The Zen Garden: A feasibility randomized controlled trial of an m-health tool for blended Competitive Memory Training for self-esteem problems

Low self-esteem is a prevalent symptom across different mental disorders (APA,
2013), including personality disorders, internalizing (e.g., anxiety,
depression and eating disorders) and externalizing disorders (aggressive
behavior and substance abuse), and is generally considered a non-specific risk
factor in physical and mental health (Mann et al., 2004). The relationship
between low self-esteem and mental health is very complex; self-esteem can
increase the individual*s vulnerability to mental health problems but can also
be the consequence of a variety of mental health difficulties, such as panic
disorders and anxiety (Fennell, 2009; Sowislo & Orth, 2013). Addressing low
self-esteem directly with targeted psychological interventions has the
advantage of being a broad-spectrum treatment strategy that may consequentially
improve or prevent the development and the exacerbation of mental health and
social problems, especially when low self-esteem represents a vulnerability
factor for the evolution of ongoing mental health and social problems.

An intervention that has proven quite successful in improving self-esteem is
the Competitive Memory Training (COMET, Korrelbom et al., 2008, 2009b). The
COMET is a brief (i.e., 7 or 8 sessions) trans-diagnostic cognitive
intervention aimed at changing the maladaptive cognitive emotional networks
underlying the expression of different psychopathological symptoms, including
low self-esteem, but also obsessions, panic, worry and rumination. Originally
designed to enhance low self-esteem (Korrelboom et al., 2008, 2009b), the COMET
is based on Brewin (2006)*s notion that emotional and cognitive representations
of concepts (such one*s self-image) are stored in long-term memory and compete
with each other to be activated (i.e., competitive memory retrieval hierarchy).
Some of these representations may be or become dysfunctional, and when too high
in the retrieval hierarchy, they become more dominant and easily retrievable.
The aim of any cognitive (behavioral) therapy is then to change the relative
retrievability of different representations of emotional concepts, rather than
directly modifying negative information. COMET targets dysfunctional thoughts
by strengthening the positive cognitive emotional networks that are in
competition with the more readily activated dysfunctional (negative) thoughts.

In the context of low self-esteem problems, the goal of COMET is to create
stronger networks of functional self-referent representations that would
compete with the dysfunctional ones, and to influence the relative activation
of these *positive* networks by repeatedly relating them to personalized cues
that are known to trigger the patient*s dysfunctional ideas and cognitions
(i.e., conditioned stimuli) about his/her self-image (i.e.,
counterconditioning). Such COMET protocol for self-esteem problems has proved
efficacious in addition to treatment as usual in enhancing self-esteem and
reducing depression in patients with eating disorders (Korrelboom et al.
2009a), personality disorders (Korrelboom et al. 2011), depressive disorders
(Korrelboom et al. 2012) and anxiety disorders (Staring et al., 2016), and has
been further extended to target worry and rumination (Ekkers et al., 2011) and
panic (Korrelboom et al., 2014).

The COMET protocol for low self-esteem is currently deployed in the clinical
practice (on top of treatment as usual) and the strong effectiveness and short
duration prompted the development of an in-house web-based version of the
therapy to be used in a blended format in one of the leading institutions
delivering such intervention in the Netherlands (PsyQ). The main goal behind
the idea of creating an e-health COMET program was to improve efficiency and
cost-effectiveness of the treatment and increase its accessibility to patients
in their daily routine. The blended COMET program was designed to alternate
face-to-face sessions with the patients* independent use of the online platform
every other week (i.e., weeks 2, 4 and 6 of the COMET protocol) and for daily
exercises. However, the first pilots of the online program were not successful:
a general lack of motivation to complete the online program was observed, with
large drop-outs, very low adherence to the daily homework, and an overall
negative feedback about the layout and structure of the online program.
Therapists involved in the project also highlighted the difficulty of patients
to stay motivated and keep up with the exercises and to autonomously go through
the progressive stages of the protocol.

The high degree of therapist guidance, accountability and interactivity; the
need of consistent exercising and rehearsing during daily homework, and the
intensive nature of the intervention, are all factors that seem to lower
compliance to treatment when delivered outside the face-to-face setting, as in
the case of the negative results of the first online COMET program at PsyQ.
Further feedback by COMET experts and therapists pointed out that also in the
full face-to-face version of the therapy patients do not completely comply or
have difficulty in complying with the schedule of daily practice of the
exercises learned during the therapy sessions (i.e., a minimum of 5-7 times a
day for 5-10 minutes). Treatment adherence and intensive daily practice, which
are necessary elements for a successful outcome, seem to be a general issue in
the COMET clinical practice.

