Mobile exergaming in Type 2 Diabetes mellitus – innovative ways to overcome physical inactivity and increase exercise adherence

Background
Type 2 diabetes has developed into a worldwide pandemic in recent years that is associated with
vast comorbidity and mortality and has created an enormous financial strain on health care
systems all around the globe. Physical inactivity is known to be one of the most important risk
factors for the development of type 2 diabetes and responsible for much of the diabetes‐related
comorbidity. An increase in regular physical activity is therefore an essential component of a
successful type 2 diabetes treatment. However, despite the proven benefits of regular physical
activity, the vast majority of patients with type 2 diabetes remain inactive – often due to low
motivation and lack of physical activity enjoyment. A recent and promising approach to motivate
sedentary individuals to be more physically active and sustainably adhere to regular physical
activity is exergames. These games integrate physical activity and personal health information into
a game or game‐like setting and thereby promote physical activity through playful and enjoyable
challenges.
Aims
The aims of this Ph.D. project were: (1) to systematically review the current evidence for the
effectiveness of exergaming in overweight and type 2 diabetes and (2) to evaluate the suitability
of the Wii Fit Plus exergame to improve cardiorespiratory fitness in individuals with type 2
diabetes. The aim was further (3) to develop a behavior change technique‐based smartphone
game that delivers individualized exercise and physical activity promotion with the intention to
motivate inactive type 2 diabetes patients to become sustainably physically active and to plan a
24‐week randomized controlled trial evaluating the game’s effectiveness. In addition, it was aimed
(4) to assess the accuracy of a commercial activity wristband and of iOS and Android smartphones
to measure steps during various walking conditions as those devices were intended to be used to
measure the primary outcome (steps per day) in the 24‐week randomized controlled trial. Finally,
the aims were (5) to examine the effectiveness of the smartphone game to increase daily physical
activity and improve glycemic control and aerobic capacity as well as (6) to evaluate the game’s
suitability to increase the intrinsic physical activity motivation and elicit sustained improvements
in physical activity adherence in inactive individuals with type 2 diabetes over a 24‐week period.
Methods
In this Ph.D. project, one systematic review and three studies were conducted. In the systematic
review, electronic bibliographic databases (PubMed, Embase, Web of Science, OpenGrey, and the
Cochrane Central Register of Controlled Trials) were searched up to March 2015. Randomized
controlled trials and cross‐sectional studies published in English in peer‐reviewed journals were
included. Included studies were required to have analyzed the effects of exergames on objectively
measured intensity parameters of physical activity (oxygen uptake, energy expenditure, and heart
rate) in overweight adults with and without type 2 diabetes.
In study 1, data collected from a study conducted by Prof. Schmidt‐Trucksäss before the start of
my Ph.D. was analyzed to evaluate the cardiorespiratory exertion (maximum and mean oxygen
uptake) of 12 elderly individuals with type 2 diabetes during 10‐minute bouts of different Wii Fit
Plus exercises. Oxygen uptake values were compared to the maximally reached values during a
previously conducted cardiopulmonary exercise test on a treadmill. Correlations between oxygen
uptake values reached during exergame activity and those reached during the all‐out exercise test
were analyzed using Spearman’s rank correlation coefficient.
Following this preparatory work and considering its results, I collaborated with an international
team consisting of sports scientists, gamification researchers, professional game developers, and
clinical professionals in developing an innovative smartphone game with the aim to encourage a
healthier, more active lifestyle through gamified physical activity in inactive type 2 diabetes
patients. To evaluate the effectiveness of the game to sustainably increase physical activity in the
target group, I planned a 24‐week randomized controlled trial with daily physical activity (steps
per day) as the primary outcome.
To verify the validity of the devices used to measure the primary outcome (steps per day) in the
24‐week intervention study, I conducted a validation study in an additional investigation of this
Ph.D. project. In the study, 20 participants were equipped with 3 iPhone SE smartphones (placed
in pants pocket, shoulder bag, and backpack), 1 Samsung Galaxy S6 Edge (pants pocket), 1 Garmin
Vivofit 2 wristband, and 2 ActiGraph wGTX+ devices (worn at wrist and hip) and instructed to walk
for five minutes at four predefined walking speeds (1.6, 3.2, 4.8, and 6.0 km/h) and to complete
an outdoor walking course. Video observation was used as gold standard. Validity was evaluated
by comparing each device with the gold standard using mean absolute percentage errors.
In the main study (ClinicalTrials.gov identifier NCT02657018) of this Ph.D. project, 36 inactive,
overweight type 2 diabetes patients between 45 and 70 years of age were randomly assigned to
either the intervention group or the control group. Participants in the intervention group received
the self‐developed smartphone game, and participants in the control group were instructed to
implement the recommendations from the baseline lifestyle counseling autonomously during the 24‐week intervention period. Before and after the intervention period, steps per day were
objectively measured during six consecutive days. In addition, glycemic control (HbA1c) was
measured by analysis of venous blood, and aerobic capacity (oxygen uptake at the first ventilatory
threshold) was assessed during an all‐out exercise test on a bike ergometer. Intrinsic physical
activity motivation was assessed with an abridged 12‐item version of the Intrinsic Motivation
Inventory (IMI). In the intervention group, adherence to the game‐proposed physical activity
recommendations during the intervention period was additionally assessed via the phonerecorded
game usage data. Analyses of covariance with adjustment for the respective preintervention
values were used to compare changes in outcome parameters (steps per day, HbA1c,
oxygen uptake at the first ventilatory threshold, and IMI score) between the two groups.
