ORIGINAL RESEARCH

published: 17 May 2019

doi: 10.3389/fpsyg.2019.01099

Effects of Mindfulness Based

Cognitive Therapy (MBCT) and
Compassion Focused Therapy (CFT)
on Symptom Change, Mindfulness,
Self-Compassion, and Rumination in
Clients With Depression, Anxiety, and
Stress
Anna Dora Frostadottir 1* and Dusana Dorjee 2,3
1
MSc in Mindfulness-Based Approaches, Director of the Mindfulness Centre in Reykjavik, Reykjavik, Iceland, 2 Department of
Education, Psychology in Education, University of York, York, United Kingdom, 3 School of Psychology, Bangor University,
Bangor, Wales

Objectives: Over the past decade there has been an increasing interest in exploring
self-compassion as a related and complementary construct to mindfulness. Increases
in self-compassion may predict clinical outcomes after MBCT and cultivation of
Edited by: compassion toward self and others is central to CFT. This pilot study compared
Joshua Fredrick Wiley, the impact of MBCT applying implicit self-compassion instructions and CFT
Monash University, Australia
employing explicit self-compassion instructions on symptom change, mindfulness,
Reviewed by:
Andrew James Greenshaw,
self-compassion, and rumination.
University of Alberta, Canada Method: This non-randomized wait-list controlled study (N = 58) with two intervention
Paul Gilbert,
NHS England, United Kingdom arms (MBCT N = 20, CFT N = 18, Control N = 20) assessed the outcomes of clients
*Correspondence: with depression, anxiety, and stress symptoms from before to after the interventions and
Anna Dora Frostadottir at one month follow up (MBCT N = 17, CFT N = 13, Control N = 13).
annadora@nuvitundarsetrid.is
Results: Both treatments resulted in significant increases in mindfulness and
Specialty section: self-compassion and decreases in rumination, depression, anxiety, and stress.
This article was submitted to
Furthermore, MBCT enhanced mindfulness for people who were initially high in
Psychology for Clinical Settings,
a section of the journal rumination, whereas CFT enhanced mindfulness across the board.
Frontiers in Psychology
Conclusion: The findings suggest that both MBCT and CFT, and hence implicit
Received: 31 August 2018
Accepted: 26 April 2019
or explicit self-compassion instructions, produce similar clinical outcomes with CFT
Published: 17 May 2019 enhancing mindfulness regardless of client’s rumination level.
Citation:
Keywords: mindfulness, self-compassion, rumination, depression, anxiety, stress, MBCT, CFT
Frostadottir AD and Dorjee D (2019)
Effects of Mindfulness Based
Cognitive Therapy (MBCT) and
Compassion Focused Therapy (CFT)
INTRODUCTION
on Symptom Change, Mindfulness,
Self-Compassion, and Rumination in Mindfulness meditation practices are increasingly being incorporated into clinical treatments for
Clients With Depression, Anxiety, and a variety of mental health problems with positive results in reducing emotional distress and
Stress. Front. Psychol. 10:1099. promoting psychological well-being (Hofman et al., 2010; Keng et al., 2011; Piet and Hougaard,
doi: 10.3389/fpsyg.2019.01099 2011; Goyal et al., 2014). Mindfulness has its roots in Buddhism and is most often defined as

Frontiers in Psychology | www.frontiersin.org 1 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

“the awareness that emerges through paying attention on module which aimed at cultivating present-moment-focused
purpose, in the present moment, and non-judgmentally to the attention and body awareness, the “Affect” module which aimed
unfolding of experience moment by moment” (Kabat-Zinn, at cultivating loving-kindness, gratitude and compassion and the
2003, p. 145) in the secular therapeutic context. The two most “Perspective” module which aimed at cultivating meta-cognitive
widespread mindfulness-based programs are Mindfulness-Based skills. Their findings indicated that present-moment mindfulness
Stress Reduction (MBSR; Kabat-Zinn, 1990) and Mindfulness- practices were not sufficient and only explicit socio-affective
Based Cognitive Therapy (MBCT; Segal et al., 2002). The practices were able to increase self-compassion.
MBSR is intended for a wide range of clinical and non-clinical Contrary to that finding and despite the implicit nature in
populations with promising cumulative evidence for its effects mindfulness practices in MBSR and MBCT, research has found
(Grossman et al., 2004; Chiesa and Serretti, 2009; Bohlmeijer that participation in MBSR (Bergen-Cico and Cheon, 2013) and
et al., 2010). MBCT is based on MBSR and also integrates MBCT (Rimes and Wingrove, 2011) leads to significant increases
cognitive approach and instructions. It was developed as a in both mindfulness and self-compassion. Looking closer into the
relapse prevention for people with recurrent depression and mechanisms of self-compassion in MBCT, Kuyken et al. (2010)
has been found to reduce the risk of depression relapse by investigated the effects of MBCT compared to maintenance
approximately half (Teasdale et al., 2000; Ma and Teasdale, antidepressant medication (m-ADM) on mindfulness, self-
2004). It has also been shown effective for people dealing with compassion, cognitive reactivity, and relapse risk for depression.
anxiety, stress, irritability, and exhaustion (Hofman et al., 2010; They found that increased mindfulness and self-compassion
Khoury et al., 2013). mediated the positive effect of MBCT on depressive symptoms
Aside from mindfulness, another construct stemming from at 15 months follow-up. They also found that cognitive reactivity
Buddhist psychology that is also being increasingly incorporated was higher for MBCT participants compared to the m-ADM
into meditation practices in clinical treatments is compassion group, but only predicted poorer outcome for the m-ADM
toward self and others (Germer, 2009; Gilbert, 2009a; Neff, group, and not for the MBCT group. Further exploration of their
2011). The standard definition of compassion is derived from findings indicated that increased self-compassion moderated and
the writings of the Dalai Lama (1995) who defined it as “a reduced the link between cognitive reactivity and relapse risk
sensitivity to suffering in self and others with a commitment to in the MBCT group, serving as an important protective factor.
try to alleviate and prevent it.” In the therapeutic context the Based on this finding the second edition of the MBCT manual
construct of self-compassion is often used as defined by Kristin now explicitly states that MBCT aims to cultivate mindfulness
Neff (2015). According to Neff, self-compassion involves meeting and self-compassion (Segal et al., 2013).
ourselves with warmth and understanding when we suffer rather In contrast to the implicit aspects of self-compassion in
than ignoring our pain or criticizing ourselves—just as we would mindfulness practices in MBSR and MBCT, the practices of
treat a good friend. self-compassion meditation (Germer, 2009; Gilbert, 2009a;
Research has shown that self-compassion is associated with Neff, 2011) explicitly aim to cultivate compassion toward self
decreased psychological distress and increased positive affect and others. One approach incorporating explicit compassion
(Neff and Dahm, 2014). A recent meta-analysis by MacBeth instructions is Compassion Focused Therapy developed
and Gumley (2012) documented a large effect size for the by Gilbert and colleagues (CFT; Gilbert, 2009a). CFT is
relationship between greater self-compassion and lower levels an “integrated and multimodal approach that draws from
of mental health problems such as depression, anxiety, and evolutionary, social, developmental and Buddhist psychology,
stress. More self-compassionate people also report lower levels and neuroscience” (Gilbert, 2009b, p. 199). It was originally
of rumination (Neff, 2003a) and self-criticism (Ehret et al., developed for people with long-term emotional problems,
2015), which are known risk factors for depression and anxiety often associated with high levels of shame and self-criticism,
(Blatt, 1995; Nolen-Hoeksema, 2000); indicating that self- and delivered as an individual therapy. The CFT seeks to
compassion might be a possible emotion regulation strategy for help individuals develop compassion for self and others
emotional difficulties. (Gilbert, 2017). It involves educating them about how the
In the context of MBSR and MBCT, the cultivation of brain operates in terms of three types of emotion-regulation
self-compassion is implicit and an indirect intention of the systems; the threat system, the drive system, and the soothing
programs. In fact, Segal et al. (2002) who developed MBCT advise system. The CFT suggests that self-compassion deactivates the
mindfulness teachers against explicitly discussing or teaching threat system and activates the self-soothing system (Gilbert
self-compassion in the program. Rather, they suggest that and Irons, 2005). Although there are many overlapping
participants learn the principles of self-compassion implicitly features of CFT and MBCT, i.e., cultivating mindfulness, body
by the kind and compassionate embodiment of the teachers awareness, and grounding mentalization training and the
(Segal et al., 2013). Indeed, according to Germer and Barnhofer use of psychoeducation, it is very clear that MBCT puts the
(2017) the second part of Kabat-Zinn’s definition of mindfulness, primary focus on cultivating mindfulness whereas CFT puts it
non-judgemental acceptance, can be taken as indicating that on cultivating compassion toward self and others.
compassion toward self and others and mindfulness are Although CFT for groups has not yet been manualized. Gilbert
intrinsically linked. However, a recent study by Hildebrandt and Procter (2006) developed the first group-based version of
et al. (2017) explored the effects of three different 3-month CFT for patients with severe long-term difficulties in a cognitive-
mental training modules on self-compassion, The “Presence” behavioral-based day center. Following 12 weekly sessions, they

