Transcript Changes in Clinical Depression

The Efficacy of Prescribed Casual Video Games in
Reducing Clinical Depression and Anxiety©
Carmen V. Russoniello, PhD*1, Matthew Fish, MS (c)*, Kevin O'Brien, PhD*,
Vadim Pougatchev, MD** & Eugene Zirnov, MD, PhD**
* East Carolina University
** Biocom Technologies
1 Corresponding Author
[email protected]
This study was underwritten by PopCap Games, Inc
The Cost of Depression

According to the World Health Organization (2011) depression
is a common diagnosable disorder, affecting approximately 121
million people worldwide. Depression was the leading cause of
disability and the 4th leading contributor to the global burden
of disease in 2000. By the year 2020, depression is projected to
reach 2nd place of the ranking of burden for diseases for all
ages and both sexes. Depression is currently the 2nd cause of
disability in the age category 15-44. Unfortunately, fewer than
25 % of those affected have access to effective treatments.
The Cost of Depression

In the United States an estimated 20.9 million American adults
(9.5% of the U.S. population age 18 or older) suffer from a
mood disorder, and more than two thirds of those (14.8 million
U.S. adults) are cases of major depression. Depression is the
leading cause of disability in the U.S. for people aged 15 to 44.
Depressive disorders often co-occur with anxiety disorders, and
approximately 40 million American adults (about 18 percent of
all U.S. adults) have an anxiety disorder (National Institute of
Mental Health 2011).
Participants
Participants
were Adults (=>18) that
signed an Institutional Review Board
approved informed consent agreeing to
participate and met the criteria score for
inclusion (PHQ9 score=>5).
Participants
Qualifying participants then completed the POMS,
State/Trait Anxiety Inventory, psychological
assessments, demographic profile sheet, and the
remaining components of the Patient Health
Questionnaire (PHQ). The participants also gave a small
saliva sample for biochemical testing. At this point
participants opened an envelope containing a random
assignment to the control or experimental groups.
Experimental Group
If the participant was assigned to the experimental group
they were given a choice of three popular casual video
games to play. Research has demonstrated that freedom
to choose is an important precursor to experiencing the
full benefits of recreation participation. The participant
then played the games of their choice for 30 minutes
while being recorded.
Experimental Group
 In
addition to the two lab sessions scheduled one month
apart, the experimental group was instructed to play the
casual video game of their choice at home for at least 30
minutes 3x per week (At least 24 hours between
sessions) for one month. Participants were asked to keep
a log of the amount of time spent playing the game
during the month. The average game playing time for the
experimental group was minimum 30 max 68 minutes
and the mean 40.7 minutes.
Control Group

If the person was assigned to the control group biosensors were
placed by the researcher and baseline psychophysiological data
was recorded for 6 minutes.

The control participant was then instructed to surf the National
Institutes of Mental Health consumer web site on depression
for 30 minutes while psychophysiology data was being
recorded.
http://www.nimh.nih.gov/health/topics/depressionindex.shtml
Methods
 All
sessions were conducted in a room with
minimal distractions (blank walls, no outside
view, minimal noise). The researcher
administered psychological assessments and
connected the participants to physiological
monitoring equipment following the same
procedure for both groups. All participants sat in
the same chair in front of the same computer.
Methods
If randomized to the control group the participants was
brought to the NIMH web site and instructed to surf the
page for 30 minutes. If randomized into the prescribed
game group the participants preferred game was brought
up on the screen and they were instructed to play for 30
minutes. In both cases the researcher stayed in the room
monitoring all physiological signals for integrity on a
separate monitor.
Data Collection
 Data
was collected from participants on 4 occasions
 Time 1 Pre Session 1 (the initial baseline)
 Time 2 Post Session 1
 Time 3 Pre Session 2 (one month after the initial
baseline)
 Time 4 Post Session 2
Psychological Measurement of Depression

To quantify changes in depression symptoms participants
completed the PHQ-9 a self administered health questionnaire
at the beginning to establish baseline and end of the study one
month later. The Patient Health Questionnaire-9 (PHQ-9) is a
nine-item depression scale and a component of the PHQ that
was originally a component of the Primary Care Evaluation of
Mental Disorders (PRIME-MD) a widely used diagnostic
screening tool in Primary care. The PHQ-9 has since been
validated and widely used as a brief diagnostic and severity
measure for depression in participant populations with
different disorders.
PHQ-9

