15

.67

to -1

0.33

)***

Boys

%M

PA

2.83

(1.1

3 to

4.4

2)**

* 1.

97 (-

0.11

to 4

.05)

-7

.39

(-9.4

6 to

-5.3

3)**

* -8

.95

(-11.

07 to

-6.8

2)**

*

%VP

A 1.

11 (-

1.44

to 3

.66)

-0

.38

(-3.2

4 to

2.4

9)

-9.8

2 (-1

2.66

to -6

.99)

***

-14.

23 (-

17.2

0 to

-11.

26)*

**

Girl

s

%M

PA

4.98

(3.2

6 to

6.7

1)**

* 5.

60 (3

.52

to 7

.67)

***

-6.0

2 (-8

.14

to -3

.91)

***

-5.8

8(-8

.09

to -3

.66)

***

%VP

A 4.

20 (1

.71

to 6

.70)

***

1.45

(-1.

39 to

4.3

0)

-8.2

8 (-1

1.22

to 5

.33)

***

-11.

43 (-

14.5

8 to

-8.2

7)**

*

***p

< 0

.001

, **p

< 0

.01,

*p<

0.05

. CI =

Con

fiden

ce in

terv

als.

The

val

ue re

flect

s the

per

cent

age

chan

ge in

the

cont

ribut

ion

of re

cess

and

lunc

htim

e to

chi

ldre

n’s d

aily

phy

sical

act

ivity

leve

ls be

twee

n Ye

ar 1

(T0)

and

Year

3 (T

1) a

nd Y

ear 1

(T0)

and

Yea

r 5 (T

2). A

pos

itive

v

alue

refle

cts a

n in

crea

se in

the

cont

ribut

ion

of re

cess

or l

unch

time

to c

hild

ren’

s dai

ly p

hysic

al

activ

ity a

t eith

er T

1 or

T2

com

pare

d to

T0,

whi

lst a

neg

ativ

e v

alue

refle

cts a

dec

reas

e in

the

cont

ribut

ion

of re

cess

or l

unch

time

to c

hild

ren’

s dai

ly p

hysic

al

activ

ity. A

ll m

odel

s are

adj

uste

d fo

r sex

, stu

dy, m

ater

nal e

duca

tion,

num

ber o

f val

id d

ays,

dai

ly w

ear t

ime

and

brea

k pe

riod

dura

tion.

17

Contribution of recess and lunchtime to daily physical activity

Younger children engaged in 68.4 (± 18.4) minutes and 31.0 (± 19.9) minutes of daily MPA

and VPA at T1, and in 51.4 (± 19.1) minutes and 17.1 (13.9) minutes of daily MPA and VPA at

T2, respectively. Older children engaged in 37.8 (± 16.3) minutes and 12.8 (± 11.5) minutes

of daily MPA and VPA at T1, and in 28.1 (± 14.5) minutes and 6.3 (± 6.1) minutes of daily

MPA and VPA at T2, respectively. The relative contribution of recess and lunchtime to

children’s weekday physical activity levels are shown in Figure 1. The results of the

multilevel analysis are shown in Table 3. In the younger cohort, the multilevel analyses

revealed significant increases in the contribution of recess and lunchtime to boys’ and girls’

daily MPA between T0 and T1 and between T0 and T2. A significant increase in the

contribution of lunchtime to girls’ daily VPA was also observed between T0 and T1. In the

older cohort, significant decreases were observed in the contribution of recess and

lunchtime to boys’ and girls’ daily MPA and VPA between T0 and T1 and T0 to T2.

[Insert Figure 1 here]

DISCUSSION

This study examined changes in children’s recess and lunchtime physical activity levels and

the contribution of recess and lunchtime to children’s daily physical activity levels in two age

groups over a 5-year period. There were increases in boys’ and girls’ sedentary time and

decreases in MPA and VPA in recess and lunchtime in both cohorts, supporting previous

studies that reported age-related declines in daily physical activity levels[14, 30] and the

notion that declines in daily physical activity start at a young age.[14]

18

The decreases in MPA and VPA observed after 3 years (T1) in the younger cohort are, in

general, lower than short-term observations in control children in European-based

intervention studies. For example, after a 3-month period in Belgium, girls’ MPA and VPA

decreased by 11.5% and 7.1% during recess and 9.9% and 5.3% during lunchtime,

respectively, while decreases were observed in boys’ recess MPA (1%) and VPA (2.6%) and

lunchtime VPA (5.6%).[16] Similarly, in the UK, decreases of 6.5% in control children’s

MVPA two months after baseline measures were reported,[31] which coincided with the

sample moving up one school year. Collectively, these findings suggest that physical activity

levels during recess and lunchtime decrease in both the short- and long-term. As such,

research is needed to identify factors that explain these decreases.