Frontiers in Psychology | www.frontiersin.org 2 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

found significant reductions in depression, anxiety, self-criticism, TABLE 1 | Demographic variables for participants in the three arms of the study.
shame, inferiority, and submissive behavior. Subsequent studies
MBCT CFT Controls
explored the use of group-based CFT in other clinical settings.
Similar positive effects have also been found for people with PRE-POST DATA (n = 58)
psychosis (Laithwaite et al., 2009; Braehler et al., 2013), people Total number, n 20 18 20
with personality disorder (Lucre and Corten, 2013), people with Gender, women: n (%) 19 (95%) 15 (83%) 17 (85%)
eating disorders (Goss and Allan, 2010; Gale et al., 2012; Kelly Age (in years): M (SD) 49 (11.05) 53 (10.20) 51 (9.26)
et al., 2016) and people with long-term and severe mental health PRE-POST & FOLLOW-UP DATA (n = 43)
problems (Heriot-Maitland et al., 2014). Total number, N 17 13 13
Due to the different instructions in mindfulness and self- Gender, women: n (%) 16 (94%) 10 (77%) 12 (92%)
compassion/compassion practices it is reasonable to assume
Age (in years): M (SD) 50 (11.38) 52 (10.07) 49 (8.10)
that they have different impact on people (Germer, 2009).
Researchers have begun to explore the question of “what for
whom”, i.e., to match a practitioner with either mindfulness
or compassion training. In relation to that, the findings from
acupuncture, and massage. It offers two kinds of psychotherapies,
a study conducted by Barnhofer et al. (2010) suggested that
MBCT, and CFT. The participants were recruited through
those experiencing higher levels of rumination may benefit from
convenience sampling; the treatment participants were allocated
meditations focusing on mindfulness, whereas people with lower
to the intervention starting at the time they began their
levels of rumination may benefit more from loving kindness
rehabilitation. Clients suffering from mild to moderate anxiety,
practices. We need to keep in mind that traditional Buddhist
depression and/or stress symptoms and able to complete the
literature clearly distinguishes loving kindness and compassion
4-week intervention were informed about the research during
(Wallace, 1999), however loving-kindness practices use explicit
their orientation day. Up to 15 clients are typically included in
instructions to induce compassion toward self and others.
each treatment group at the clinic and treatment participants
Cumulatively, the outlined evidence suggests that
were therefore allocated to two MBCT and two CFT groups.
mindfulness-based programs, even though they cultivate
Complete pre-post data sets were obtained from 58 participants,
self-compassion implicitly, are effective in reducing emotional
20 from the MBCT group, 18 from the CFT group, and 20 from
distress in part because they cultivate self-compassion. There
the control group. A complete pre-post and follow-up data was
is also initial evidence that group CFT, a program targeting
obtained from 43 participants, 17 from the MBCT group, 13 from
development of compassion toward self and others explicitly, can
the CFT group, and 13 from the control group. Table 1 shows the
result in reductions in anxiety, depression, and other symptoms
distribution of gender and age across the three arms of the study.
similar to the effects of MBCT, yet it is not clear if these effects
After removing outliers the complete pre-post data for all four
are mediated by self-compassion. The current study aimed
measures was from 54 participants, 19 from the MBCT group,
to directly compare the effects of implicit self-compassion
17 from the CFT group, and 18 from the control group. The
training applied in MBCT and explicit self-compassion training
complete pre-, post-, and follow-up data for all four measures was
employed in CFT on symptom change, mindfulness, self-
from 32 participants, 13 from the MBCT group, 12 from the CFT
compassion, and rumination. We predicted that both treatments
group, and 7 from the control group.
would increase self-compassion and mindfulness as well as
reduce symptoms and rumination, but gains in self-compassion
would be greater after CFT in comparison to MBCT. If this later Procedure
prediction was supported, this may indicate that mindfulness- The study was approved by the ethics committee in the School
based programs would benefit from incorporating explicit of Psychology at Bangor University in the UK as well as
self-compassion practices into them. Furthermore, given the from the Ethic Committee at the Directorate of Health in
preliminary evidence suggesting that people respond differently Iceland. Treatment participants were asked to complete the
to meditation practices based on their tendency to ruminate, we pre- and post-treatment questionnaires at the beginning of
wanted to explore whether CFT and MBCT differentially impact the first session and at the end of the last session in the
on participants’ outcomes depending on their baseline levels of MBCT and CFT groups. At the follow-up assessment 1 month
rumination. It could be expected that participants with higher after treatment participation, participants completed follow-up
baseline rumination benefit more from MBCT and those with questionnaires electronically. Participants in the control group
lower rumination from CFT. also completed the questionnaires at the same time points as
treatment participants electronically.
METHODS The treatments were delivered by two experienced therapists,
one of them is a clinical psychologist and the other is a psychiatric
Participants nurse. They have both been trained in MBCT at Bangor
The study was conducted in a residential rehabilitation and University and in CFT by Paul Gilbert. The clinical psychologist
health clinic in Iceland, where clients usually stay for 4 weeks. was also under Paul Gilberts’ supervision. To ensure quality of the
The clinic offers a holistic approach in treating their clients with treatments they delivered, they had regular supervision sessions
a variety of programs, e.g., psychoeducation, exercise, nutrition, and provided each other with feedback in terms of adherence to