The information gathered during this phase of desk-research and interviews with
multiple COMET practitioners grounded a new round of design and development of
an e-health version of the COMET. The main idea was to keep the original
blended format of the e-health version protocol (i.e., mix of face-to-face and
independent work) but improve the design, content, usability and efficiency of
the digital component of the therapy. A gamified e-health version of the COMET
was then developed as a mobile smartphone application based on the concept of a
playful garden.
A user-centered and iterative design approach has then been used to design and
develop the app, by actively involving the end users of the app in the
development process. The app design was reviewed and evaluated based on
formative research (i.e., focus groups) conducted with a small group of COMET
patients and therapists, who evaluated different game concepts and mock-ups in
the early design phase. The concept of the Zen Garden was then put into
production, with a continuous iterative evaluation of multiple, progressive
prototypes of the app. These iterative rounds of evaluation have been conducted
both in-house and with experienced COMET therapists, in order to reach a final
version of the app approved both in terms of contents and intended use as an
integral part of a revised COMET protocol, blending the face-to-face and
digital components (more details presented in the Intervention section).

The resulting Zen Garden mobile app incorporates and gamifies the main
principles of four out of the seven stages of the COMET protocol for
self-esteem problems (i.e., weeks 2, 3, 4 and 6). The app is based on a playful
garden concept to help enhancing the therapy experience and motivating clients
to adhere to the treatment. The garden adopts the aspects of playful
interaction, collection and progression as key elements of the game. Patients
are encouraged to expand and grow their own garden of positive self-images,
symbolized as plants and flowers, by collecting and *planting* positive
self-referenced resources (stories, images, audios, etc.) to progressively add
to the flowers. By later revisiting and reviewing them guided by a narrator
voice throughout the therapy stages, the plants would blossom and grow bigger
and the patients would strengthen the (counter)memories associated with their
positive self-images. Overall, it is expected that this application would help
supporting and improving the efficiency of the treatment in a blended format.
Patients would autonomously engage with a proportion of the therapy outside the
clinical setting, therefore empowering their self-reliance and self-management,
but also reducing therapeutic costs in terms of time and effort for both the
patient and the staff and health care institution.

Goal of this first pilot study is to explore the feasibility, acceptability and
indications of clinical effects of the blended version of the COMET protocol
for self-esteem, in comparison with the standard protocol of COMET therapy.
While the standard COMET therapy involves 7 face-to-face sessions, the blended
version combines a lower amount of face-to-face sessions (i.e., 3 instead of 4)
with the use of the Zen Garden mobile application throughout the remaining 4
sessions. The study will explore if the blended COMET treatment has the
potential to increase treatment adherence (i.e., successful completion of
homework assignments) and achieve positive therapeutic effects (increase in
self-esteem and decrease in anxiety and depressive symptoms) with a less
intensive format (i.e., fewer face-to-face contacts), compared with the
standard, more intensive face-to-face COMET protocol.

Specifically, the study will address the following research questions:
1) What is the adherence and retention rate to the blended COMET protocol in
comparison to the standard version (e.g., amount of completed homework
exercises and drop-out rate)?
2) What is the patients* experience and acceptability with the blended COMET
therapy in comparison to the standard COMET therapy?
3) What is the patients* experience with the blended COMET therapy and the
mobile app?
4) What are the effects of the blended COMET protocol on self-esteem?
5) What are the secondary effects of the blended COMET protocol on depression
symptoms?

The study is a mixed-method, feasibility randomized clinical trial with a
two-group parallel design. Patients are randomized over the two intervention
conditions: one group will receive the standard COMET intervention for low
self-esteem (Korrelboom et al., 2009, 2011, 2012), consisting of 7 face-to-face
individual sessions with a trained therapist; and one group will receive the
blended version of the COMET, with fewer face-to-face sessions (session 1, 5,
and 7) in conjunction with the independent use of the Zen Garden mobile app
during homework weeks (weeks 2, 3, 4, and 6). Therapists involved in the study
will deliver both therapy versions.

