Trial Information

Interactive Videogame Bikes and Their Effect on Exercise Adherence and Health Related Fitness Among Families

Short title: Fit Game: Fitness & Game Bike Adherence, Motivation and Exercise Study

Study period: 3 years Clinical phase: Phase III trial

Background:

Cancer is a significant economic burden in Canada, through direct costs to the health-care
system (e.g., care and rehabilitation related to disease) and indirectly through lost
economic output (e.g., missed work due to illness, premature death). At least half of all
new cancer cases and deaths worldwide can be prevented, thus it is imperative that primary
prevention become a focus of research. Recent research suggests a strong relationship
between physical activity and many of the most prevalent forms of cancer including breast,
lung, and colon cancer (i.e., the more an individual exercises the less likely they are to
develop cancer). Despite these findings, over half of the Canadian population fails to meet
levels of physical activity recommended as a preventative measure. Additionally, it has been
found that the largest declines in physical activity occur early in life; thus, promotion
efforts targeting critical transitions to physical inactivity early in life are paramount.
Findings from a Canadian Community Health Survey note that only 21% of Canadian youth are
accumulating enough daily activity to meet international guidelines for optimal growth and
development. As well, national cross-sectional and cohort studies on physical
inactivity/overweight prevalence demonstrate that the most prominent deflection point is
between ages 25 to 35, and this has been linked convincingly to the demands of parenthood.
Thus, two important target groups for disease prevention are parents and their children
through family-based physical activity initiatives. Unfortunately, interventions of this
type are limited and have resulted in little changes in physical activity. Previous studies
have focused heavily on education about the benefits/barriers of physical activity, followed
by a self-monitoring and self-regulatory (e.g., goal-setting) component. One area that has
been overlooked when trying to increase physical activity participation among youth, despite
its reliable and robust association with physical activity, is the modification of affective
expectations or judgements (expected pleasure or enjoyment).

The introduction of new, enjoyable, and engaging exercise activities may present a novel
approach to increase physical activity. One group of activities with this potential is
interactive exercise video gaming including games such as the Game Bike system, Sony
PlayStation EyeToy, and Nintendo Wii. These games allow players to interact physically
(using leg, arm, or whole-body movement) in response to some on-screen virtual activity and
provide a controlled opportunity for physical activity and exercise in a family environment.

Emerging evidence suggests that these games can significantly increase energy expenditure
similar to moderate to vigorous intensity physical activities that can translate into
health-related fitness improvements. Our systematic review of existing active video game
interventions highlight the potential of this approach to increase physical activity in
children and young people. Additionally, our previous research with these games has also
demonstrated the health-related fitness gains even when compared to standard cycling
conditions.

Despite these positive effects, there is very little information present to understand
adherence to exercise videogames (EV). Our own research with university-aged males showed
that an intervention group using an interactive Game bike attended 30% more sessions than a
control group using traditional stationary exercise bikes. Even fewer studies on EV have
evaluated the motivational properties of these games and the potential reasons for this
improved adherence over control physical activity conditions. Only two studies have measured
psychological constructs to examine effects of EV on motivation. Results of our previous
work suggests that EV can effectively change affective judgments about physical activity and
subsequent behaviour unlike most prior intervention efforts.

Despite early positive findings, EV research has notable limitations. First, the populations
employed in EV research have been limited to convenience samples of male undergraduate
students. Research needs to expand to other samples in order to examine the reach of EV.
Second, EV research has almost been exclusively conducted in a laboratory setting. While
helpful for initial pilot/efficacy phase research, EV research needs to be conducted in
ecologically valid locations. The family home seems an excellent test for whether EVs can
still affect psychological, behavioural, and fitness outcomes when situated in a
naturalistic location with other leisure-time stimuli. Finally, the length of EV trials has
been limited to six-week longitudinal tests or single exposure examinations. Trials of
longer duration would be very helpful to examine continuing interest in EV and subsequent
adherence. It may be that EV provides a powerful novel experience in the short term but
wanes similar to other exercise initiatives across time. Our proposed study will overcome
these past limitations and advance the current knowledge of EV.