Frontiers in Psychology | www.frontiersin.org 3 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

the treatments. However, no other formal external examination self-attention, the tendency to rehash, re-evaluate or dwell on past
was put into place to ensure the fidelity of the treatments. events or experiences. Trapnell and Campbell (1999) reported
MBCT was delivered following the treatment protocol high internal consistency of the RRQ (α = 0.94).
(Segal et al., 2002) modified for four weeks: it included Depression Anxiety and Stress Scales—Short Form (DASS-21;
eight two-hour-long sessions over four consecutive weeks (two Lovibond and Lovibond, 1995) evaluates the severity symptoms
sessions per week). Session content included guided mindfulness of Depression, Anxiety, and Stress. It consists of 21 items in which
practices (i.e., body scan, yoga, sitting meditations of breath, individuals are required to indicate the presence of a symptom
body, sound, thoughts, and exploring difficulties); enquiry over the past week from 0 (did not apply to me at all over
into participants’ experiences of these practices; homework the last week) to 3 (applied to me very much or most of the
review and teaching/discussion of cognitive skills modified for time over the past week). Henry and Crawford (2005) reported
depression, anxiety, and stress. An adequate dose of MBCT was good discriminant and convergent validity for DASS-21 and high
defined as participation in at least four of the eight MBCT group internal consistency for depression (α = 0.88), anxiety (α = 0.82),
sessions as recommended (Segal et al., 2002). and stress (α = 0.90) scales.
CFT was delivered following the generic treatment outline To date, normative data has not been created for an Icelandic
(Gilbert, 2009a) which the clinical psychologist based on the population for these measures. Therefore, all four questionnaires
MBCT manual (Segal et al., 2002) under Gilberts’ supervision: have been translated into Icelandic and then back-translated to
it included eight 2-h long sessions over 4 consecutive confirm the accuracy of the initial translation. It was expected
weeks (two sessions per week). Session content included that Icelandic participants would respond in similar ways to
guided mindfulness, compassion, and self-compassion practices; Western participants for which the norms have been established.
enquiry into participants’ experiences of these practices; review
of homework; imagery; videos; experiential exercises; and
teaching/discussion of the three interconnecting emotion- Data Analysis
regulation systems. An adequate dose of CFT was defined as All data analyses were conducted using the Statistical Package
participation in at least four of the eight CFT group sessions. for the Social Science (SPSS), version 22.0 (SPPS Inc, Chicago,
The control group was offered to attend MBCT or CFT after IL, USA). Extreme outliers were removed from the data and the
the intervention groups completed their programs. non-extreme outliers were winsorized on the upper and lower
bounds of the group (95%) on a measure by measure basis.
Research Design Internal consistency was calculated for each measure. Between-
This pilot study followed a pre-post design with a control group group comparisons at the pre-treatment time for all the measures
and two intervention arms. The assessments were conducted at examined possible baseline differences. Correlations among the
the baseline, after the 4-weeks long interventions and at 1 month dependent measures at pre-treatment time were also investigated
follow up. Participants who were in the treatment groups either for expected convergent/divergent patterns. A 2 (time: pre, post)
received the 4-week MBCT or the 4-week CFT. × 3 (Group, MBCT, CFT, Control) mixed ANOVA examined
changes in the scores of the dependent measures from pre to
Measures post. A 3 (time: pre, post, follow-up) × 3 (group: MBCT, CFT,
Five-Facets of Mindfulness (FFMQ; Baer et al., 2006) consists Control) mixed ANOVA with a smaller sample due to attrition
of 39 times, rated on a five-point Likert-scale, assessing the five at the latter two time points examined changes in the scores
facets of mindfulness of Observing, Describing, Non-judging of from pre, to post and follow-up. Significant interactions for all
inner experience, Acting with awareness, and Non-reactivity to the measures were investigated further for predicted differences
inner experience. Carmody and Baer (2008) reported that the using paired-samples t-tests.
FFMQ subscales are sensitive to change and have high internal
consistency (α = 0.75–0.91).
Self-Compassion Scale (SCS; Neff, 2003b) consists of 26 items, RESULTS
rated on a five-point Likert-scale, assessing the positive and
negative aspects of the three main factors of self-compassion: Self- The DASS had high internal consistency for the Depression
Kindness vs. Self-Judgement; Common Humanity vs. Isolation; (α = 0.84), Anxiety (α = 0.77), and Stress (α = 0.79) subscales.
and Mindfulness vs. Over Identification. Birnie et al. (2010) The FFMQ had high internal consistency for the Observing (α =
reported high internal consistency for SCS subscales (α = 0.77– 0.77), Describing (α = 0.75), Acting with Awareness (α = 0.83),
0.81) and overall high convergent and discriminant validity. Nonjudging (α = 0.90), and Nonreacting (α = 0.73) subscales.
This measure has been criticized because of its factor structure The SCS had high internal consistency for the Self-Kindness (α =
(Williams et al., 2014; Muris et al., 2016), however, it was chosen 0.82), Self-Judgement (α = 0.81), Common Humanity (α = 0.72),
to provide comparative insights into the effect sizes of self- Isolation (α = 0.84), and Over-identifying (α = 0.77) subscales
compassion. It is worthwhile to note its overlap with measuring but poor consistency for Mindfulness (α = 0.66) subscale. The
mindfulness as it has a mindfulness subscale. RRQ had high internal consistency (α = 0.91).
Reflection Rumination Questionnaire (RRQ: Trapnell and There were no significant differences between the groups
Campbell, 1999) consists of 12 items, rated on a five-point at pre-treatment time on any of the dependent measures, all
Likert-scale, assessing three aspects of rumination: ruminative ps > 0.05

Frontiers in Psychology | www.frontiersin.org 4 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

TABLE 2 | Intercorrelations of outcome measures at pre-intervention time (N = 57).

Variable DASS total DASS-D DASS-A DASS-S FFMQ SCS RRQ

DASS total 0.87** 0.81** 0.86** −0.58** −0.67** 0.56**

DASS-D 0.52** 0.63** −0.51** −0.63** 0.37**
DASS-A 0.57** −0.42** −0.47** 0.44**
DASS-S −0.52** −0.58** 0.63**
FFMQ 0.74** −0.60**
SCS −0.59**
RRQ

DASS total, total scores of Depression Anxiety and Stress Scales; DASS-D, scores for the Depression subscale of DASS; DASS-A, scores for the Anxiety subscale of DASS; DASS-S,
scores for the Stress subscale of DASS; FFMQ, Five Facet Mindfulness Questionnaire; SCS, Self-Compassion Scale; RRQ, Reflection Rumination Questionnaire. **p < 0.01.

