Received: 3 October 2017 Revised: 14 November 2017 Accepted: 17 December 2017

DOI: 10.1111/eip.12540

BRIEF REPORT

Feasibility and preliminary efficacy of a smartphone

application intervention for subthreshold depression
Kana Takahashi | Koki Takada | Kazuki Hirao

Department of Occupational Therapy, Kibi

International University, Takahashi, Japan
Correspondence
Kazuki Hirao, Department of Occupational
Therapy, Kibi International University, 8 Iga-
machi Takahashi, Okayama 716-8508, Japan.
Email: hirao-k@kiui.ac.jp
Funding information
Japan Society for the Promotion of Science,
Grant/Award number: JP16K16583
Aim: We developed a motion picture-reproducing smartphone application (app) designed to
improve self-confidence in individuals with subthreshold depression (StD) and examined the
feasibility and preliminary efficacy of an app intervention for individuals with StD.
Methods: We conducted an open-label, single-arm pre-post study of 22 young adults with
StD. Our intervention involved app usage for at least 70 min/wk for 5 weeks. The primary out-
come measure was the adherence rate. Secondary outcome measures were safety and Center
for Epidemiologic Studies Depression Scale (CES-D) scores.
Results: No participants dropped out during the study. The adherence rate was 50%. The mean
time of app use over 5 weeks was 124.23  81.94 min/wk. No adverse events were reported.
The CES-D scores were significantly improved at 5 weeks post-intervention.
Conclusions: Although this app intervention has the potential to reduce depressive symptoms,
the app needs to be improved to increase adherence rates.

KEYWORDS

application, eHealth, mHealth, smartphone, subthreshold depression

1 | I N T RO D UC T I O N individuals with StD in finding treatment options. The use of smart-

Major depressive disorder (MDD) is one of the most common mental
health problems. The 12-month prevalence of MDD is estimated to
be 5.1% (Kessler et al., 2015), with an annual incidence rate of 3%
(Ferrari et al., 2013). This implies that approximately 60% of depres-
sion cases are new, underscoring the importance of preventing
depression. Although focusing on depression is important, the exis-
tence of subthreshold depression (StD) cannot be overlooked
(Furukawa et al., 2012). StD can be defined as a severity cut point on
a self-administered depression measure, which may vary in the dura-
tion and requirements of core symptoms, but typically requires the
absence of MDD (Kroenke, 2017). An alternative definition of StD is
the presence of 2 to 4 criterion symptoms of depression for ≥2 week
with at least one of the core symptoms (depressed mood or anhedo-
nia) (Kroenke, 2006; Rodríguez, Nuevo, Chatterji, & Ayuso-Mateos,
2012). Individuals with StD mainly face issues regarding self-
confidence (Chen et al., 2013). StD is also a strong risk factor for
MDD (Cuijpers & Smit, 2004). Therefore, StD should be seriously
considered as a target for preventive intervention and treatment
(Kessler, Zhao, Blazer, & Swartz, 1997).
Currently, owning a smartphone has become increasingly com-
mon, and the use of smartphone applications (apps) may help
phones for sharing helpful mental health resources may also aid
underprivileged individuals with StD and bridge the gap due to the
limited availability of mental health services. In this context, we
developed a motion picture-reproducing app (SPSRS) designed to
improve self-confidence in individuals suffering from StD (Takahashi
et al., 2017), based on evidence from subliminal studies (Aoyama
et al., 2017; Van den Bussche, Van den Noortgate, & Rey-
nvoet, 2009).
This study examined the feasibility and preliminary efficacy of an
app intervention for individuals with StD. Our hypotheses are as fol-
lows: (1) The adherence rate is ≥70%. (2) Participants would show
significantly greater changes in the symptoms of depression between
pre- and post-intervention.
2 | METHODS
2.1 | Study design
This was an open-label, single-arm pre-post study for testing the fea-
sibility and preliminary efficacy of an app intervention. The study was
conducted from May 2017 to June 2017 in Takahashi, Okayama,