Before and at conclusion of the intervention, participants will complete a
brief assessment session, including measures of self-esteem, depression and
anxiety symptoms. Treatment acceptability and satisfaction will be assessed
during the post-intervention assessment session. Patients assigned to the
blended COMET group will also report about their user experience with the
mobile app. Patients activity on the app during the intervention phase will be
automatically recorded on the app, while weekly forms about homework exercises
will be given to the patients receiving the standard COMET and reviewed with
the therapist during their regular therapy sessions.

Nested within the main quantitative study, we will explore app users* views of
the acceptability, usability and potential effectiveness of the app and its
integration within the therapy, in qualitative interviews. At conclusion of the
post-assessment session, a selected sub-sample of patients assigned to the
blended COMET group will be further invited to participate in individual
interviews. The nested qualitative study is conducted concurrently with the
quantitative study (i.e., *QUAN + qual* concurrent explanatory mixed-method
design; Bishop, 2014) and is aimed to complement the quantitative findings and
to capture different facets of patients* experience with the app and the
blended treatment (e.g., how, when and why patients engaged with specific
features of the app; what aspects of the app need revision or improvements;
what is the experience with using an app in combination with the face-to-face
session; how was the patients' experience with the integration of the app in
the therapy stages).

COMET protocols for sef-esteem primarily involve 1) the identification of the
negative self-opinion, 2) the identification of positive characteristics and
behaviors that are incompatible with the negative self-opinion, and 3)
enhancing the retrievability and experiencing of these positive
characteristics.
Both the standard and the blended version of the COMET therapy are fully
manualized for both therapists and patients. At the start of the therapy, all
patients are provided with a comprehensive treatment manual in which the
rationale, background and specific procedures, and homework assignments of the
treatment are specified. The patient manual for the blended COMET protocol
includes the integration of the app use into the therapy steps and an
additional *tutorial* on how to use the app.
Therapists will fill in a checklist including a list of treatment procedures
detailed for each COMET session (cf. Staring et al., 2016). The researchers
will then screen the checklists to evaluate treatment integrity.

The standard COMET
The standard COMET therapy lasts 7 sessions of 45-60 minutes each and consists
of several steps, all aimed at strengthening the patient*s positive
self-opinions and making them more competitive. All steps are repetitively
practiced during the therapy sessions and in homework assignments. First, the
negative self-image is identified. Next, a credible but incompatible and more
positive self-image, based on personal characteristics, acts, and experiences,
is formulated. This positive self-image is strengthened by writing small
self-referent stories about real-life instances in which these positive
personal characteristics had been active. During the next few sessions, these
instances are made more emotionally salient by imagining these scenes, backing
them up with positive self-verbalizations (session 2), body posture and facial
expression (session 3), and music (session 4); all these are selected by the
patients themselves and believed to promote the experience of positive
self-esteem. Then, in sessions five and six counter-conditioning is introduced
in order to form new associations between cues that formerly activated low
self-esteem and the newly enhanced representations of positive self-esteem. In
the final (seventh) session, after discussion of the latest homework, the
therapy progress and achieved goals are evaluated and the generalization of the
new representations of positive self-esteem in the future is discussed.