Correlations among the dependent measures at pre-treatment = 7.40, p < 0.002) was significant. As shown in Table 3,
time are shown in Table 2. All the relationships were in the participants in the MBCT group had significantly higher post-
expected directions. treatment SCS scores t (18) = −3.92, p < 0.001, d = −0.67,
95% CI [−17.38, −5.26]. Participants in the CFT group also
Pre-post Evaluations had significantly higher post-treatment SCS scores t (16) =
DASS −2.20, p < 0.05, d = −0.39, 95% CI [−10.40, −0.19]. No
The main effect of time F (1, 51) = 49.04, p < 0.001) and the significant change was found in SCS scores in the control
main effect of group F(2,51) = 3.90, p < 0.03) were significant. group (p > 0.05).
The interaction of time by group F (1, 51) = 6.67, p < 0.001)
was also significant. As shown in Table 3, participants in the RRQ
MBCT group had significantly lower post-treatment total DASS The main effect of time was significant F (1, 51) = 17.95, p < 0.001)
scores t (18) = 5.72, p < 0.001, d = 1.65, 95% CI [9.92, 21.45], and the main effect of group was not significant F (2, 51) = 0.08, p
and significantly lower post-treatment scores on the depression > 0.5). The interaction of time by group F (1, 51) = 7.40, p < 0.002)
t (18) = 4.19, p < 0.001, d = 1.11, 95% CI [2.50, 7.50], anxiety was significant. As shown in Table 3, participants in the MBCT
t (18) = 4.34, p < 0.001, d = 1.11, 95% CI [1.74, 5.00], and group had significantly lower post-treatment RRQ scores t (18) =
stress t (18) = 6.62, p < 0.001, d = 2.06, 95% CI [4.99, 9.64] 3.73, p < 0.002, d = 0.87, 95% CI [3.49, 12.51]. Participants in
subscales of DASS. Participants in the CFT group also had the CFT group also had significantly lower post-treatment RRQ
significantly lower post-treatment total DASS scores t (16) = 6.91, scores t (16) = 3.49, p < 0.003, d = 0.67, 95% CI [1.98, 8.14].
p < 0.001, d = 1.98, 95% CI [8.40, 15.84], and significantly No significant change was found in RRQ scores in the control
lower post-treatment scores on the depression t (16) = 6.53, p group (p > 0.05). The mean scores for pre- and post-
< 0.001, d = 1.59, 95% CI [3.14, 6.16], anxiety t (16) = 3.05, treatment scores on the dependent measures for participants
p < 0.008, d = 0.96, 95% CI [0.57, 3.19], and stress t (16) = in the treatment groups and the control group are shown
5.78, p < 0.001, d = 1.85, 95% CI [3.54, 7.64] subscales. No in Figure 1.
significant change was found in DASS scores in the control group
(all ps > 0.05).
Pre, Post, Follow-Up Evaluations
DASS
FFMQ The main effect of time F (2, 58) = 21.85, p < 0.001) and the
The results revealed a significant main effect of time F (1, 51) = main effect of group F (2, 29) = 5.69, p < 0.01) were significant.
30.80, p < 0.001) and a significant main effect of group F (2, 51) The interaction of time by group F (4, 58) = 6.20, p < 0.01)
= 0.01, p > 0.05). The interaction of time by group F (1, 51) was also significant. As shown in Table 4, participants in the
= 7.40, p < 0.002) was also significant. As shown in Table 3, MBCT group had significantly lower post-treatment total DASS
participants in the MBCT group had significantly higher post- scores t (12) = 5.26, p < 0.001, d = 1.86, 95% CI [9.77, 23.61],
treatment FFMQ scores t (18) = −3.86, p < 0.001, d = −1.01, and lower post-treatment scores on the depression t (12) = 4.09,
95% CI [−24.05, −7.11]. Participants in the CFT group also had p < 0.001, d = 1.35, 95% CI [2.77, 9.08], anxiety t (12) =
significantly higher post-treatment FFMQ scores t (16) = −5.64, 4.16, p < 0.001, d = 1.39, 95% CI [1.79, 5.75], and stress
p < 0.001, d = −0.96, 95% CI [−18.30, −8.29]. No significant t (12) = 5.40, p < 0.001, d = 1.96, 95% CI [4.18, 9.82]. These
change was found in FFMQ scores in the control group reductions were maintained at follow-up as the participants
(p > 0.05). had significantly lower follow-up total DASS scores than at
pre-test t (12) = 5.06, p < 0.001, d = 1.78, 95% CI [9.11,
SCS 22.89], and similarly significantly lower follow-up scores on the
The main effect of time was significant F (1, 51) = 17.95, p < depression t (12) = 3.96, p < 0.002, d = 1.35, 95% CI [2.76,
0.001) and the main effect of group was not significant F (2, 51) 9.54], anxiety t (12) = 3.83, p < 0.002, d = 1.39, 95% CI [1.66,
= 0.08, p > 0.5). The interaction of time by group F (1, 51) 6.04], and stress t (12) = 5.65, p < 0.000, d = 1.79, 95% CI [3.69,

Frontiers in Psychology | www.frontiersin.org 5 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

TABLE 3 | Pre- and post-test scores for the MBCT, CFT, and control groups analyzed with paired-samples t-tests.

MBCT CFT Controls

(N = 19) (N = 17) (N = 18)

Outcome Pretest Posttest Pretest Posttest Pretest Posttest

M (SD) M (SD) M (SD) M (SD) M (SD) M (SD)

DASS total 26.74 (11.81) 11.05 (6.40)** 20.59 (6.88) 8.47 (5.23)** 23.44 (12.03) 20.39 (11.75)
Depression 9.11 (5.42) 4.11 (3.30)** 7.94 (3.36) 3.29 (2.59)** 9.33 (5.04) 8.00 (4.90)
Anxiety 5.68 (3.87) 2.32 (1.83)** 3.06 (2.44) 1.18 (1.29)** 5.17 (4.44) 4.11 (3.76)
Stress 11.95 (4.22) 4.63 (2.73)** 9.59 (3.32) 4.00 (2.69)** 8.94 (4.54) 8.28 (4.76)
FFMQ 111.47 (18.94) 127.05 (10.93)** 112.41 (14.33) 125.71 (13.34)** 119.50 (11.04) 119.83 (15.22)
SCS 71.42 (18.98) 82.74 (14.60)** 73.82 (13.60) 79.12 (13.87)* 75.00 (14.27) 75.56 (14.40)
RRQ 44.63 (10.70) 36.63 (7.41)** 44.82 (6.30) 39.76 (8.57)** 40.56 (9.12) 41.11 (10.05)

M, mean; SD, standard deviation.

T-test significant at *p < 0.05 and **p < 0.001.

TABLE 4 | Pretest, post-test, and follow-up mean scores for the MBCT, CFT, and control groups analyzed with paired sample t-tests.