The PHQ-9 has demonstrated the ability to detect depression
symptoms and change over time. In one study the PHQ-9 scores
discriminated well between participants with any versus no
depressive disorder, as well as between participants with and
without major depression. A PHQ-9 score > or =10 had 91%
sensitivity and 89% specificity for major depression, and 78%
sensitivity and 96% specificity for any depression diagnosis. Since
then the PHQ -9 has been validated as a screening tool as well as
an outcome indicator.
PHQ-9

The PHQ-9 uses check marks in a Likert type scale and is Interpreted as
follows:

0-4 None

5-9 Mild depression

10-14 Moderate depression

15-19 Moderately severe depression

20-27 Severe depression
Physiological Measurement of Depression
 Heart
rate variability (HRV) provides an accurate
assessment of autonomic nervous system
functioning based upon variability in the interbeat interval of heart beats. A robust HRV is
associated with balance between the
sympathetic and parasympathetic branches of
the autonomic nervous system (ANS).
Physiological Measurement of Depression
 In
people who are depressed heart rate variability is
negatively affected and can be assessed using a power
spectrum analysis. In this study HRV changes were used
as a measure of ANS change to determine the
effectiveness of the intervention on the ANS. HRV was
recorded during the entire session using a small ear clip
sensor using photoplethesmography technology.
Analysis followed international HRV standards for
measurement.
Physiological Measurements of Depression

Changes in the Brain function were recorded using a 10
channel electroencephalography device. EEG symmetry
has been extensively used to study the effects of
different stimuli on the brain and it is now generally
accepted that hypoactivity in the left hemisphere is
associated with negative affect and stress while increases
in right hemisphere alpha power are associated with
mania states, euphoria and excitement.
Physiological Measurements of Depression
It was hypothesized that casual video games would
also result in increased left and right frontal activity
and overall symmetry when compared to controls.
A
software computer screen was developed that
automatically collected 6 minutes of baseline and
30 minutes of intervention providing normative
data and R/L brain alpha ratio scores.
Participants
 There
a total of 59 participants included in the study.
Twenty-nine participants served as controls surfing
the National Institutes of Mental Health’s Web Page
on Depression. The experimental group consisted of
30 participants with eighteen choosing to play
Bejeweled 2; seven participants choosing to play
Peggle and five participants choosing to play
Bookworm Adventures (BWA).
Participants
 There
were 30 females and 29 males in the
study. Participants ranged in age from 18 to 74
with average age calculated at 30. Sixty three
percent were gamers and thirty one percent
were not. Twenty three percent were on
antidepressant medications and seventy three
percent were not.
Participants
65% of the participants were EuropeanAmericans; 23% percent were African
Americans, 3.3. % Hispanic Americans,
1.6 % Asian Americans and 3.3% other.

RESULTS
PHQ-9 Differences between Experimental and Control Groups
Time
1
2
3
4
Mean Diff
-.345
3.13
2.85
3.13
Std. Err
1.33
1.36
1.23
1.08
Sig
.797
.024
.024
.005
PHQ-9 scores did not differ at time 1 (the initial baseline) but did so after time
2 or post session one; time 3 or baseline for session two obtained 1 month after
initial data collection and time 4 or post second session.
Changes in Clinical Depression Pre-Post Study
(PHQ-9) Score
Pre Study Control
Post Study Control
Count
Percent
Count
Percent
Pre Study Experimental Count
Percent
Post Study Experimental Count
Percent
Minimal Minor Moderate Severe Total
12
9
6
2
29
41.4% 31.0%
20.7% 6.9%
100.0%
18
9
1
1
29
62.1% 31.0%
3.4% 3.4%
100.0%
Minimal
14
46.7%
26
86.7%
Minor Moderate Severe Total
9
3
4
30
30.0%
10.0%
13.3% 100.0%
4
0
0
30
13.3%
0%
0% 100.0%
PHQ SCORES
Changes in Clinical Depression within CVG Group
100
90
80
70
60
50
40
30
20
10
0
PHQ post
Minimal
PHQ pre
PHQ post
Minimal
42.9
100
Mild
PHQ pre
Moderate
Mild
50
0
Severe
Moderate
7.1
0
Severe
0
0
Summary of Changes in Clinical Depression

