Deakin Research Online
Deakin University’s institutional research repository
DDeakin Research Online
Research Online
This is the published version (version of record) of:
Graves, Lee E. F., Ridgers, Nicola D., Atkinson, Greg and Stratton, Gareth 2010-11, The
effect of active video gaming on children's physical activity, behavior preferences and body
composition, Pediatric exercise science, vol. 22, no. 4, pp. 535-546.
Available from Deakin Research Online:
http://hdl.handle.net/10536/DRO/DU:30031997
Copyright : © 2010, Human Kinetics, Inc.
Reproduced with the kind permission of the publisher.
Pediatric Exercise Science, 2010, 22, 535-546
© 2010 Human Kinetics, Inc.
The Effect of Active Video Gaming on
Children’s Physical Activity, Behavior
Preferences and Body Composition
Lee E.F. Graves, Nicola D. Ridgers, Greg Atkinson,
and Gareth Stratton
Liverpool John Moores University
Active video game interventions typically provide children a single game that
may become unappealing. A peripheral device (jOG) encourages step-powered
gaming on multiple games. This trial evaluated the effect of jOG on children’s
objectively measured PA, body fat and self-reported behaviors. 42 of 58 eligible
children (8–10 y) randomly assigned to an intervention (jOG) or control (CON)
completed the trial. Intervention children received two jOG devices for home use.
Analyses of covariance compared the intervention effect at 6 and 12 weeks from
baseline. No differences were found between groups for counts per minute (CPM;
primary outcome) at 6 and 12 weeks (p > .05). Active video gaming increased
(adjusted change 0.95 (95% CI 0.25, 1.65) h·d-1, p<.01) and sedentary video
gaming decreased (-0.34 (-1.24, 0.56) h·d-1, p > .05) at 6 weeks relative to CON.
No body fat changes were observed between groups. Targeted changes in video
game use did not positively affect PA. Larger trials are needed to verify the impact
of active video games on children’s PA and health.
Sedentary time in youth is negatively associated with metabolic health (7),
and youth spending more than 2 hr∙d-1 in sedentary screen-based behaviors are at
greater risk for overweight (16). Children play sedentary video games for 30–45
min∙d-1 (21), and this prevalence is highest in lower socioeconomic status (SES)
groups (9). Financial constraints and parental concerns of outdoor safety may
contribute to fewer sport, play and PA opportunities outside the home (9). Active
video games that incorporate movement into video gaming (12,18) may be an
innovative method for encouraging PA and discouraging sedentary behavior in
children from deprived areas.
Research into active video game use in the home and its affect on PA and sedentary time is limited. In 9–12 year olds, a dance simulation game was insufficiently
motivating to maintain regular use in the home over 12 weeks (3). Similarly in 9–18
year olds, home use of a dance game decreased over 6 months (19). Participants
Graves, Ridgers, Atkinson, and Stratton are with the Research Institute for Sport and Exercise Sciences,
Liverpool John Moores University, Liverpool, UK.
535
536 Graves et al.
suggested dance games were insufficiently immersive, fun, or graded in terms of
difficulty to maintain motivation for long-term use (3,19). In ten 10–14 year olds
provided an EyeToy and dance mat for home use, no significant increase in objective
and subjectively measured total, moderate and vigorous PA was found compared
with ten controls at 12 weeks (25). This was despite an increase in active video
gaming, decrease in total video gaming, and, significant decrease in sedentary video
gaming in intervention children relative to controls (25). Similarly, no significant
change was found compared with controls in objectively measured sedentary time
or PA at 10 weeks in forty 7–8 year olds provided with a dance game for home
use, despite lower self-reported sedentary screen time and mean use of the game
for 60 min∙wk-1 (20). While these studies may be insufficiently powered to detect
changes in PA, they suggest active video games may not increase PA or reduce
sedentary time in the long-term.
Participants in previous interventions were typically provided a single game
for use on a specific console, and like any sedentary video game, over familiarization may contribute to decreased motivation for sustained use (23). An alternative
approach may be to reengineer PA into video games young people own and play.
A peripheral device enabling this is jOG (New Concept Gaming Ltd, Liverpool,
UK). jOG encourages step-powered video gaming of light-to-moderate intensity
(unpublished data from our group) and is compatible with PlayStation (PS) 2, PS3
(Sony Co, Tokyo, Japan) and Nintendo Wii (Nintendo Co Ltd, Minami-ku Kyoto,
Japan) consoles. Use of jOG in the home and its impact on PA is unknown. This
randomized controlled trial (RCT) evaluated the effects of jOG on the PA (counts
per minute; CPM) of 8–10 year old children from low SES groups. A secondary
aim was to assess the effects on subcomponents of habitual PA, behavior preferences and body fat.