The blended COMET
The blended version of the COMET therapy involves the use of the Zen Garden app
along the therapy steps, involving the completion of progressive exercises and
homework to strengthen the activation and rehearsal of the positive
self-referent events chosen by the patient. During the first COMET session, the
therapist introduces the app to the patient and its main goal. Upon logging in
for the first time, patients are asked to enter maximum five different negative
self-esteem elements. Each of the negative self-images would act as an *area*
in the Zen Garden. Positive self-images counteracting the negative self-esteem
evaluations are planted as flowers in these areas to help participants build a
network of positive self-evaluations and images. To plant a flower,
participants need to enter a credible statement of their positive self-image
(i.e. *I regularly help other people*). The plants could be grown further by
progressively attaching self-referenced resources (i.e., self-reference
stories, images and audios) to the plants, similarly to what is done along
steps 2-4 of the standard COMET protocol.
In each week, participants are provided with a task list, which will help in
guiding them on their goals for the therapy. Notifications are sent
periodically through push notification on the mobile phone, to remind patients
about completing the therapy tasks. In addition, a voice narrator in the form
of a virtual avatar is present in the application and will guide users through
the various therapy stages (e.g., explaining the importance of each step in the
therapy, giving instructions on what to do, etc.).
In order to review their positive self-images, patients will need to click on
the previously added positive self-image resources, which are attached to the
flowers in the form of *spirits* floating around them. The narrator voice will
guide participants through the review process, adopting a narrative style that
is meditative in nature (e.g., *close your eyes* try to visualize the memories*
etc.). This is done to help strengthen the memories of their positive
self-images.
The three face-to face sessions in the blended COMET protocol last between 45
and 60 minutes each, except for the first one, which lasts between 60 and 90
minutes to allow the patient to get acquainted with the app together with the
therapist. In the first session, the patient will install the app and will then
be further familiarized with the apps* functions by the therapist. Together,
they will identify the negative self-image(s) and enter it (or them) as an area
in the Zen Garden app, followed by planting the first flower symbolizing the
first self-referent positive image/story. The first session will be concluded
with an overview of the tasks the patient can expect for the following weeks.
The second session is scheduled four weeks after the first session. By that
time, the patients will have practiced with self-referent stories, imagination,
pictures, self-verbalization, body posture, facial expression and music using
the Zen Garden app. The patient will share the *garden* of positive images in
the app with the therapist at the beginning of the second session (session 5 in
the standard COMET protocol). The therapist and the patient will discuss the
progress the patient made with the app and any hardship the patient may have
experienced while doing the daily exercises. In the second part of the session,
the therapist will continue with the standard content of the session, i.e., the
counter-conditioning. Counter-conditioning is used to form associations between
cues that previously triggered low self-esteem and the enhanced positive images
the patients have been practicing. Patients will continue practicing the
counter-conditioning exercises at home while using the app. The second session
will be concluded with an overview of the tasks the patient can expect for the
following week.
During the third and last session of the blended COMET protocol, the patient
will share the entire garden with the therapist as to evaluate the progress
that has been made. At conclusion, the generalization of the new positive
self-esteem in the future will be discussed.

All clients participating in the study will receive an active treatment
alongside their treatment as usual. Clients randomized to the blended COMET
condition will have access to a therapy-supportive app that could enhance their
therapy progression and adherence (i.e., homework) while reducing the intensity
and costs of weekly face-to-face sessions. The results of this study will also
provide insights in the acceptability and experience, app usage pattern and
relation with the therapy progression and goals of a blended, less intensive
therapy protocol, compared to the standard, more intensive protocol.

The burdens that are connected to participation in this study may impose some
extra work to patients, which is yet minimal and fully related to the treatment
goals. In addition to their regular therapy, patients will only be asked to
fill in a few questionnaires at the beginning and end of the study. Patients
are under the consistent care and monitoring of their therapist, including
agreed weekly contacts outside of therapy sessions. This is particularly
relevant for clients assigned to the blended COMET condition during the
*homework* weeks of independent use of the app (weeks 2, 3, 4, and 6), where
fewer therapist-patient face-to-face contacts will occur.
A less intensive face-to-face therapy protocol may increase the risk of
sub-optimal therapy outcome(s) or rebound of negative symptoms, due to the
lower amount of direct contact with the therapist, fewer formal therapy
sessions and their substitution with an m-health tool. If a patient shows signs
of intense distress or exacerbation of pathological symptoms the therapist will
inform the research staff and the treating specialist, the patient will be
immediately withdrawn from the study and, if deemed appropriate by the treating
specialist, start over with the standard COMET intervention. Furthermore,
clients receiving the blended COMET intervention are free to request and
discuss with their therapist the possibility of receiving the full COMET
intervention at conclusion of the study.

The burden for therapists involved in the study is compensated by the reduced
intensity of the blended COMET protocol. Although being involved in the study
requires the therapists to carry out a list of extra tasks, the delivery of the
blended COMET protocol includes only 3 out of the 7 standard face-to-face
sessions, hence saving around four hours for each patient in the blended COMET
group. These hours are then used to compensate for the time spent on the extra
activities.
All therapists will deliver both versions of the COMET protocol. Before
starting the study, they will be trained to integrate and use the app within
the blended COMET protocol by the main COMET designer, Prof. Kees Korrelboom,
and the main researcher at PsyQ, who is experienced with the Zen Garden app.
Besides delivering the therapy, extra tasks the therapists are asked to do
involve proposing the study to potential patients, collecting the weekly diary
forms from clients in in the standard COMET therapy, and extracting the routine
summary about each session/contact with the clients. Therapists can always
refer tot the researcher in place at PsyQ who will support them across all the
tasks they are required to do in the study procedure.