The experimental group saw significant reductions in
depression across the board, with all seven subjects previously
classified as suffering from moderate to severe depression
moving to the minor or minimal depression categories. At the
same time, the number of subjects classified as having minor
depression dropped from nine to four. Further analyses
revealed changes between males and females as well as in
younger and older participants.
Changes in Depression Symptoms in Females Pre Post Prescribed Game Play
Changes in Clinical Depression
Changes in Depression Symptoms in Males Pre-Post Study
100
80
60
40
20
PHQ post
0
Minimal
PHQ pre
PHQ post
Minimal
42.9
100
Mild
PHQ pre
Moderate
Mild
50
0
Severe
Moderate
7.1
0
Severe
0
0
Changes in Depression Symptoms < 25 Years Old
Changes in Depression Symptoms => 25 Years Old
Measurement of Overall Mood

To quantify overall mood changes participants completed a self
administered psychological assessment of mood using the Profile of Mood
States (POMS) assessment. The POMS is a factor analytically derived
inventory that measures six subscales: tension, depression, anger, vigor,
fatigue, and confusion and can be used with “Last Week” and “Right Now.”
administrations. Internal consistency for the POMS has been reported at .90
or above. Test re-test reliability is reported between .68 and .74 for all
factors. Construct and predictive validity have been established in four
areas: brief psychotherapy, controlled outpatient drug studies, studies of
response to emotional conditions and studies of concurrent validity
coefficients and other POMS correlates.
Profile of Mood States (POMS)

Overall mood during game play was also measured. The POMS measures
Psychological Tension, Anger, Depression, Vigor, Fatigue and Confusion.
Cumulatively, these six aspects of mood are combined to form “Total Mood
Disturbance,” (TMD)
Time
1
2
3
4
Mean Diff
.252
24.4
24.0
29.6
Std. Err
11.2
7.4
10
7.7
Sig
.982
.002
.020
.000
A decrease in TMD indicates a positive change in mood. In terms of TMD during game play the experimental group experienced a 65%
reduction in TMD and this was significant from control after each measure except for the initial baseline
Changes in POMS Categories

The following slides depict the differences between the video
game group and the six categories of the POMS.
In general there were Decreases in:
 Tension: 49.6%
 Anger: 55%
 Depression: 50%
 Fatigue: 58%
 Confusion: 50%
 and a 33% Increase in Vigor
Reductions in Tensions
16
14
POMS SCORES
12
10
8
6
4
2
0
Video Game
Control
pre Sess1
15.1
13.4
post Sess1
4.6
8.3
pre Sess2
7.8
10.8
post Sess2
2.2
6.6
Video game group differed from control p=.396; p=.009; p=.091; p=.001
Depressed Mood
20
18
POMS SCORES
16
14
12
10
8
6
4
2
0
pre Sess1
Video Game
Control
16.9
18.5
pre Sess2
post Sess1
4.6
8.8
8.5
14.1
post Sess2
2
7
Video game group differed from control p=.673; p=.090; p=.063; p=.044
Changes in Anger
14
POMS SCORES
12
10
8
6
4
2
0
Video Game
Control
pre Sess1
12.8
13.5
post Sess1
1.1
3
pre Sess2
7
10
post Sess2
0.73
3.2
Video game group differed from control p=.778; p=.082; p=.161; p=.065
Changes in Vigor
18
16
POMS SCORES
14
12
10
8
6
4
2
0
Video Game
Control
1
12.5
12.6
2
13.7
8.5
3
16.1
11.5
4
17.1
8.6
Video game group differed from control p=.970; p=.012; p=.016; p=.000
Changes in Fatigue
14
POMS SCORES
12
10
8
6
4
2
0
Video Game
Control
pre Sess1
11.7
13
post Sess1
3.9
9.1
pre Sess2
6.8
10.6
post Sess2
2.4
8.3
Video game group differed from control p=.475; p=.002; p=.025; p=.000
Changes in Confusion
12
POMS SCORES
10
8
6
4
2
0
Video Game
Control
pre Sess1
12
11.5
post Sess1
4.3
8.5
pre Sess2
6
9.9
post Sess2
2.9
7
Video game group differed from control p=.827; p=.005; p=.001; p=.001
Changes in Anxiety