Methods
Participants and Settings
Pupils in years (grades) 4 and 5 (8–10 y) from three primary schools in low SES
areas in North West England were invited to participate. The study received institutional ethics approval.. Parents/guardians and children gave written informed
consent and assent respectively. Parents/guardians completed medical questionnaires on behalf of their child. Children completed a video gaming survey. Children
were eligible if they were free from the presence of chronic disease and metabolic
disorders, owned a PS2 or PS3 video game console and self-reported playing these
for ≥ 2 hr∙wk-1.
Intervention Design
The RCT was conducted for 12 weeks (January-April 2009). Following baseline
measures, eligible children were randomly assigned at the individual level to an
intervention (jOG) or control (CON) group. Randomisation, stratified by gender,
was achieved by drawing folded sheets of paper, each marked with a participant’s
code, from a hat. Allocation alternated between groups (1st, 3rd, 5th child into jOG
group). Participants and researchers were not blinded to the experimental group.
Intervention participants were given two jOG devices for home-use and familiarized
Active Video Gaming Trial in Children 537
with it during a school-based session. jOG packaging contained instructions for use.
Two devices were given to discourage sedentary play during multiplayer gaming.
jOG links a hip-worn pedometer to a standard console controller and encourages
gamers to step on the spot to use directional controls to generate onscreen character
movement in games (http://www.newconceptgaming.com/products/ps2-compatiblejog/). For every step the pedometer records, 1-s of onscreen movement is obtained.
For continuous gaming, sustained stepping is required. jOG includes an option to
disable the device, allowing games to be played seated while jOG is connected
to the console. Though children were not told this, the jOG instruction manual
contained this information. Participants and their parents were encouraged to play
their PS video games in a step-powered manner with jOG, rather than seated with
or without jOG connected to the console. Though discouraged, participants could
unplug jOG and play games seated. Participants kept the devices after the trial.
The CON group were asked to continue playing their video games as normal and
received two jOG devices upon trial completion.
Instruments and Procedures
At baseline (0 weeks), midtest (6 weeks) and postintervention (12 weeks) participants attended university laboratories for a series of measures. Habitual PA was
also measured at these specified time points.
Behavior Preference Survey. Participants self-reported how much time they
spent in minutes in the following behaviors from waking until lunch, lunch until
dinner, and dinner until bedtime: sedentary (seated) video gaming, active video
gaming (other than jOG step-powered gaming at 6 and 12 weeks), TV viewing,
computer/internet use for pleasure, working on a computer, reading for pleasure
and doing homework. Questions were asked for a typical school and weekend
day, allowing calculation of a weighted weekly estimate (h∙d-1). This approach is
reliable and demonstrates appropriate predictive validity in youth (16). At 6 and 12
weeks children similarly reported time spent playing video games in only a steppowered manner with jOG. This contributed to the active video gaming variable.
Total video gaming time (sedentary, active) and time spent in productive (working
on the computer, reading, doing homework) and leisure (total video gaming,
TV, computer/internet use) behaviors was calculated. Productive behaviors have
been defined to increase knowledge that helps pupils improve their education and
awareness, while leisure behaviors lack these gains (10).
Habitual Physical Activity Assessment and Data Analysis. PA was measured for
7 days every 5 s by an accelerometer (ActiGraph GT1M, ActiGraph Ltd, Pensacola,
FL, USA) worn on the right hip. ActiGraph is valid and reliable for use in child
studies (5). At each time point participants were given the same monitor. Data were
downloaded using manufacturer software (v.3.2.2, ActiGraph Ltd, Pensacola, FL,
USA) and checked for compliance by a data reduction program (Mahuffe, MRC,
Cambridge, UK). Sustained bouts of 20 min of 0 counts indicated monitor removal
(2). Missing counts were removed from the calculation of daily wear time. To be
retained for analyses participants had to provide ≥ 9 hr of wear time for ≥ 3 days
at 0, 6 and 12 weeks. These criteria have shown acceptable reliability in similar
aged children (22). Data were analyzed using individually calibrated activity
count thresholds determined from a treadmill (H P Cosmos, Traunstein, Germany)
538 Graves et al.
protocol performed during each laboratory visit. Participants wore an ActiGraph
(5-s epochs, right hip) and following treadmill familiarization, walked at 4 km∙h-1
for 3 min, rested for 30 s, then jogged at 8 km∙h-1 for 3 min. Speeds of 4 km∙h-1
and 8 km∙h-1 were selected as they differentiated between walking and jogging in
children this age (13). In previous studies in youth, activity counts from 4 km∙h-1
treadmill walking have been used to analyze PA data (1), and such an individual
calibration approach has provided a valid indication of free-living PA (6).