Early Intervention in Psychiatry. 2018;1–4. wileyonlinelibrary.com/journal/eip © 2018 John Wiley & Sons Australia, Ltd 1
2 TAKAHASHI ET AL.
Japan. The study was approved by the Ethics Committee of the Kibi
International University and registered at clinicaltrials.gov (Identifier:
NCT03110016).
2.2 | Participants and procedure
Participants were recruited from Kibi International University. The
inclusion criteria were individuals (1) with an age of 18 to 24 years,
(2) with a Center for Epidemiologic Studies Depression Scale (CES-D)
(Shima, Shikano, Kitamura, & Asai, 1985) score of ≥16 and (3) who
owned a smartphone with the iOS operating system. The exclusion
criteria were individuals (1) with a lifetime history of MDD, (2) with a
lifetime history of bipolar disorder, (3) who were currently receiving
treatment for a mental health problem from a mental health profes-
sional, and (4) who had experienced a major depressive episode during
the 2 weeks prior to the study, as ascertained using the Mini-
International Neuropsychiatric Interview (M.I.N.I) (Sheehan & Lecrubier,
2003). Accordingly, 22 participants were included in the study. All par-
ticipants provided written informed consent. The researcher helped the
participants to install the app in their smartphones, and a brief explana-
tion of the app was provided. Because the app was designed to be
user-friendly and self-explanatory, other than the embedded standard
tutorials in the app, no further information or training was offered to
the participants during the study period.
2.3 | Intervention
The smartphone app (ie, SPSRS) used in the study was available in
Japanese language and was free of charge for smartphones with iOS
9.0 or higher. The motion picture-reproducing app allows users to
search for and watch videos by using keywords, similar to a general
video playback app. This app uses a YouTube API. Therefore, partici-
pants can freely watch their favourite videos. The difference between
this app and YouTube apps is that in all videos, the SPSRS is pro-
grammed to display common words, such as “can,” “let us try,” “good
luck,” “able” and “do not worry,” that can improve self-confidence in
young adults with StD (Takahashi et al., 2017). These words are ran-
domly displayed in the four corners of the screen (for 17 ms each).
Soon afterwards, positive words such as “nice,” “great,” “fantastic,”
“satisfactory” and “enjoyable” are displayed (Takarada & Nozaki,
2014). These words appear in the middle of the screen (for 150 ms
each). These words are repeatedly displayed every 5 seconds. Our
intervention involved app use for at least 70 minutes per week over
5 weeks. The app focuses on increasing the motivation for beha-
vioural activation.
2.4 | Outcome measures
2.4.1 | Primary outcomes
Adherence rate
The adherence rate was calculated as the number of participants who
used the SPSRS app for 70 minutes per week for 5 weeks divided by
the total number of participants.
2.4.2 | Secondary outcomes
Usability and satisfaction
Usability (ease of use and ease of learning) and satisfaction were
measured using the USE Questionnaire (Lund, 2001) at 5 weeks after
intervention. The USE questionnaires were constructed as seven-
point Likert rating scales. Participants were asked to rate their agree-
ment with the statements, ranging from strongly disagree to strongly
agree (a score of 4 implied neither disagreement nor agreement). The
USE questionnaires were scored as disagree <4 < agree.
Safety
Safety was measured as the experience of major depressive episodes
according to the M.I.N.I (Sheehan & Lecrubier, 2003) at baseline and
1, 2, 3, 4 and 5 weeks after initiating the intervention. We reported
the number of participants with serious adverse events.
Preliminary efficacy
Preliminary efficacy assessments were conducted at baseline and
after 5 weeks. These included depressive symptoms assessed using
the CES-D (Shima et al., 1985), inflammatory biomarkers assessed
based on salivary interleukin (IL)-6 levels (Dowlati et al., 2010), gen-
eral self-efficacy assessed using the General Self-Efficacy Scale
(GSES) score (Sakano & Tohjoh, 1986), mental health assessed using
the 12-item General Health Questionnaire (GHQ-12) score
(Nakagawa & Daibo, 1985) and social anxiety assessed using the Lie-
bowitz Social Anxiety Scale (LSAS) score (Asakura et al., 2002).
2.5 | Statistical analyses
The analysis was conducted based on the intention-to-treat principle.
Paired t-tests were used to detect differences in the pre- and post-
intervention values of the outcome measures (CES-D, GSES, GHQ-
12, LSAS scores and IL-6 levels). The magnitude of the within-group
effect was determined as the effect size based on Cohen’s d for each
scale ([Meanpre − Meanpost]/SDpooled) (Cohen, 1988). Effect sizes
were categorized as small (0.20-0.49), medium (0.50-0.79), and large
(≥0.80) (Cohen, 1988). All statistical tests were two-tailed, and an
alpha value of <0.05 was considered statistically significant. Statistical
analyses were performed using IBM SPSS Statistics 24.0 for Win-
dows (IBM Corp., Armonk, New York).
3 | RE SU LT S
3.1 | Baseline characteristics
No participants dropped out during the study. Table 1 shows the
demographic data of the participants at baseline.
3.2 | Primary outcome
3.2.1 | Adherence rate
Eleven participants completed the 5-week app intervention, that is,
adherence throughout the intervention period (50% adherence rate).
The mean time for app use over 5 weeks was 124.23  81.94 min/wk.
TAKAHASHI ET AL. 3