MBCT CFT Controls

(N = 13) (N = 12) (N = 7)

Outcome Pretest Posttest Follow-up Pretest Posttest Follow-up Pretest Posttest Follow-up
M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD)

DASS total 29.23 (11.48) 12.54 (5.44)** 13.23 (5.43)** 19, 17 (7.30) 8.33 (6.24)** 10.67 (4.40)* 21.14 (10.85) 22.43 (6.71) 19.43 (6.05)
Depression 10.69 (5.41) 4.77 (3.00)** 4.54 (3.53)* 7.33 (3.47) 3.25 (2.80)** 4.08 (2.68)* 8.57 (5.29) 8.43 (3.74) 5.86 (3.24)
Anxiety 6.31 (3.43) 2.54 (1.71)** 2.46 (1.90)* 2.92 (2.54) 1.50 (1.38) 1.50 (1.68) 3.71 (3.35) 3.86 (3.13) 3.57 (2.76)
Stress 12.23 (4.38) 5.23 (2.52)** 6.23 (1.83) ** 8.92 (3.50) 3.58 (3.00)** 5.08 (2.75)* 8.86 (4.18) 10.14 (2.91) 10.00 (4.28)
FFMQ 108.92 (15.06) 125.69 (11.09)** 126.85 (10.29)** 114.92 (14.82) 126.83 (12.19)** 124.92 (9.34)* 115.71 (8.22) 110.14 (14.38) 111.71 (12.89)
SCS 66.77 (17.41) 81.69 (13.14)* 81.92 (13.25)** 76.42 (14.28) 81.92 (11.60) 84.75 (7.14)* 75.00 (16.29) 73.71 (14.42) 75.43 (13.25)
RRQ 46.77 (9.02) 36.54 (8.62)** 38.15 (5.84)** 43.08 (6.78) 38.25 (9.15)* 38.25 (6.97)* 45.14 (10.95) 44.57(10.53) 45.86 (8.26)

M, mean; SD, standard deviation.

T-test significant at *p < 0.05 and **p < 0.001.

8.31]. No significant change was found in the post to follow- FFMQ

up DASS scores in the MBCT group (p > 0.05). Participants The main effect of time was significant F (2, 58) = 11.33, p < 0.001),
in the CFT group had significantly lower post-treatment total but the main effect of group was not significant F (2, 29) = 1.92, p >
DASS scores t (11) = 4.76, p < 0.001, d = 1.60, 95% CI [5.82, 0.5). The interaction of time by group F (4, 58) = 7.14, p < 0.003)
15.84], and significantly lower post-treatment scores on the was significant. As shown in Table 4, participants in the MBCT
depression t (11) = 4.42, p < 0.001, d = 1.29, 95% CI [2.05, group had significantly higher post-treatment FFMQ scores t (12)
6.12], and stress t (11) = 4.28, p < 0.001, d = 1.64, 95% CI = −14.19, p < 0.001, d = −1.27, 95% CI [−25.49, −8.05].
[2.59, 8.07] subscales. However, no significant change was found This increase was maintained at follow-up as the participants
in their post-treatment scores on the anxiety subscale of DASS in the MBCT group had significantly higher follow-up FFMQ
(p > 0.5). The reductions were maintained at follow-up as the scores in comparison to pre-treatment t (12) = −6.25, p < 0.000,
participants in the CFT group had significantly lower follow- d = −1.39, 95% CI [−24.17, −11.68]. No significant change
up total DASS scores in comparison to pre-treatment t (11) = was found between the post to follow-up FFMQ scores in the
3.03, p < 0.02, d = 1.41, 95% CI [2.32, 14.68], and similarly MBCT group (p > 0.05). Participants in the CFT group also had
significantly lower follow-up scores on the depression t (11) = significantly higher post-treatment FFMQ scores t (11) = −4.52,
3.15, p < 0.01, d = 1.05, 95% CI [0.98, 5.52], and stress t (11) p < 0.001, d = −0.88, 95% CI [−17.72, −6.12]. This increase
= 3.01, p < 0.02, d = 1.22, 95% CI [1.03, 6.63]. However, no was maintained at follow-up as the participants in the CFT group
significant change was found from pre-treatment to follow-up had significantly higher follow-up FFMQ scores in comparison
scores on the anxiety subscale (p > 0.05). No significant changes to pre-treatment t (11) = −3.08, p < 0.02, d = −0.81, 95% CI
were found in the post to follow-up DASS scores in the CFT [−17.15, −2.85]. No significant change was found between the
group (all ps > 0.05). No significant changes were found in post to follow-up FFMQ scores in the CFT group (p > 0.05).
the pre, post and follow-up DASS scores in the control group No significant change was found in the pre, post and follow-up
(all ps > 0.05). FFMQ scores in the control group (all ps > 0.05).

Frontiers in Psychology | www.frontiersin.org 6 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

FIGURE 1 | Changes in mean scores for all measures from pre to post-treatment in the MBCT, CFT, and Control groups.

SCS between the post and follow-up SCS scores in the CFT group (p
The main effect of time was significant F (2, 58) = 9.83, p < 0.004), > 0.05). No significant change was found in the pre, post, and
but the main effect of group was not significant F (2, 29) = 0.69, follow-up SCS scores in the control group (all ps > 0.05).
p > 0.5). The interaction of time by group F (4, 58) = 3.64, p <
0.04) was significant. As shown in Table 4, participants in the RRQ
MBCT group had significantly higher post-treatment SCS scores he main effect of time was significant F (2, 58) = 12.99, p < 0.001),
t (12) = −4.08, p < 0.002, d = −0.97, 95% CI [−22.89, −6.96]. but the main effect of group was not significant F (2, 29) = 1.21,
This increase was maintained at follow-up as the participants in p > 0.5). The interaction of time by group F (4, 58) = 4.13, p <
the MBCT group had significantly higher follow-up SCS scores 0.03) was significant. As shown in Table 4, participants in the
in comparison to pre-treatment t (12) = −4.32, p < 0.001, d = MBCT group had significantly lower post-treatment RRQ scores
−0.98, 95% CI [−22.79, −7.52]. No significant change was found t (12) = 4.74, p < 0.001, d = 1.16, 95% CI [5.52, 14.94]. This
between the post and follow-up SCS scores in the MBCT group (p reduction was maintained at follow-up as the participants in the
> 0.05). Participants in the CFT did not have significantly higher MBCT group had significantly lower follow-up RRQ scores in
post-treatment SCS scores (p > 0.05) but they had significantly comparison to pre-treatment t (12) = 4.12, p < 0.001, d = 1.13,
higher follow-up SCS scores t (11) = −2.66, p < 0.03, d = 95% CI [4.06, 13.17]. No significant change was found between
−0.74, 95% CI [−15.23, −1.44]. No significant change was found the post and follow-up RRQ scores in the MBCT group (p >

Frontiers in Psychology | www.frontiersin.org 7 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

0.05). Participants in the CFT group also had significantly lower Impact of Pre-treatment Rumination Levels
post-treatment RRQ scores t (11) = 2.61, p < 0.03, d = 0.60, 95% on the Treatment Outcomes
CI [0.76, 8.91]. This reduction was maintained at follow-up as In order to investigate whether the tendency to ruminate
the participants in the CFT group had significantly lower follow- affected how participants responded to the two interventions,
up RRQ scores in comparison to pre-treatment t (11) = 3.01, p the participants in the MBCT and in the CFT groups were
< 0.02, d = 0.70, 95% CI [1.30, 8.36]. No significant change further split into two subgroups based on a median split
was found between the post and follow-up RRQ scores in the of their RRQ scores at pre-test. Higher than median scores
CFT group (p > 0.05). No significant change was found in the determined higher rumination tendency and lower than median
pre, post and follow-up RRQ scores in the control group (all ps scores determined lower rumination tendency. This resulted in
> 0.05). The mean scores on the dependent measures across time an additional factor of rumination which was included in the
for participants in the treatment groups and the control group following analyses. A repeated-measures three-way ANOVAs
are shown in Figure 2. In line with abovementioned results it with time (pre, post) as within-subject factor and intervention
shows that the effects of both treatments are mostly maintained (MBCT, CFT, Control) and rumination (high, low) as between-
at 1month follow-up. subject factors, were conducted.