For each speed, the mean of the middle 2 min of data were calculated as the
individual’s count threshold (13). At each monitoring period, individual thresholds
and a sedentary threshold of 100 CPM (28) established the time spent per valid
day sedentary, between sedentary and 3.99 km∙h-1 (PA<4), 4 km∙h-1 and 7.99 km∙h-1
(PA4–7.99) and ≥ 8 km∙h-1 (PA≥8) for each participant. Individual thresholds were
applied at each time point, as leg length and stature, factors influencing counts
generated during locomotion (15), increased in both groups from week 0–12 (p <
.05), as did sitting stature in the jOG group at week 6 (p < .05, one-way repeatedmeasures ANOVA).
Steps, CPM, total PA (TPA: > 100 CPM) and time spent ≥ 4 km∙h-1 (PA≥4)
were also determined for each valid day. The volume of each PA component per
day was calculated (total volume from valid days/number of valid days). Data were
then adjusted for wear time (volume/mean wear time of valid days) to account for
significant differences in sedentary time and PA<4 at baseline between participants
providing ≥ 13.0 hr∙d-1, and, 10.0–10.9 hr∙d-1 and 11.0–11.9 hr∙d-1 of valid mean
wear times. Thus, PA data were presented as min∙h-1 or steps∙h-1 of wear time.
Anthropometry and Maturation Assessment. Stature and sitting stature
were measured to the nearest 0.1 cm using a Leicester Height Measure (Seca
Ltd, Birmingham, UK) and body mass to the nearest 0.1 kg using a calibrated
mechanical flat Seca scale (Seca Ltd, Birmingham, UK) with participants wearing
light clothing and without shoes (17). Body mass index (BMI) was calculated as
mass divided by stature (kg∙m-2). Maturity status was estimated by calculating
years from attainment of peak height velocity (PHV; 24). Stature, sitting stature,
leg length, mass, chronological age and their interactions were used in genderspecific equations to predict each participant’s maturation offset (y to PHV; 24).
Body Fat Assessment. Subtotal and trunk body fat % were assessed by dualenergy x-ray absorptiometry (DXA, Hologic QDR series Discovery A, Bedford,
MA, USA). DXA assessment of body composition has been validated against
hydrodensitometry (30) and is a gold standard for assessing fat mass in youth (11).
Participants were scanned in the supine position while wearing a t-shirt and shorts.
The scanner was calibrated daily. The coefficient of variation (mean baseline, mid
and post measures) for repeated measurements of subtotal body fat was 0.48% and
for trunk body fat 1.17%.
Statistical Analysis
Dependent on data normality by group, independent t tests or Mann-Whitney tests
assessed baseline group comparability for age, maturation offset, number of inactive/
active video game consoles in the home and typical video gaming setting. Analysis
of covariance (ANCOVA) compared the intervention effect at 6 and 12 weeks from
baseline on the primary outcome CPM and all secondary outcome variables. The
Active Video Gaming Trial in Children 539
intervention group (jOG vs CON) was the independent variable and the variable
change score (mid or post minus baseline) the dependent variable. Covariates in
all analyses were the baseline value for the variable to control for any imbalances
at baseline (29) and gender as the stratification factor. Change in maturity offset
(post minus baseline) was included as a covariate in ANCOVA for trunk and
subtotal body fat %, PA<4, PA4–7.99, PA>4, steps and CPM at 12 weeks. This was
included as significant correlations (p < .05; Spearman’s) were observed between
change in maturity offset and these dependent variables. Adjusted change scores
and 95% confidence intervals (CI) for the difference in change between groups are
presented unless stated otherwise. To account for missing PA data a per protocol
analyses (PPA) was conducted. For the primary outcome, the PPA was compared
with intention-to-treat (ITT) analysis, as a sensitivity analysis. To treat missing data,
the monotone imputation technique and ten imputation sets were used. Imputation
was based on all 58 randomized participants. Statistical analyses were performed
using SPSS v.17.0 and statistical significance was set at p ≤ .05.
Effects observed from ANCOVA were additionally evaluated for clinical
importance through prespecification of a minimum clinically important difference
(MCID). This approach makes inferences based on meaningful magnitudes and has
been advised by researchers (14). MCID was determined through a distributionbased method as a Cohen’s d (standardized difference between change scores
between groups) of 0.2 between-subjects SDs (4). However, the SD of pooled
baseline data were used to negate the possibility of individual differences from the
intervention influencing 6 and 12 week SD. The adjusted change score for a given
variable was interpreted as beneficial or detrimental if it exceeded the MCID in
the appropriate direction.