Despite declines in the younger children’s activity levels during recess and lunchtime over

time, the contribution of recess and lunchtime to daily physical activity generally increased

over 3 years (T1), and decreases were mainly observed in VPA after 5 years (T2).

Interestingly, while physical activity decreased during recess and lunchtime, overall daily

physical activity also decreased, but at a greater magnitude, during other parts of the day.

Notably, the contribution to recess and lunchtime to daily physical activity in the younger

cohort at each time point was greater than in previous studies that used objective

measures.[12, 32-33] The present data suggest that recess and lunchtime provide an

important and salient contribution to daily activity levels for Australian primary school

children.

The largest decreases in MPA and VPA and increases in sedentary time were observed in the

older cohort between baseline (T0) and 3 years (T1), coinciding with the transition between

19

primary and secondary school. Early adolescence marks a shift from active play behaviours

to social behaviours,[30] characterised by low physical activity levels. The present study

somewhat supports this, particularly as sedentary time increased between baseline and the

5-year follow-up. This may be explained further by the secondary school physical

environment and the availability of different spaces and equipment for use during recess

and lunch. However, there is currently a dearth of literature that investigates the activity

preferences and behaviours of adolescents during recess and lunchtime. A recent review

stated that research into adolescents’ physical activity levels and behaviour during recess

and lunchtime is needed.[8] While that review excluded intervention studies that specially

targeted MVPA and mainly focusing on the developmental value of recess, no

developmental and health benefits of recess on adolescents was reported suggesting a lack

of data in this area.

Large decreases in the contribution of recess and lunchtime to daily physical activity were

noted over time in the older cohort. However, after 5 years (T2), lunchtime still contributed

10.7% and 6.6% towards daily MPA and VPA, respectively. These data suggest that recess

and lunchtime do contribute a small amount to adolescents’ daily physical activity levels,

and that interventions to increase physical activity and decrease sedentary behaviour during

recess and lunchtime at secondary schools are warranted. To date, there has been no such

published intervention studies in secondary schools. While research is needed to examine

what types of recess (e.g. structured versus unstructured) would be beneficial,[8] such

school-based interventions would be attractive from a public health perspective, particularly

as they would not be competing with screen-based sedentary leisure time activities that

become increasing prevalent in older adolescents outside the school environment.[30, 34]

20

Research is needed to identify what activities adolescents enjoy participating in during

recess and lunchtime, and whether these can be incorporated into the design of

interventions that aim to increase physical activity during these time periods.[35] Based on

the current results, researchers and practitioners should consider using sex and age-

appropriate activities if these discrete periods are to provide opportunities to engage in

physical activity at school.

To the best of our knowledge, this is the first study to investigate changes in recess and

lunchtime physical activity and the contribution of these discrete time periods to daily

physical activity levels as children age. Strengths of this study include measuring a large

population of children over a substantial length of time in two age groups. There are,

however, several limitations. Firstly, data were collected using 60-s epochs, which may

underestimate VPA in children due to the sporadic nature of their physical activity

behaviours.[36] Second, it is not known whether different year groups had their recess and

lunchtime periods together or separately in the different schools. It is possible that this may

influence activity due to the number of children on the playground at one time, which

affects the amount of space children have to engage in different activities. Lastly,

accelerometers do not provide behavioural information, so it is not known what activities

the children engaged in, which could be important for informing the development of future

interventions.

CONCLUSION

21

Understanding patterns of physical activity in discrete time periods of the day provides an

opportunity to inform future intervention development. In this study, physical activity levels

during recess and lunchtime decreased in both groups over time, particularly during the

transition from primary to secondary school. Interventions are needed in both primary and

secondary schools to contribute to overall physical activity levels. Further research is

needed to examine behaviours during recess and lunchtime, and determine how these

change and influence physical activity over time.