Correlations between game use (min) and change in steps per day as well as change in workload
(W) at the first ventilatory threshold were analyzed using Spearman’s rank correlation coefficient,
and a linear regression model was used to assess the relationship between total in‐game training
(min) and change in IMI total score.
Results
Publication 1: Effects of Exergaming on Physical Activity in Overweight Individuals. [1]
The systematic review showed that exergames are able to increase physical activity among
overweight individuals. However, the magnitude of this increase as well as the intensity of the
exergame‐related physical activity and consequently their suitability to offer a guidelineconcordant
training is greatly dependent on the console and the game mode as well on the
duration of play. The majority of the included studies were of poor or moderate methodological
quality. No studies exist that objectively assess the effect of exergames on daily physical activity
in type 2 diabetes, and the cross‐sectional nature of all included studies does not permit
judgement on whether exergames are suitable to increase physical activity in the long term. A
general lack of adequate individualization of the intensity and difficulty level was noted for all
exergames, making most game modes unsuitable for an effective training and potentially unsafe
for inactive individuals with chronic diseases and a likely reduced fitness level.
Publication 2: Cardiorespiratory Exertion While Playing Video Game Exercises in Elderly
Individuals With Type 2 Diabetes. [2]
In elderly type 2 diabetes patients, carefully selected Wii Fit Plus exercises elicit peak values in
oxygen uptake that correspond to 60% of the maximally reached values during an all‐out treadmill
test. Mean values were just above 40% for all exercises and were thus in a range that corresponds
to current exercise guidelines. A moderate‐to‐strong positive correlation between the peak
oxygen uptake during exergame play and the maximal value reach during the exercise test was
found, indicating that subjects with a higher fitness level were able to exercise at a higher intensity
during exergame play than individuals with a lower fitness level.
Publication 3: Mobile Exergaming for Health – Effects of a Serious Game Application for
Smartphones on Physical Activity and Exercise Adherence in Type 2 Diabetes Mellitus – Study
Protocol for a Randomized Controlled Trial. [3]
The self‐developed and innovative game takes the proven motivational benefit of exergames into
consideration and further extends them by integrating established behavior change techniques
into an elaborate storyline. Shortcomings with regard to the suitability and effectiveness of the
exercise regimen of current exergames are addressed by offering a fitness level‐adjusted training
with an individualized rate of intensity progression that is based on the performance in established
and in‐game‐conducted exercise tests.
Publication 4: Validity of Activity Trackers, Smartphones, and Phone Applications to Measure
Steps in Various Walking Conditions. [4]
The Garmin Vivofit 2 and iPhone SE showed good accuracy for treadmill walking ≥3.2 km/h (<3%
deviation from video recordings) and for free walking thus being suitable to measure steps at
normal, fast, and even slow walking speeds. The Samsung Galaxy S6 Edge showed good accuracy
(<3% deviation from video recordings) for treadmill walking ≥4.8 km/h and for free walking,
demonstrating suitability to measure steps during normal and fast walking. The good validity of
the iPhone SE was found independently of the phone’s position (pants pocket, backpack, shoulder
bag), suggesting a broad versatility of the device for physical activity measurement in various
settings.
Publication 5: Publication 5 Behavior Change Technique‐based Smartphone Game Sustainably
Increases Daily Physical Activity in Type 2 Diabetes Patients – A Randomized Controlled Trial. [5]
In persistently inactive type 2 diabetes patients, a novel self‐developed smartphone game that
follows a cognitive/behavioral approach and incorporates individualized exercise regimens and
physical activity recommendations elicited significant and clinically meaningful increases in daily
physical activity (+4,000 steps per day) over a 24‐week period, classifying the participants as
sufficiently active post‐intervention (average of 9,782 steps per day). The increase in daily physical
activity was accompanied by an increase in aerobic capacity with a significantly higher oxygen
uptake and workload at the first ventilatory threshold. Total duration of in‐game training was
positively correlated with the change in steps per day and the change in workload at the first
ventilatory threshold. Glycemic control (HbA1c) did not change during the intervention period;
however, as indicated by the significant adjusted difference in HbA1c of ‐0.9 percentage points in
favor of the intervention group, there is reason to assume that regular use of the game supports
the stabilization of glycemic control in medically well‐controlled patients. In the control group,
none of the outcome parameters changed significantly during the intervention period.
Publication 6: Effectiveness of a Behavior Change Technique‐based Smartphone Game to
Improve Intrinsic Motivation and Physical Activity Adherence in Type 2 Diabetes Patients: A
Randomized Controlled Trial. [6]
The 24‐week randomized controlled trial further showed that participants in the intervention
group significantly increased their IMI‐measured intrinsic physical activity motivation during the
intervention period. Participants in the control group did not show any significant changes in their
intrinsic physical activity motivation. The analysis of the usage data revealed that participants in
the intervention group used the game for an average of 143.1 (SD 59.1) minutes of in‐game
training per week, which underlines the game’s potential in motivating formerly inactive type 2
diabetes patients to meet and sustainably adhere to established physical activity
recommendations.
Conclusions
A novel smartphone game that incorporates established motivational elements and personalized
physical activity recommendations into an elaborate storyline elicits clinically relevant increases
in daily physical activity and aerobic capacity and contributes to a medically well‐regulated
glycemic control in persistently inactive type 2 diabetes patients. The game is therefore a suitable
treatment option to motivate inactive individuals to increase daily physical activity that may be
relevant for other inactive target groups with and without chronic diseases. Future research
should further refine the game design as well as the motivational and physical activity‐related
content to make it an even more comprehensive treatment tool for various target groups.