Results
Informed consent was received from 69 of 295 children invited to participate. One
child was withdrawn (by parent) before randomisation and ten were ineligible.
Thus, 58 children were randomized. Due to incomplete PA data for 16 children,
complete data sets were available for 22 (13 m, 9 f; 76%) of the 29 (19 m, 10 f), and
20 (15 m, 5 f; 69%) of the 29 (20 m, 9 f) participants in the jOG and CON groups,
respectively. Thirteen of these 42 children did not provide ≥ 1 valid weekend day of
PA data at each time point (31%: 10 jOG, 3 CON). The remaining 29 participants
(69%: 12 jOG, 17 CON) provided ≥ 1 valid weekend day and ≥ 3 weekdays at
each time point. At baseline jOG and CON groups, respectively, were comparable
(p > .05) in age (mean (SD) 9.2 (0.5) vs 9.2 (0.5) y), maturation offset (-3.2 (0.9)
vs -3.1 (0.8) y to PHV), number of inactive (5.8 (2.2) vs 6.6 (2.4)) and active (1.2
(0.9) vs 1.5 (0.8)) video game consoles in the home, and typical setting of gaming
(home: 100% vs 96%).
Habitual Physical Activity
Similar to the PPA (Table 1), ITT analyses found no significant difference between
groups for CPM at 6 (ITT p = .17) and 12 weeks (ITT p = .38). There were no other
statistically significant intervention effects on PA variables. At 6 and 12 weeks, the
adjusted change score between groups relative to CON for PA4–7.99, PA≥4, steps and
CPM satisfied MCID criterion for detriment.
540
562.2
(148.1)
768.4
(160.1)
21.1 (3.2)
526.4
(114.7)
790.3
(177.7)
20.4 (3.5)
7.0 (2.3)
0.8 (0.6)
6.1 (2.3)
13.5 (3.4)
39.6 (3.5)
jOG
592.7
(145.9)
841.6
(195.6)
21.5 (2.8)
7.7 (3.0)
0.9 (0.8)
6.8 (2.8)
13.8 (3.0)
38.5 (2.7)
CON
Midtest (6 weeks)
619.2
(205.3)
867.1
(193.8)
21.5 (3.5)
7.9 (3.1)
1.2 (0.8)
6.8 (2.9)
13.5 (3.3)
38.5 (3.5)
jOG
663.2
(242.2)
935.2
(178.5)
22.3 (3.1)
8.2 (2.9)
1.1 (0.8)
7.1 (2.7)
14.1 (3.6)
37.7 (3.1)
CON
Postintervention
(12 weeks)
-57.6 (-123.5–8.4)
-58.1 (-159.1–42.9)
-0.7 (-2.2–0.7)
-0.9 (-2.2–0.4)
-0.0 (-0.3–0.3)
-0.9 (-2.1–0.3)
0.3 (-1.0–1.6)
0.7 (-0.7–2.1)
Adjusted change
0–6 (95% CI) **
-50.4 (-186.1–85.3)
-85.5 (-183.2–12.3)
-0.4 (-1.9–1.1)
-0.6 (-2.2–1.0)
0.1 (-0.3–0.6)
-0.8 (-2.3–0.7)
0.2 (-1.3–1.7)
0.4 (-1.1–1.9)
Adjusted change
0–12 (95% CI) **
* Baseline, midtest and postintervention values are unadjusted mean (SD).
** Change scores and 95% CIs are the differences between groups (relative to CON) after adjustment by ANCOVA for the baseline value and gender. ANCOVA for
PA<4, PA4–7.99, PA>4, steps and CPM at 12 weeks are also adjusted for change in maturation offset (post minus baseline).
PA, physical activity; TPA, total physical activity (PA<4, PA4), CPM, counts per minute.