Objectives:

The primary objective of this study is to evaluate the effect of an interactive exercise
video bike (i.e., Active Cycle) in comparison to a stationary bike (Active Cycle without the
videogame controllers) in front of a TV on physical fitness, use of the bikes, and
perceptions of the bikes. We will also explore whether season (winter/summer), age
(parents/kids) and gender (males/females) affect the use of the various bikes.

Study population:

The targeted population will be inactive families within the Greater Victoria Area, British
Columbia and the Greater Halifax Area, Nova Scotia, Canada.

Number of subjects:

A total of 160 families will be recruited (n=80 per group). 120 families will be recruited
at the Greater Victoria site. The remainder will be recruited from Greater Halifax.

Each family in the EV-interactive condition will receive a videogame bike (Hogan Health
Industries, West Jordan, Utah) that will be linked into the family's Sony Playstation 2 or
3® (Sony Computer Entertainment America Inc, Foster City, California). If the family does
not own a Sony Playstation 2 or 3® it will be provided to them for the duration of the
intervention. Families will choose five videogames from a variety of Sony Playstation 2 or
3® videogames. At 3 months, families will be given the opportunity to select five new
videogames, if they wish. Each family in the control group will receive a traditional
stationary bike (i.e., an Active Cycle without the videogame controllers) which will be
placed in front of their television.

Statistical methods:

Study power:

A sample size of 160 families (80 per group) will be recruited to detect a small-medium
effect size (f2 = .10) in adherence to physical activity (primary outcome) with a type one
error of .05, an average correlation of .75 across time for our dependent variable (DV) of
interest, and a power of .80. Our sample size also considers the main 2 (group) x 2
(parent/child) x 4 (time)repeated measures design and a potential 15% attrition rate. The
prediction-based research will be examined by group condition as well as via the collapsed
sample for mediation analyses. Considering an average of 5 predictor independent variables
(Theory of Planned Behavior (TPB) or Self-Determination Theory (SDT) models), and using a
small-medium effect size (f2 = .10) we will have sufficient power (.80) to evaluate these
predictors at an alpha of .05.

Statistical analysis:

Intention to treat analysis will be used to evaluate the main treatment effect. Missing data
in the primary outcome measures will be imputed using a pre-specified approach. Research
question 1 will be analyzed using a 2 (condition) x 4 (time) repeated measures factorial
ANOVA on the primary outcome of child adherence to the bikes. A child (i.e., the target
child) from each household in the eligibility range will serve for this analysis (chosen
through randomization procedures). Post hoc examinations using Tukey follow-up procedures
will be utilized if necessary.

Our secondary objectives (parent; parent/child; gender; season; fitness variables, etc.),
will also be analyzed using a variant of this design with the addition of factors. Cluster
analysis/Hierarchical Linear Modelling will be used for parent/child collinearity. Our pilot
study (r = .21) and prior research suggests limited collinearity but it is appropriate to
explore findings with these approaches given these are naturally clustered environments
(i.e., family home). Prediction using our models (TPB, SDT) and questions of mediation will
be achieved via multiple regression analyses following standard procedures for these types
of tests.

The qualitative analyses will incorporate the following processes: 1) Invite participants to
review the transcripts of interviews, and summarize their perception of the data, for
accuracy, and check for the trustworthiness of the data; 2) Conduct a thematic analysis
using a reciprocal coding approach where researchers engage in open dialogue about themes
and data interpretation. In doing so, each transcript is first reviewed independently, then
through dialogue composite themes and related critical issues are developed; and 3) Manage
the data using the NVivo software program. NVivo enables theory building, testing and
elaboration. With NVivo, 'free nodes' can be created during the coding process, capturing
participants' perspectives and the investigators' critical issues.