TABLE 1 Baseline characteristics of the participants 3.3.3 | Preliminary efficacy

Mean SD Table 2 shows the means and standard deviations (SD) of outcome
Age (y) 20 0.62 measures at the baseline (pre) and 5-weeks (post) time points. The
Sleep duration (h) 6.21 0.97 mean total CES-D scores significantly improved from 19.18 (SD,
3.05) to 14.50 (SD, 6.35) over the study period (P = .001). The post-
intervention within-group effect size was Cohen’s d = 0.94, which
N Percentage
corresponds to the large effect of the intervention. No significant
Sex
within-group differences in other outcome measures were observed.
Male 16 72.7

Female 6 27.3
Smoking habit 4 | DI SCU SSION
Smoker 5 22.7
To the best of our knowledge, this is the first study to evaluate the
Non-smoker 17 77.3
feasibility of a motion picture-reproducing app designed to improve
Drinking habit self-confidence in individuals suffering from StD. Despite good
Drinker 1 4.5 usability, the adherence rate was 50%. Typical reasons for not using
Non-drinker 21 95.5 the app included “related videos are not displayed” and “no habit of
Exercise habit watching motion pictures.” The app had a low satisfaction rate of
55%. One of the reasons was that participants were dissatisfied with
Present 9 40.9
the lack of related videos such as those on YouTube. We believe that
Absent 13 59.1
was the reason why the adherence rate declined. As a solution, we
thought that the app should display related videos such as those on
Typical reasons for not using the app included "no habit of watching YouTube. To solve “no habit of watching motion pictures” problem, a
motion pictures" and “related videos are not displayed.” pop-up message prompting the use of the app is required. Through
these improvements, by eliminating participants’ dissatisfaction and
putting in a habit of watching videos, the adherence rate and user
3.3 | Secondary outcomes satisfaction may also increase.
3.3.1 | Usability and satisfaction The results confirm that this app intervention is safe. Further-
more, this app could be a valid tool and may contribute to the reduc-
Totally, 91% (20/22) of participants agreed that the app was easy to
tion of depressive symptoms in young adults with StD. Our pre-post
use. The mean score of ease of use was 5.10  0.80 points. Further-
effect size of 0.94 (large effect) for the CES-D was larger than the
more, 100% (22/22) of participants agreed that the app was easy to
pre-post effect size of 0.51 (medium effect) reported in a previous
learn from. The mean score of ease of learning was 6.78  0.47
study (Walsh, Eisenlohr-Moul, & Baer, 2016) that employed a 4-week
points. Additionally, 55% (12/22) of participants agreed that the
mindfulness-based intervention. In addition, the results of this study
experience of using the app was satisfactory. The mean score of sat-
considerably exceed the measurement error of CES-D (Ohno et al.,
isfaction was 4.42  1.08 points.
2017). It has been shown that unconsciously presented information
can affect behaviour (Aoyama et al., 2017; Van den Bussche et al.,
3.3.2 | Safety 2009). In addition, behavioural activation mechanism theory suggests
No experience of major depressive episodes and no adverse events that behavioural activation decreases the symptoms of depression
that could have been related to intervention participation were (Manos, Kanter, & Busch, 2010). Therefore, we believe that positive
reported. words used in this app activate action and reduce the symptoms of