556.5
(151.2)
20.7 (3.9)
TPA
CPM
7.0 (3.4)
7.1 (2.4)
PA4
805.6
(191.7)
1.0 (1.1)
0.8 (0.6)
PA8
Steps
(steps∙h-1)
6.1 (3.2)
6.3 (2.5)
PA4–7.99
14.0 (2.8)
13.6 (2.9)
PA<4
38.9 (3.2)
CON
39.3 (3.9)
jOG
Sedentary
(min∙h-1)
Baseline (0 weeks)
Table 1 Physical Activity Values for the jOG and CON Groups*
Active Video Gaming Trial in Children 541
Behavior Preferences
Step-powered video gaming was significantly greater at week 6 than 12 in the
jOG group (Table 2). A significant increase in active video gaming at 6 weeks
was observed in the jOG group relative to CON. No other statistically significant
intervention effects on behavior variables were found. For active video gaming at
6 and 12 weeks, the adjusted change score between groups satisfied MCID criteria
for benefit (0.14 hr∙d-1 increase). The adjusted change score at 6 weeks between
groups for inactive video gaming satisfied criteria (0.25 hr∙d-1 decrease) for benefit, though the 12-week change score met this criterion in the opposite direction
(detriment). The adjusted change score for total video gaming at 6 and 12 weeks
between groups satisfied criteria for detriment (0.25 hr∙d-1 increase), as did the
change score for leisure behaviors at 6 weeks (0.60 hr∙d-1 increase). The adjusted
change score for TV viewing between groups at 12 weeks satisfied criteria for
benefit (0.37 hr∙d-1 decrease).
Anthropometrics and Body Fat
Compared with CON, the jOG group had a statistically significant increase in stature
at 6 weeks (Table 3). No other significant or clinically relevant changes were found.
Discussion
This RCT examined the short-term effects of a peripheral active video gaming device
(jOG) on the PA, behavior preferences and body fat of 8–10 year old children. No
significant difference in CPM was observed at 6 or 12 weeks between intervention
and control groups. However, a nonsignificant but detrimental change in CPM, steps,
PA4–7.99 and PA≥ 4 at 6 and 12 weeks was observed relative to controls. At 6 weeks,
sedentary video gaming decreased and active video gaming significantly increased
in the jOG group relative to controls, with the increase in active gaming largely due
to jOG use. Overall time spent playing video games also increased. At 12 weeks an
increase in total video gaming was similarly observed relative to controls, due to an
increase in sedentary and active video gaming. Thus, these findings may suggest
an increase in total video gaming is detrimental to PA, regardless of whether the
main contributor is active, or active and sedentary gaming. It is acknowledged that
sample size calculations were not conducted prospectively for the trial. Thus, the
trial may have been insufficiently powered to detect significant effect sizes for PA
variables practically different between groups based on MCID analyses.
In previous home-based active video game interventions a similar trend for
increased active video gaming and reduced sedentary video gaming (25) or sedentary
screen time (20) has been observed. Thus, provision of a peripheral device or active
video game appears to encourage children to experiment with the device and devote
time away from sedentary screen-based activities. However, no positive effect on PA
or sedentary time was observed in any intervention. In the present trial there were
no significant differences between groups at 6 weeks for TV use, reading, doing
homework or working on a computer. However, relative to controls, a nonsignificant
but detrimental increase in time spent in leisure behaviors (TV, total video gaming,
computer/internet use) was found. Relative to controls, the intervention group had
a decrease in TV viewing and increase in total video gaming, with the latter due to
542
2.84
(1.05)
-
0.71
(0.75)
3.57
(1.12)
2.58
(1.68)
1.21
(1.10)
7.04
(2.70)
Inactive video
gaming
Step-powered
jOG gaming
Active video
gaming
Total video
gaming
TV viewing
Productive
behaviors
Leisure behaviors
6.98
(3.34)
1.81
(1.57)
2.85
(2.10)
3.51
(1.46)
0.97
(0.59)
-
2.56
(1.47)
CON
6.21
(3.21)
0.61
(0.57)
2.18
(1.74)
2.94
(2.01)
1.50
(1.53)
0.75
(0.71)
1.37
(1.39)
jOG
5.16
(3.11)
0.81
(0.78)
2.41
(1.89)
2.56
(1.72)
0.79
(0.69)
-
1.78
(1.56)
CON
Midtest (6 weeks)
4.55
(2.21)
0.65
(0.36)
1.44
(1.17)
2.86
(1.76)
0.89
(1.16)
0.38
(0.55)†
1.98
(1.43)
jOG
5.10
(2.10)
0.67
(0.46)
2.26
(1.78)
2.56
(1.35)
0.71
(0.65)
-
1.86
(1.09)
CON
Postintervention
(12 weeks)
0.66 (-1.39–2.71)
-0.15 (-0.59–0.28)
-0.15 (-1.10–0.79)
0.55 (-0.60–1.70)
0.95 (0.25–1.65)‡
-
-0.34 (-1.24–0.56)
Adjusted change 0–6
(95% CI) **
-0.23 (-1.43–0.97)
-0.04 (-0.30–0.21)
-0.71 (-1.58–0.15)
0.62 (-0.18–1.41)
0.46 (-0.04–0.95)
-
0.37 (-0.32–1.05)
Adjusted change 0–12
(95% CI) **
* Baseline, midtest and postintervention values are unadjusted mean (SD).