WHAT IS KNOWN ON THIS TOPIC

School recess and lunchtime provide daily opportunities for children to engage in physical

activity. Boys engage in more physical activity during recess and lunchtime than girls. Little is

known concerning how children’s recess and lunchtime and their contribution to daily

physical activity change longitudinally.

WHAT THIS STUDY ADDS

Physical activity levels during recess and lunchtime and their contribution to daily physical

activity decrease over time, with larger decreases associated with the transition from

primary to secondary school. Strategies for increasing recess and lunchtime physical activity,

particularly in adolescents, warrant attention.

ACKNOWLEDGEMENTS

22

We gratefully acknowledge the contribution of all project staff, especially Dr. Michelle

Jackson, Dr. Amanda Telford, Sophie Thal-Janssen and Anna Stzendur for the collection of

data in both studies, and Eoin O’Connell for the analysis of the raw data.

COMPETING INTERESTS

The authors declare that they have no competing interests.

FUNDING

NR is supported by an Alfred Deakin Postdoctoral Research Fellowship. AT and DC are

supported by VicHealth Public Health Research Fellowships. JS is supported by a National

Heart Foundation of Australia and sanofi-aventis Career Development Award. Data

collection for HEAPS was funded by the Victorian Health Promotion Foundation (baseline)

and Australian Research Council (follow-ups; ID DP 0664206) and for CLAN by the Financial

Markets Foundation for Children (baseline) and the National Health and Medical Research

Council (follow-ups; ID 274309).

COPYRIGHT STATEMENT

The Corresponding Author has the right to grant on behalf of all authors and does grant on

behalf of all authors, an exclusive licence (or non-exclusive for government employees) on a

worldwide basis to the BMJ Publishing Group Ltd and its Licensees to permit this article (if

accepted) to be published in British Journal of Sports Medicine and any other BMJPGL

products to exploit all subsidiary rights, as set out in our licence

(http://group.bmj.com/products/journals/instructions-for-authors/co_owned_licence.pdf).

23

FIGURE CAPTION

Figure 1: Contribution of recess and lunch to children’s school day physical activity at (a)

baseline (T0), (b) follow-up 1 (T1) and (c) follow-up 2 (T2). Data are unadjusted means.

REFERENCES

1. Andersen L, Harro M, Sardinha L, et al. Physical activity and clustered cardiovascular

risk in children: A cross-sectional study (The European Youth Heart Study). Lancet

2006;368:299-304.

2. Janssen I, LeBlanc A. Systematic review of the health benefits of physical activity and

fitness in school-aged children and youth. Int J Behav Nutr Phys Act 2010;7:40.

3. Salmon J, Campbell KJ, Crawford DA. Television viewing habits associated with

obesity risk factors: a survey of Melbourne schoolchildren. MJA 2006;184:64-7.

4. Ekelund U., Anderssen SA, Froberg, K. et al. Independent associations of physical

activity and cardiorespiratory fitness with metabolic risk factors in children: the

European youth heart study. Diabetologia 2007;50:1832-40

5. Fein AJ, Plotnikoff RC, Wild C, et al. Perceived environment and physical activity in

youth. Int J Behav Med 2004;11:135-142.

6. Beighle A, Morgan CF, Le Masurier G, et al. Children's physical activity during recess

and outside of school. J Sch Health 2006;76:516-20.

7. Guinhouya BC, Lemdani M, Vilhelm C, et al. How School Time Physical Activity Is the

"Big One" for Daily Activity Among Schoolchildren: A Semi-Experimental Approach. J

Phys Act Health 2009;6:510-9.

8. Ramstetter CL, Murrary R, Garber AS. The crucial role of recess in schools. J Sch

Health 2010;80:517-26.

24

9. Ridgers ND, Stratton G, Fairclough SJ. Assessing physical activity during recess using

accelerometry. Prev Med 2005;41:102-7.

10. Willenberg L, Ashbolt R, Holland D, et al. Increasing school playground physical

activity: a mixed methods study combining environmental measures and children's

perspectives. J Sci Med Sport 2010;13:210-6.

11. Zask A, van Beurden E, Barnett L, et al. Active school playgrounds - Myth or reality?

Results of the "Move It Groove It" Project. Prev Med 2001;33:402-8.

12. Mota J, Suva P, Santos M, et al. Physical activity and school recess time: differences

between the sexes and the relationship between children's playground physical

activity and habitual physical activity. J Sports Sci 2005;23:269-75.