TABLE 2 Comparison of the pre- and post-intervention values of outcome measures

Pre Post ESa,b

Mean (SD) Mean (SD) Mean differences (SD) t P d 95% CI for ES
CES-D 19.18 (3.05) 14.50 (6.35) 4.68 (5.96) 3.69 .001 0.94 0.30, 1.54

IL-6 (pg/mL) 2.90 (6.85) 1.27 (5.80) 1.63 (4.20) 1.83 .08 0.26 −0.34, 0.85
GSES 6.36 (3.30) 7.23 (3.65) −0.87 (2.03) −2.00 .06 0.25 −0.35, 0.84

GHQ-12 14.18 (4.75) 12.41 (4.16) 1.77 (4.91) 1.69 .11 0.40 −0.21, 0.99
LSAS 50.36 (21.95) 50.14 (23.49) 0.22 (16.06) 0.07 .95 0.01 −0.58, 0.60

Abbreviations: CI, confidence interval; CES-D, Center for Epidemiologic Studies Depression Scale; ES, effect size; GHQ-12, 12-item General Health Ques-
tionnaire; GSES, General Self-Efficacy Scale; IL-6, level of salivary interleukin-6; LSAS, Liebowitz Social Anxiety Scale; SD, standard deviation.
a
Effect sizes are calculated such that positive values represent the improvement in all variables.
b
Effect sizes are calculated using pooled standard deviations of pre-treatment and post-treatment values.
4 TAKAHASHI ET AL.
depression. There are several limitations to this study. To begin, this
app was only available on iOS. Therefore, we plan to make it available
to Android users in a future study. In addition, our study had a rela-
tively small sample size, there was gender imbalance in the sample,
and the study lacked a control group. Additional controlled trials with
large, gender-balanced sample sizes are warranted.
ORCID
Kazuki Hirao http://orcid.org/0000-0002-2467-7564
RE FE R ENC E S
Aoyama, Y., Uchida, H., Sugi, Y., Kawakami, A., Fujii, M., Kiso, K., …
Hirao, K. (2017). Immediate effect of subliminal priming with positive
reward stimuli on standing balance in healthy individuals: A random-
ized controlled trial. Medicine, 96, e7494.
Asakura, S., Inoue, S., Sasaki, F., Sasaki, Y., Kitagawa, N., Inoue, T.,. R., …
Koyama, T. (2002). Reliability and validity of the Japanese version of
the Liebowitz social anxiety scale. Seishin Igaku (Clinical Psychiatry), 44,
1077–1084.
Chen, L., Wang, L., Qiu, X. H., Yang, X. X., Qiao, Z. X., Yang, Y. J., &
Liang, Y. (2013). Depression among Chinese university students: Prev-
alence and socio-demographic correlates. PLoS ONE, 8, e58379.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd
ed.). New Jersey: Erlbaum.
Cuijpers, P., & Smit, F. (2004). Subthreshold depression as a risk indicator
for major depressive disorder: A systematic review of prospective
studies. Acta Psychiatrica Scandinavica, 109, 325–331.
Dowlati, Y., Herrmann, N., Swardfager, W., Liu, H., Sham, L., Reim, E. K., &
Lanctôt, K. L. (2010). A meta-analysis of cytokines in major depres-
sion. Biological Psychiatry, 67, 446–457.
Ferrari, A. J., Somerville, A. J., Baxter, A. J., Norman, R., Patten, S. B.,
Vos, T., & Whiteford, H. A. (2013). Global variation in the prevalence
and incidence of major depressive disorder: A systematic review of
the epidemiological literature. Psychological Medicine, 43, 471–481.
Furukawa, T. A., Horikoshi, M., Kawakami, N., Kadota, M., Sasaki, M.,
Sekiya, Y., … Iwasa, K. (2012). Telephone cognitive-behavioral therapy
for subthreshold depression and presenteeism in workplace: A ran-
domized controlled trial. PLoS ONE, 7, e35330.
Kessler, R. C., Sampson, N. A., Berglund, P., Gruber, M. J., Al-Hamzawi, A.,