** Change scores and 95% CIs are the differences between groups (relative to CON) after adjustment by ANCOVA for the baseline value and gender.
† p = .01 (different from midtest).
‡ Significant (p < .01).
Productive behaviors (reading for pleasure, doing homework, working on a computer); Leisure behaviors (TV viewing, total video gaming, computer/internet use for
pleasure).
jOG
(h∙d-1)
Baseline (0 weeks)
Table 2 Behavior Variable Values for the jOG and CON Groups*
543
26.7
(9.4)
20.7
(8.8)
Subtotal body
fat (%)
Trunk body fat
(%)
21.9
(8.3)
27.2
(9.0)
19.7
(4.3)
1.38
(0.05)
38.0
(10.6)
CON
20.8
(9.2)
26.7
(9.6)
18.8
(4.5)
1.35
(0.06)
34.6
(10.2)
jOG
21.8
(8.4)
27.4
(9.1)
19.9
(4.3)
1.39
(0.06)
38.7
(10.6)
CON
Midtest(6 weeks)
21.4
(9.8)
27.1
(9.7)
18.9
(4.6)
1.36
(0.06)
35.3
(10.4)
jOG
22.1
(9.1)
27.2
(9.2)
19.7
(4.2)
1.40
(0.06)
39.0
(10.6)
CON
Postintervention
(12 weeks)
0.5 (-0.6–1.6)
-0.1 (-0.7–0.6)
-0.2 (-0.6–0.2)
0.005 (0.001–0.009) ‡
-0.2 (-0.9–0.5)
Adjusted change 0–6
(95% CI) **
0.9 (-0.5–2.3)
0.6 (-0.4–1.5)
0.0 (-0.5–0.5)
0.003 (-0.002–0.009)
0.2 (-0.8–1.2)
Adjusted change 0–12
(95% CI) **
* Baseline, midtest and postintervention values are unadjusted mean (SD).
** Change scores and 95% CIs are the differences between groups (relative to CON) after adjustment by ANCOVA for the baseline value and gender. ANCOVA for
trunk and subtotal body fat % at 12 weeks are also adjusted for change in maturation offset (post minus baseline).
† p = .01 (different from CON).
‡ Significant (p = .02). 3 decimal places reported to improve interpretation of ANCOVA.
BMI, body mass index.
18.9
(4.5)
1.34
(0.06)†
Stature(m)
BMI (kg∙m-2)
34.1
(10.0)
Body mass
(kg)
jOG
Baseline (0 weeks)
Table 3 Anthropometric and Body Fat Values for the jOG and CON Groups*
544 Graves et al.
increased active gaming. Subsequent analysis indicated that computer/internet use
for pleasure increased in the jOG group relative to controls by 0.3 (-0.4, 1.1) h∙d-1
at 6 weeks. Thus, the nonsignificant difference in sedentary time may be due to
intervention children allocating more time to this alternative sedentary pursuit. If
active video games are to increase PA and reduce sedentary time, interventions may
need to consider concurrently available sedentary behaviors young people enjoy.
Self-reported use of jOG at 12 weeks was half that at 6 weeks, indicating a novelty effect observed in previous trials (3,19,20). These studies noted that boredom
with a dance game led to decreased use over time (3,19). In the current study, jOG
was compatible with multiple video games children owned, which likely minimized
the effect of over-familiarization with a single game. Gaming with jOG however
requires continual stepping on the spot, with physical feedback from the working
muscles inherent. Physical feedback such as using a joystick was found to act as a
‘reality check’ during gaming that distracted adolescents and adults from the game
(32). This reminded the gamer of their surroundings (32) and prevented immersion
into the game and sensations of time loss (33). Thus, step-powered gaming may
have been distractive and felt like exercise over time, with physical sensations
overriding enjoyment. It is also possible that children were unable to sustain the
level of stepping required. These factors may have reduced the perceived value of
using jOG, with sedentary gaming the more reinforcing alternative and preferred
choice (8). The decrease in sedentary gaming at 6 weeks in the jOG group, and
subsequent increase at 12 weeks when jOG use declined supports this. Achieving
an optimal balance between physical input and immersion may be important if
active video games are to be enjoyed and become the preferred choice over sedentary equivalents (8).
No significant effects on BMI and body fat were observed, with findings for
BMI similar to those of Maloney et al. (20). In contrast, Ni Mhurchu et al. (25)
reported reductions in waist circumference relative to controls at 12 weeks. However,
the study was insufficiently powered to detect differences in anthropometrics (25),
and like the present RCT failed to monitor energy intake across the trial. Energy
intake during screen time contributes to obesity (27) and may have confounded
body composition findings. Future screen-based trials should account for changes
in energy intake.