13. Ridgers ND, Stratton G, Fairclough SJ. Physical activity levels of children during school

playtime. Sports Med 2006;36:359-71.

14. Nelson MC, Neumark-Stzainer D, Hannan PJ, et al. Longitudinal and secular trends in

physical activity and sedentary behavoir during adolescence. Pediatrics

2006;118:e1627-e34.

15. Sherar LB, Esliger DW, Baxter-Jones ADG, et al. Age and gender differences in youth

physical activity: does physical maturity matter? Med Sci Sports Exerc 2007;39(5):830-

5.

16. Verstraete S, Cardon G, De Clercq D, et al. Increasing children's physical activity

levels during recess in elementary schools: the effects of providing game equipment.

Eur J Public Health 2006;16:415 - 9.

17. Cleland V, Timperio A, Salmon J, et al. Predictors of time spent outdoors among

children: 5-year longitudinal findings. J Epidemiol Community Health 2010;64:400-6.

25

18. Pearson N, Timperio A, Salmon J, et al. Family influences on children’s physical

activity and fruit and vegetable consumption. Int J Behav Nutr Phys Act 2009;6:34.

19. Corder K, Ekelund U, Steele RM, et al. Assessment of physical activity in youth. J

Appl Physiol 2008:105:977-87.

20. Trost SG, Ward DS, Moorhead SM, et al. Validity of the Computer Science and

Applications (CSA) activity monitor in children. Med Sci Sports Exerc 1998:30:629-33.

21. Mo F, Turner M, Krewski D, et al. Physical inactivity and socioeconomic status in

Canadian adolescents. Int J Adolesc Med Health 2005;17:49–56.

22. Mattocks C, Hines M, Ness A, et al. Associations between sex-typed behaviour at age

3½ and levels and patterns of physical activity at age 12: the Avon Longitudinal Study

of Parents and Children. Arch Dis Child 2010;95:509-12.

23. Catellier DJ, Hannan P, Murray DM, et al. Imputation of missing data when

measuring physical activity by accelerometry. Med Sci Sports Exerc

2005;11(Suppl):S555-S562.

24. Freedson PS, Sirard J, Debold E, et al. Calibration of the Computer and Science

Applications, Inc (CSA) accelerometer. Med Sci Sports Exerc 1997:29(Suppl):S45.

25. Harrell JS, McMurray RG, Baggett CD, et al. Energy costs of physical activities in

children and adolescents. Med Sci Sports Exerc 2005;37:329-36.

26. Treuth MS, Schmitz K, Catellier DJ, et al. Defining accelerometer thresholds for

activity intensities in adolescent girls. Med Sci Sports Exerc 2004;36:1259-66.

27. Goldstein H. Multilevel statistical models (2nd Ed.). London: Arnold 1995:1-30.

28. Twisk JWR. Applied multilevel analysis. Cambridge: Cambridge University Press 2006:

108-124.

26

29. Quené H, van den Bergh H. On multi-level modelling of data from repeated

measures designs: A tutorial. Speech Commun 2004;43:103-121.

30. Hardy LL, Bass SL, Booth ML. Changes in sedentary behavior among adolescent girls:

a 2.5 year prospective cohort study. J Adolesc Health 2007;40:158-65.

31. Stratton G, Mullan E. The effect of multicolor playground markings on children's

physical activity level during recess. Prev Med 2005;41:828-33.

32. Ridgers ND, Toth M, Uvacsek M. Physical activity levels of Hungarian children during

school recess. Prev Med 2009;49:410-2.

33. Tudor-Locke C, Lee Sm, Morgan CF, et al. Children's pedometer-determined physical

activity during the segmented school day. Med Sci Sports Exerc 2006;38:1732-8.

34. Janz KF, Burns TL, Levy SM. Tracking of activity and sedentary behaviors in childhood:

The Iowa bone development study. Am J Prev Med 2005;29(3):171-8.

35. Bradley CB, McMurray RG, Harrell JS et al. Changes in common activities of 3rd

through 10th graders: the CHIC study. Med Sci Sports Exerc 2000;32:2071-8.

36. Edwardson CL, Gorely T. Epoch length and its effect on physical activity intensity.

Med Sci Sports Exerc 2010;42:928-34.