Andrade, L., … Gureje, O. (2015). Anxious and non-anxious major
depressive disorder in the World Health Organization world mental
health surveys. Epidemiology and Psychiatric Sciences, 24, 210–226.
Kessler, R. C., Zhao, S., Blazer, D. G., & Swartz, M. (1997). Prevalence, cor-
relates, and course of minor depression and major depression in the
National Comorbidity Survey. Journal of Affective Disorders, 45, 19–30.
Kroenke, K. (2006). Minor depression: Midway between major depression
and euthymia. Annals of Internal Medicine, 144, 528–530.
Kroenke, K. (2017). When and how to treat subthreshold depression.
JAMA, 317, 702–704.
Lund, A. M. (2001). Measuring usability with the USE Questionnaire12.
Usability Interface, 8, 3–6.
Manos, R. C., Kanter, J. W., & Busch, A. M. (2010). A critical review of
assessment strategies to measure the behavioral activation model of
depression. Clinical Psychology Review, 30, 547–561.
Nakagawa, Y., & Daibo, I. (1985). Manual of the Japanese version of the
general health questionnaire. Tokyo: Nihon Bunka Kagakusha.
Ohno, S., Takahashi, K., Inoue, A., Takada, K., Ishihara, Y., Tanigawa, M., &
Hirao, K. (2017). Smallest detectable change and test-retest reliability
of a self-reported outcome measure: Results of the Center for Epide-
miologic Studies Depression Scale, general self-efficacy scale, and
12-item general health questionnaire. Journal of Evaluation in Clinical
Practice, 23, 1348–1354.
Rodríguez, M. R., Nuevo, R., Chatterji, S., & Ayuso-Mateos, J. L. (2012).
Definitions and factors associated with subthreshold depressive con-
ditions: A systematic review. BMC Psychiatry, 12, 181.
Sakano, Y., & Tohjoh, M. (1986). The general self-efficacy scale (GSES):
Scale development and validation. Japanese. Journal of Behavior Ther-
apy, 12, 73–82.
Sheehan, D.V., Lecrubier, Y. (2003). Mini-International Neuropsychiatric
Interview, Japanese Version 5.0.0. Seiwa Shoten, Tokyo (Otsubo, T.,
Miyaoka, H., Kamijima, K., translated from English; in Japanese).
Shima, S., Shikano, T., Kitamura, T., & Asai, M. (1985). New self-rating
scale for depression. Seishin Igaku (Clinical Psychiatry). 27, 717–723.
Takahashi, K., Takada, K., Inoue, A., Ohno, S., Tanigawa, M., Ishihara, Y., …
Hirao, K. (2017). Identification of common words to improve
self-confidence in Japanese students with subthreshold depression.
International Journal of Adolescent Medicine and Health. https://doi.
org/10.1515/ijamh-2017-0018. in press.
Takarada, Y., & Nozaki, D. (2014). Maximal voluntary force strengthened
by the enhancement of motor system state through barely visible
priming words with reward. PLoS ONE, 9, e109422.
Van den Bussche, E., Van den Noortgate, W., & Reynvoet, B. (2009).
Mechanisms of masked priming: A meta-analysis. Psychological Bulletin,
135, 452–477.
Walsh, E., Eisenlohr-Moul, T., & Baer, R. (2016). Brief mindfulness training
reduces salivary IL-6 and TNF-α in young women with depressive symp-
tomatology. Journal of Consulting and Clinical Psychology, 84, 887–897.
How to cite this article: Takahashi K, Takada K, Hirao K.
Feasibility and preliminary efficacy of a smartphone applica-
tion intervention for subthreshold depression. Early Interven-
tion in Psychiatry. 2018;1–4. https://doi.org/10.1111/eip.
12540