FIGURE 2 | Changes in mean scores from pre- through post-treatment and to 1 month follow-up in the MBCT, CFT, and control groups.

Frontiers in Psychology | www.frontiersin.org 8 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

DASS rumination scores and significant increase in mindfulness and

No significant three-way time × group × rumination self-compassion scores. However, only the MBCT group showed
interactions were found for either the total DASS scores or significant reductions in anxiety scores whereas the CFT group
the DASS subscales scores (all ps > 0.05). did not manifest a significant change.
Overall, findings of the current study are consistent with a
FFMQ number of studies that have shown that MBCT and CFT are
There was a significant three-way time × group × rumination effective at enhancing mindfulness and self-compassion and at
interaction F (1.48) = 8.65, p < 0.001). Participants in the MBCT reducing depression, anxiety, stress, and rumination (Teasdale
group who were high in rumination had significantly higher post- et al., 2000; Ma and Teasdale, 2004; Gilbert and Procter, 2006;
treatment FFMQ scores t (9) = −5.46, p < 0.001, d = −2.14, 95% Gilbert, 2009a; Goss and Allan, 2010; Kuyken et al., 2010; Lowens,
CI [−38.75, −16.05], whereas the ones low in rumination did 2010; Rimes and Wingrove, 2011). This is the first study that has
not show any significant change (p > 0.05). Participants in the directly compared the effects of MBCT and CFT and according to
CFT group who were high in rumination had significantly higher our findings there were no significant differences in their effects,
post-treatment FFMQ scores t (8) = −4.11, p < 0.003, d = −1.21, except for follow up reductions in anxiety with significant change
95% CI [−22.73, −6.38], and the ones low in rumination had also in the MBCT group only. Interestingly, both MBCT and CFT
significantly higher post-treatment FFMQ scores t (7) = −3.69, p resulted in significant improvements for self-compassion with a
< 0.008, d = −0.95, 95% CI [−19.48, −4.27]. Participants who medium effect size for the MBCT group and small effect size in
were both high and low in rumination in the control group did the CFT group which is contrary to the prediction that explicit
not show any significant change in FFMQ scores (all ps > 0.05). cultivation of self-compassion leads to larger enhancements
(Gilbert, 2009a).
SCS Our findings partly support the notion that people differ
There was no significant three-way time x group x rumination in their response to the treatments based on their tendency
interaction (p > 0.05). to ruminate as participants who were high in rumination at
pre-test in the MBCT group showed a significant increase in
DISCUSSION mindfulness at post-test and those with low rumination did
not show improvements. Such selective increase in mindfulness
This pilot study aimed to investigate the effects of for the high rumination participants lends further support
MBCT applying implicit self-compassion instructions and to previous results with recurrently depressed participants in
CFT employing explicit self-compassion instructions on remission where only those with high rumination positively
depression/anxiety/stress symptom change, mindfulness, self- responded to a brief mindfulness induction (Barnhofer et al.,
compassion, and rumination. Participants in both treatment 2010). This interesting finding is more broadly in line with the
groups showed significant increases in mindfulness and self- pervious evidence that MBCT is particularly effective for people
compassion and decreases in rumination, depression, anxiety, with depression (Teasdale et al., 2000; Ma and Teasdale, 2004)
and stress at post-test, whereas there were no changes reported as rumination is considered to be one of the key risk factors
for participants in the control group. The significant findings for depression (Nolen-Hoeksema, 2000). However, findings from
for the symptom change and mindfulness scores in the two the current study also surprisingly revealed that both high
treatment groups were accompanied by large effect sizes. The and low rumination participants in the CFT group showed
significant reductions in rumination were associated with large significant increases in mindfulness at post-test which is contrary
effect size in the MBCT group and medium effect size in the to Barnhofer’s study (2010) where only participants with low
CFT group. Importantly, both participants in the CFT group rumination benefited from brief loving kindness instructions.
and in the MBCT group showed significant improvements Importantly, we haven’t found differences between the high and
for the self-compassion scores at the post-test, with a medium low rumination groups for either MBCT or CFT in symptom
effect size in the MBCT group and a small effect size in the change and self-compassion scores which somewhat limits
CFT group. Furthermore, the exploration of the effects of the findings of differences in mindfulness gains and requires
MBCT and CFT on the participant outcomes depending on further investigation.
their pre-treatment tendency to ruminate revealed that MBCT
participants who were high in rumination, but not those low Limitations
in rumination, showed significantly increased mindfulness Overall, interpretation of the findings of the current study needs
scores. Interestingly, for CFT participants mindfulness scores to take into account several limitations. First, the study did
at post-test significantly increased in both rumination groups. not employ a randomized controlled design. While this raises
However, all these significant findings need to be interpreted some concerns with regard to comparability of the groups, there
with caution given the non-randomized design of this study and were no significant differences found in any of the measures
a relatively small sample size. between the groups at pre-treatment. Second, the treatment
In examining the effects of CFT and MBCT from pre- groups were in a residential rehabilitation and health clinic
treatment to follow-up with a reduced sample of participants due during the treatment period, in which they were offered variety of
to lack of response at this third time point (N = 32), participation other beneficial programs along with MBCT and CFT. This could
in both treatments led to a significant decrease in total DASS and have confounded the findings to some extent. Third, there is also

Frontiers in Psychology | www.frontiersin.org 9 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

the possibility that the findings of this study have been impacted support the hypothesis that participants differ in their response
by the small sample size and therefore need to be interpreted with to meditation practices and that MBCT may be more effective at
caution (even though the combination of significant findings and enhancing mindfulness for people who are high in rumination.
large/medium effect sizes is reassuring). Shorter time-frame for Surprisingly, we have found that CFT may be effective at
delivery of MBCT and CFT (half in comparison to the usual 8- enhancing mindfulness for both those with high and low
week format, yet equal number of sessions was covered) may rumination. Further research is needed to conclusively elucidate
have also impacted the results since it may take time to develop the similarities and differences between effects of implicit and
mindfulness and self-compassion skills. In addition, difficulties explicit self-compassion instruction on participant outcomes.
may arise with replicating this study with a larger sample, as
CFT is not manualized like MBCT. In addition, no external ETHICS STATEMENT
examination was put into place in order to ensure the fidelity of
the treatments. Finally, the follow-up time in the current study The study was approved by the Research Ethics &amp;
was limited to one month which does not allow for informed Governance Committee at Bangor University and the Ethic
conclusions about possible longer-term effects of MBCT and CFT Committee at the Directorate of Health in Iceland.
and their comparison.
Future studies should employ a randomized controlled design AUTHOR CONTRIBUTIONS
with a larger sample, in both clinical and non-clinical settings
and with longer and repeated follow-ups. If conducted in a AF designed and executed the study, conducted the data analyses,
residential rehabilitation clinic, it would be recommended to and wrote the manuscript. DD guided and supervised the
compare the treatment groups to a control group that is also design/execution of the study and data analyses, she edited
in the clinic at the same time but not receiving MBCT or CFT. the manuscript.
It would be interesting to compare MBCT to a standardized
manualized self-compassion program. Future studies could also ACKNOWLEDGMENTS
include assessments such as behavioral or psychophysiological
markers to bypass some of the limitations of self-reports. Impact The authors would like to thank Dr. Ragnar P. Olafsson at the
of individual differences at the baseline, such as levels of University of Iceland for his support with the ethics application
rumination, on treatment outcomes needs further investigation. to the Ethic Committee at the Directorate of Health in Iceland
In conclusion, the findings from the current study suggest and his helpful guidance throughout the process. They would
that both MBCT and CFT are effective at enhancing mindfulness also like to thank the manager at the NFLI for his approval to
and self-compassion and at reducing depression, anxiety, stress, conduct the study at the medical private rehabilitation center
and rumination in clients with anxiety, depression, and stress and Margret Arnljotsdottir, clinical psychologist, and Bridget
difficulties. It seems that the implicit way of cultivating self- Yr McEvoy, psychiatric nurse, for all their hard work and
compassion in MBCT is just as effective as cultivating self- diligence in running the programs so skillfully and collecting
compassion explicitly in CFT. Furthermore, our findings partly the data.