The experimental group saw significant reductions in both state
and trait anxiety. Subjects in the experimental group
experienced a significant decrease in their state anxiety score
between Time 1 (session 1 baseline) and Time 3 (session 2
baseline). Likewise, subjects also experienced a significant
decrease in their trait anxiety score from Time 1 and Time 3.
Subjects within the control group did not experience any
significant change in their anxiety levels.
RESULTS
STAI-S Differences between Experimental and Control Groups
Time
1
2
3
4
Mean Diff
3.28
8.48
8.34
11.64
Std. Err
2.98
2.72
3.06
2.72
Sig
.275
.003
.009
.000
STAI-S scores did not differ at time 1 (the initial baseline) but did so after time 2 or post session one; time 3 or baseline
for session two obtained 1 month after data collection and between time 2 and time 4 or post second session were
statistically significant.
RESULTS
STAI-T Differences between Experimental and Control Groups
Time
1
2
Mean Diff
3.14
7.86
Std. Err
3.22
3.24
Sig
.333
.019
STAI-T scores did not differ at time 1 (the initial baseline) but did so at time 2
or baseline for session two obtained 1 month after data collection.
STAI-S Pre-Post Study Control Group
30
25
Pre Study STAI-S
20
Post Study STAI-S
15
10
Post Study STAI-S
5
Pre Study STAI-S
0
Non-Clinical Anxiety
Clinical Anxiety
STAI-S Pre-Post Study Experimental Group
30
25
Pre Study STAI-S
20
Post Study STAI-S
15
10
Post Study STAI-S
5
Pre Study STAI-S
0
Non-Clinical Anxiety
Clinical Anxiety
STAI-T Pre-Post Study Control Group
30
25
Pre Study STAI-T
20
Post Study STAI-T
15
10
Post Study STAI-T
5
Pre Study STAI-T
0
Non-Clinical Anxiety
Clinical Anxiety
STAI-T Pre-Post Study Experimental Group
30
25
Pre Study STAI-T
20
Post Study STAI-T
15
10
Post Study STAI-T
5
Pre Study STAI-T
0
Non-Clinical Anxiety
Clinical Anxiety
STAI-S Pre-Post Study Control Group
STAI-S Control Group
50
40
30
20
STAI-S Control Group
10
STAI-S Control Group
0
Time 1
Time 2
Time 3
Time 4
STAI-S Pre-Post Study Experimental Group
STAI-S Experimental Group
50
40
30
20
STAI-S Experimenatl Group
10
STAI-S Experimenatl Group
0
Time 1
Time 2
Time 3
Time 4
STAI-S Experimental and Control Comparison
45
40
35
Experimental Group
30
Control Group
25
Control Group
20
Time 1
Experimental Group
Time 2
Time 3
Time 4
STAI-T Pre-Post Study Control Group
STAI-T Control Group
50
40
30
STAI-T Control Group
20
10
STAI-T Control Group
0
Time 1
Time 2
STAI-T Pre-Post Study Experimental Group
STAI-T Experimental Group
50
40
30
STAI-T Experimental Group
20
10
STAI-T Experimental Group
0
Time 1
Time 2
STAI-T Experimental and Control Comparison
50
45
40
Experimental Group
35
Control Group
30
Control Group
25
20
Experimental Group
Time 1
Time 2
Discussion
 The
results of this randomized clinical study clearly
demonstrate the intrinsic value of certain casual
games in terms of significant, positive effects on
depression and anxiety symptoms. In our opinion
these findings support the use of prescribed casual
video games for treating depression and anxiety as an
adjunct to, or even replacement for, standard
therapies including medication.
Discussion
Equally
important, the data supports
the hypothesis that video game play
provokes physiological and biochemical
changes consistent with positive
changes in mood and anxiety.
Discussion
Finally,
given the fact that significant
changes were demonstrated before and
after game play as well as before and after
the month long study, the data supports
both the short and long term efficacy of
prescribed game play.