This study had several limitations. The small samples were likely underpowered
to detect changes in BMI or body fat. The 12-week trial provided an indication of
short-term behavior change but not long-term sustainability. Only 29 of 42 participants providing valid PA data had ≥1 weekend day at each time point. However,
no significant differences were found for PA variables at any time point between
those providing 3 valid days (with no weekend day) and 4–7 valid days (data not
shown), supporting the inclusion criteria of any 3 days. Factors that may have
influenced treadmill activity counts across time points include technical issues with
monitors or their positioning (15). These were protected against as best as possible
by one researcher conducting and one monitor being used for, each protocol. The
self-report survey may have been influenced by social demands and biases, and
poor child recall of activity (31). The use of memory cards and console hard drives
may provide objective data on video game use, which might be an opportunity for
future research. Lastly, qualitative data on children’s perception of jOG may have
indicated why the device was or was not used, and highlights an area for future
research to consider.
Active Video Gaming Trial in Children 545
Targeted changes in video game use following provision of an active video
game peripheral device were not accompanied by increased PA or decreased sedentary time in children. Increased total video gaming, due to active or active and
sedentary gaming, may be detrimental to PA in children. Preventing the allocation
of time to other sedentary pursuits and providing PA opportunities away from the
screen appear important if active video games are to benefit PA and health. The
novelty effect observed for use of the peripheral device supports the call for new
active video games that attract children and sustain their interest (26). Longer
interventions in larger samples are needed to verify the effects of active video
gaming on PA and health.
Acknowledgments
Thanks to the pupils and schools for their participation and Jennifer Cain, Kate Eccles,
Richard Yarwood, Craig Davies and James Byrne for assisting data collection. Thanks to
New Concept Gaming Co Ltd for equipment support.
References
1. Andersen, L.B., M. Harro, L.B. Sardinha, et al. Physical activity and clustered cardiovascular risk in children: a cross-sectional study (The European Youth Heart Study).
Lancet. 368:299–304, 2006.
2. Catellier, D.J., P.J. Hannan, D.M. Murray, et al. Imputation of missing data when measuring
physical activity by accelerometry. Med. Sci. Sports Exerc. 37(S11):S555–S562, 2005.
3. Chin A Paw, M.J.M., W.M. Jacobs, E.P.G. Vaessen, S. Titze, and W. van Mechelen. The
motivation of children to play an active video game. J. Sci. Med. Sport. 11(2):163–166,
2007.
4. Cohen, J. Statistical power analysis for the behavioural sciences, 2nd ed. Hillsdale,
NJ: Lawrence Erlbaum Associates, 1988.
5. Ekelund, U., M. Sjostrom, A. Yngve, et al. Physical activity assessed by activity monitor
and doubly labeled water in children. Med. Sci. Sports Exerc. 33(2):275–281, 2001.
6. Ekelund, U., J. Aman, and K. Westerterp. Is the ArteACC index a valid indicator of
free-living physical activity in adolescents? Obes. Res. 11:793–801, 2003.
7. Ekelund, U., S.A. Anderssen, K. Froberg, L.B. Sardinha, L.B. Andersen, and S.
Brage. Independent associations of physical activity and cardiorespiratory fitness
with metabolic risk factors in children: the European youth heart study. Diabetologia.
50:1832–1840, 2007.
8. Epstein, L.H., M.D. Beecher, J.L. Graf, and J.N. Roemmich. Choice of interactive
dance and bicycle games in overweight and nonoverweight youth. Ann. Behav. Med.
33(2):124–131, 2007.
9. Fairclough, S.J., L.M. Boddy, A.F. Hackett, and G. Stratton. Associations between
children’s socioeconomic status, weight status, and sex, with screen-based sedentary
behaviours and sport participation. Int. J. Pediatr. Obes. epub ahead of print:1-7, 2009.
10. Feldman, D.E., T. Barnett, I. Shrier, M. Rossignol, and L. Abenhaim. Is physical activity differentially associated with different types of sedentary pursuits? Arch. Pediatr.
Adolesc. Med. 157:797–802, 2003.
11. Goran, M.I., P. Driscoll, R. Johnson, T.R. Nagy, and G. Hunter. Cross-calibration of
body composition techniques against dual-energy x-ray absorptiometry in youth young
children. Am. J. Clin. Nutr. 63:299–305, 1996.
12. Graves, L., G. Stratton, N.D. Ridgers, and N.T. Cable. Comparison of energy expenditure in adolescents when playing new generation and sedentary computer games: cross
sectional study. BMJ. 335(7633):1282–1284, 2007.