REFERENCES of a feasibility randomized controlled trial. Br. J. Clin. Psychol. 52, 199–214.
doi: 10.1111/bjc.12009
Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., and Toney, L. (2006). Using Carmody, J., and Baer, R. A. (2008). Relationship between mindfulness practice
self-report assessment methods to explore facets of mindfulness. Assessment 13, and levels of mindfulness, medical and psychological symptoms and wellbeing
27–45. doi: 10.1177/1073191105283504 in a mindfulness-based stress reduction program. J. Behav. Med. 31, 23–33.
Barnhofer, T., Chittka, T., Nightingale, H., Visser, C., and Crane, C. (2010). State doi: 10.1007/s10865-007-9130-7
effects of two forms of meditation on prefrontal EEG asymmetry in previously Chiesa, A., and Serretti, A. (2009). Mindfulness-based stress reduction for stress
depressed individuals. Mindfulness 1, 21–27. doi: 10.1007/s12671-010-0004-7 management in healthy people: a review and meta-analysis. J. Alternat.
Bergen-Cico, D., and Cheon, S. (2013). The mediating effects of Complement. Med. 15, 593–600. doi: 10.1089/acm.2008.0495
mindfulness and self-compassion on trait anxiety. Mindfulness 2013:205. Dalai Lama (1995). The Power of Compassion. Delhi: Harper Collins.
doi: 10.1007/s12671-013-0205-y Ehret, A. M., Joormann, J., and Berking, M. (2015). Examining risk and resilience
Birnie, K., Speca, M., and Carlson, L. E. (2010). Exploring self-compassion and factors for depression: the role of self-criticism and self-compassion. Cogn.
empathy in the context of mindfulness-based stress reduction (MBSR). Stress Emot. 29, 1496–1504. doi: 10.1080/02699931.2014.992394
Health 26:359. doi: 10.1002/smi.1305 Gale, C., Gilbert, P., Read, N., and Goss, K. (2012). An evaluation of the
Blatt, S. J. (1995). “Representational structures in psychopathology,” in Rochester impact of introducing compassion focused therapy to a standard treatment
symposium on developmental psychopathology: Emotion, cognition,and programme for people with eating disorders. Clin. Psychol. Psychother.
representation, Vol. 6, eds D. Cicchetti and S. Toth (Rochester, NY: University 2012:1806. doi: 10.1002/cpp.1806
of Rochester Press, 1–34. Germer, C., and Barnhofer, T. (2017). “Mindfulness and compassion: similarities
Bohlmeijer, E., Prenger, R., Taal, E., and Cuijpers, P. (2010). The effects of and differences,” in Compassion: Concepts, Research and Applications, ed P.
mindfulness-based stress reduction therapy on mental health of adults with Gilbert (London: Routledge).
a chronic medical disease: a meta-analysis. J. Psychosomat. Res. 68, 539–544. Germer, C. K. (2009). The Mindful Path to Self-Compassion: Freeing Yourself From
doi: 10.1016/j.jpsychores.2009.10.005 Destructive Thoughts and Emotions. New York, NY: Guilford Press.
Braehler, C., Gumley, A., Harper, J., Wallace, S., Norrie, J., and Gilbert, P. (2013). Gilbert, P. (2009a). The Compassionate Mind: A New Approach to Life’S Challenges.
Exploring change processes in compassion focused therapy in psychosis: results Oakland, CA: New Harbinger.

Frontiers in Psychology | www.frontiersin.org 10 May 2019 | Volume 10 | Article 1099