546 Graves et al.
13. Hopkins, N.D., G. Stratton, T.M. Tinken, et al. Relationships between measures of
fitness, physical activity, body composition and vascular function in children. Atherosclerosis. 204(1):244–249, 2009.
14. Hopkins, W.G., S.W. Marshall, A.M. Batterham, and J. Hanin. Progressive statistics for
studies in sports medicine and exercise science. Med. Sci. Sports Exerc. 41(1):3–13,
2009.
15. Jago, R., I. Zakeri, T. Baranowski, and K. Watson. Decision boundaries and receiver
operating characteristic curves: New methods for determining accelerometer cutpoints.
J Sports Sci. 25(8):937–944, 2007.
16. Laurson, K.R., J.C. Eisenmann, G.J. Welk, E.E. Wickel, D.A. Gentile, and D.A. Walsh.
Combined influence of physical activity and screen time recommendations on childhood
overweight. J. Pediatr. 153:209–214, 2008.
17. Lohman, T.G., A.F. Roche, and R. Martorell. Anthropometric standardisation reference
manual. Champaign, IL: Human Kinetics, 1991.
18. Maddison, R., C. Ni Mhurchu, A. Jull, J. Yannan, H. Prapavessis, and A. Rodgers.
Energy expended playing video console games: An opportunity to increase children’s
physical activity? Pediatr. Exerc. Sci. 19:334–343, 2007.
19. Madsen, K.A., S. Yen, L. Wlasiuk, T.B. Newman, and R. Lustig. Feasibility of a dance
videogame to promote weight loss among overweight children and adolescents. Arch.
Pediatr. Adolesc. Med. 161(1):105–107, 2007.
20. Malone, A.E., T.C. Bethea, K.S. Kelsey, et al. A pilot of a video game (DDR) to promote physical activity and decrease sedentary screen time. Obesity (Silver Spring).
16(9):2074–2080, 2008.
21. Marshall, S.J., T. Gorely, and S.J. Biddle. A descriptive epidemiology of screen-based
media use in youth: a review and critique. J. Adolesc. 29(3):333–349, 2006.
22. Mattocks, C., A. Ness, S. Leary, et al. Use of accelerometers in a large field-based
study of children: protocols, design issues, and effects on precision. J Phys Act Health.
5(S1):S98–S111, 2008.
23. Mellecker, R.R., A.M. McManus, L.M. Lanningham-Foster, and J.A. Levine. The
feasibility of ambulatory screen-time in children. Int. J. Pediatr. Obes. 8:1–6, 2008.
24. Mirwald, R.L., A.D.G. Baxter-Jones, D.A. Bailey, and G.P. Beunen. An assessment of
maturity from anthropometric measurements. Med. Sci. Sports Exerc. 34(4):689–694,
2002.
25. Ni Mhurchu, C., R. Maddison, Y. Jiang, A. Jull, H. Prapavessis, and A. Rodgers. Couch
potatoes to jumping beans: A pilot study of the effect of active video games on physical
activity in children. Int. J. Behav. Nutr. Phys. Act. 5:8, 2008.
26. Pate, R. Physically active video gaming: an effective strategy for obesity prevention?
Arch. Pediatr. Adolesc. Med. 162(9):895–896, 2008.
27. Robinson, T.N. Reducing children’s television viewing to prevent obesity: a randomised
controlled trial. JAMA. 282(16):1561–1567, 1999.
28. Treuth, M.S., K. Schmitz, D.J. Catellier, et al. Defining accelerometer thresholds for
activity intensities in adolescent girls. Med. Sci. Sports Exerc. 36(7):1259–1266, 2004.
29. Vickers, A.J., and D.G. Altman. Statistics notes: analysing controlled trials with baseline
and follow up measurements. BMJ. 232:1123–1124, 2001.
30. Wallace, J.A., K. George, and T. Reilly. Validity of dual-energy x-ray absorptiometry
for body composition analysis. In: Contemporary Ergonomics, P.D. Bust (Ed.). London:
Taylor and Francis, 2006, pp. 513–515.
31. Welk, G.J. Physical activity assessments for health-related research. Champaign, IL:
Human Kinetics, 2002.
32. Wood, R.T.A., M.D. Griffiths, D. Chappell, and M.N.O. Davies. The structural characteristics of video games: a psycho-structural analysis. Cyberpsychol. Behav. 7(1):1–10,
2004.
33. Wood, R.T.A., M.D. Griffiths, and A. Parke. Experiences of time loss among videogame
players: an empirical study. Cyberpsychol. Behav. 10(1):38–44, 2007.