Frostadottir and Dorjee Effects of MBCT and CFT

Gilbert, P. (2009b). Introducing compassion-focused therapy. Adv. Psychiatry Lucre, K. M., and Corten, N. (2013). An exploration of group compassion-
Treat. 15, 199–208. doi: 10.1192/apt.bp.107.005264 focused therapy for personality disorder. Psychol. Psychother. 86, 387–400.
Gilbert, P. (2017). “Compassion: definitions and controversies,” in Compassion: doi: 10.1111/j.2044-8341.2012.02068.x
Concepts, Research and Applications, ed P. Gilbert (London: Routledge), 3–15. Ma, S. H., and Teasdale, J. D. (2004). Mindfulness-Based cognitive therapy
doi: 10.4324/9781315564296 for depression: replication and exploration of differential relapse prevention
Gilbert, P., and Irons, C. (2005). “Focused therapies and compassionate mind effects. J. Consult. Clin. Psychol. 72:31. doi: 10.1037/0022-006X.72.1.31
training for shame and self-attacking,” in Compassion: Conceptualisations, MacBeth, A., and Gumley, A. (2012). Exploring compassion: a meta-analysis of the
Research and Use in Psychotherapy, ed P. Gilbert (London: Routledge), associations between self-compassion and psychopathology. Clin. Psychol. Rev.
263–325. 32, 545–552. doi: 10.1016/j.cpr.2012.06.003
Gilbert, P., and Procter, S. (2006). Compassionate mind training for people with Muris, P., Otgaar, H., and Petrocchi, N. (2016). Protection as the mirror
high shame and self-criticism: overview and pilot study of a group therapy image of psychopathology: further critical notes on the self-compassion scale.
approach. Clin. Psychol. Psychother. 13, 353–379. doi: 10.1002/cpp.507 Mindfulness 7, 787–790. doi: 10.1007/s12671-016-0509-9
Goss, K., and Allan, S. (2010). Compassion focused therapy for eating disorders. Neff, K. D. (2003a). Self-compassion: An alternative conceptualization of a healthy
Int. J. Cogn. Ther. 3, 141–158. doi: 10.1521/ijct.2010.3.2.141 attitude toward oneself. Self Identity 2, 85–102 doi: 10.1080/15298860309032
Goyal, M., Singh, S., Sibinga, E. M., Gould, N. F., Rowland-Seymour, A., Sharma, Neff, K. D. (2003b). The development and validation of a scale to measure
R., et al. (2014). Meditation programs for psychological stress and well-being: self-compassion. Self Identity 2, 223–250. doi: 10.1080/15298860309027
a systematic review and meta-analysis. JAMA Intern. Med. 174, 357–368. Neff, K. D. (2011). Self-Compassion. New York, NY: William Morrow.
doi: 10.1001/jamainternmed.2013.13018 Neff, K. D. (2015). Self-Compassion: The Proven Power of Being Kind to Yourself.
Grossman, P., Niemann, L., Schmidt, S., and Walach, H. (2004). Mindfulness- New York, NY: William Morrow.
Based stress reduction and health benefits. A meta-analysis. J. Psychosomat. Res. Neff, K. D., and Dahm, K. A. (2014). “Self-Compassion: what it is, what it does,
57, 35–43. doi: 10.1016/S0022-3999(03)00573-7 and how it relates to mindfulness,” in Mindfulness and Self-Regulation, eds
Henry, J. D., and Crawford, J. R. (2005). The short-form version of the M. Robinson, B. Meier and B. Ostafin (New York, NY: Springer), 121–140.
Depression Anxiety Stress Scales (DASS- 21): construct validity and normative doi: 10.1007/978-1-4939-2263-5_10
data in a large non-clinical sample. Br. J. Clin. Psychol. 44, 227–239. Nolen-Hoeksema, S. (2000). The role of rumination in depressive disorders
doi: 10.1348/014466505X29657 and mixed anxiety/depressive symptoms. J. Abnormal Psychol. 109, 504–511.
Heriot-Maitland, C., Vidal, J. B., Ball, S., and Irons, C. (2014). A compassionate- doi: 10.1037/0021-843X.109.3.504
focused therapy group approach for acute inpatients: feasibility, initial Piet, J., and Hougaard, E. (2011). The effect of mindfulness-based cognitive
pilot outcome data, and recommendations. Br. J. Clin. Psychol. 53, 78–94. therapy for prevention of relapse in recurrent major depressive disorder:
doi: 10.1111/bjc.12040 a systematic review and meta- analysis. Clin. Psychol. Rev. 31, 1032–1040.
Hildebrandt, L. K., McCall, C., and Singer, T. (2017). Differential effects of doi: 10.1016/j.cpr.2011.05.002
attention-, compassion-, and socio-cognitively based mental practices on Rimes, K. A., and Wingrove, J. (2011). Pilot study of mindfulness-based cognitive
self-reports of mindfulness and compassion. Mindfulness 8, 1488–1512. therapy for trainee clinical psychologists. Behav. Cogn. Psychother. 39, 235–241.
doi: 10.1007/s12671-017-0716-z doi: 10.1017/S1352465810000731
Hofman, S. G., Sawyer, A. T., Witt, A. A., and Oh, D. (2010). The effect of Segal, Z. V., Williams, J. M. G., and Teasdale, J. D. (2002). Mindfulness–Based
mindfulness-based therapy on anxiety and depression: a meta-analytic review. Cognitive Therapy for Depression: A New Approach to Preventing Relapse. New
J. Consult. Clin. Psychol. 78, 169–183. doi: 10.1037/a0018555 York, NY: Guilford Press.
Kabat-Zinn, J. (1990). Full Catastrophe Living: Using the Wisdom of Your Body and Segal, Z. V., Williams, J. M. G., and Teasdale, J. D. (2013). Mindfulness–based
Mind to Face Stress, Pain and Illness. New York, NY: Delta. Cognitive Therapy for Depression: A New Approach to Preventing Relapse, 2nd
Kabat-Zinn, J. (2003). Mindfulness-based interventions in context: past, present Ed. New York, NY: Guilford Press.
and future. Clin. Psychol. 10, 144–156. doi: 10.1093/clipsy.bpg016 Teasdale, J. D., Segal, Z. V., Williams, J. M., Ridgeway, V. A., Soulsby, J. M., and
Kelly, A. C., Wisniewski, L., Martin-Wagar, C., and Hoffman, E. (2016). Group- Lau, M. A. (2000). Prevention of relapse/recurrence in major depression by
based compassion-focused therapy as an adjunct to outpatient treatment for mindfulness-based cognitive therapy. J. Consult. Clin. Psychol. 68, 615–623.
eating disorders: a pilot randomized controlled trial. Clin. Psychol. Psychother. doi: 10.1037/0022-006X.68.4.615
24, 475–487. doi: 10.1002/cpp.2018 Trapnell, P. D., and Campbell, J. D. (1999). Private self-consciousness and the
Keng, S. L., Smoski, M. J., and Robins, C. J. (2011). Effects of mindfulness on Five-Factor Model of Personality: distinguishing rumination from reflection.
psychological health: a review of empirical studies. Clin. Psychol. Rev. 31, J. Personal. Soc. Psychol. 76, 284–304. doi: 10.1037/0022-3514.76.2.284
1041–1056. doi: 10.1016/j.cpr.2011.04.006 Wallace, B. A. (1999). The Four Immeasurables: Cultivating a Boundless Heart.
Khoury, B., Lecomte, T., Fortin, G., Masse, M., Therien, P., Bouchard, V., Ichaca, NY: Snow Lion Publications, 192.
et al. (2013). Mindfulness-based therapy: a comprehensive meta-analysis. Clin. Williams, M. J., Dalgleish, T., Karl, A., and Kuyken, W. (2014). Examining the
Psychol. Rev. 33, 763–771. doi: 10.1016/j.cpr.2013.05.005 factor structures of the Five Facet Mindfulness Questionnaire and the Self-
Kuyken, W., Watkins, E., Holden, E., White, K., Taylor, R. S., Byford, S., et al. Compassion Scale. Psychol. Assessment 26, 407–418. doi: 10.1037/a0035566
(2010). How does mindfulness-based cognitive therapy work? Behav. Res. Ther.
48, 1105–1112. doi: 10.1016/j.brat.2010.08.003
Laithwaite, H., O’Hanlon, M., Collins, P., Doyle, P., Abraham, L., Porter, S., et al. Conflict of Interest Statement: The authors declare that the research was
(2009). Recovery After Psychosis (RAP): a compassion focused programme conducted in the absence of any commercial or financial relationships that could
for individuals residing in high security settings. Behav. Cogn. Psychother. 37, be construed as a potential conflict of interest.
511–526. doi: 10.1017/S1352465809990233
Lovibond, P. F., and Lovibond, S. H. (1995). The structure of negative emotional Copyright © 2019 Frostadottir and Dorjee. This is an open-access article distributed
states: comparison of the Depression Anxiety Stress Scales (DASS) with under the terms of the Creative Commons Attribution License (CC BY). The use,
the beck depression and anxiety inventories. Behav. Res. Ther. 33, 335–342. distribution or reproduction in other forums is permitted, provided the original
doi: 10.1016/0005-7967(94)00075-U author(s) and the copyright owner(s) are credited and that the original publication
Lowens, I. (2010). Compassion focused therapy for people with bipolar disorder. in this journal is cited, in accordance with accepted academic practice. No use,
Int. J. Cogn. Ther. 3, 172–185. doi: 10.1521/ijct.2010.3.2.172 distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Psychology | www.frontiersin.org 11 May 2019 | Volume 10 | Article 1099