Mindfulness Training in a Heterogeneous Psychiatric Sample:

Outcome Evaluation and Comparison of Different Diagnostic Groups

Elisabeth H. Bos,1,2 Ria Merea,1 Erik van den Brink,1 Robbert Sanderman,3
and Agna A. Bartels-Velthuis1,2
1
Lentis Mental Health Organization, Center for Integrative Psychiatry, Groningen, the
Netherlands
2
University of Groningen, University Medical Center Groningen, University Center for
Psychiatry, Groningen, the Netherlands
3
University of Groningen, University Medical Center Groningen, Health Psychology Section,
Groningen, the Netherlands

Objectives: To examine outcome after mindfulness training in a heterogeneous psychiatric outpa-

tient population and to compare outcome in different diagnostic groups. Method: One hundred
and forty-three patients in 5 diagnostic categories completed questionnaires about psychological symp-
toms, quality of life, and mindfulness skills prior to and immediately after treatment. Results: The
mixed patient group as a whole improved significantly on all outcome measures. Differential improve-
ment was found for different diagnostic categories with respect to psychological symptoms and quality
of life: Bipolar patients did not improve significantly on these measures. This finding could be explained
by longer illness duration and lower baseline severity in the bipolar category. Conclusion: Mind-
fulness training is associated with overall improvement in a heterogeneous outpatient population. Dif-
ferences in outcome between diagnostic categories may be ascribed to differences in illness duration
and baseline severity.  C 2013 Wiley Periodicals, Inc. J. Clin. Psychol. 70:60–71, 2014.

Keywords: mindfulness training; depressive disorder; anxiety disorder; adjustment disorder; bipolar
disorder

Mindfulness refers to “the awareness that emerges through paying attention on purpose, in the
present moment, and nonjudgmentally to the unfolding of experience moment by moment”
(Kabat-Zinn, 2003, p. 145). The practice of mindfulness includes observing and attending to
internal and external experiences as they occur in the here and now and cultivating a noneval-
uative and open attitude to these experiences (Kabat-Zinn, 1982, 1990; Bishop et al., 2004).
Increased mindfulness is thought to improve psychological functioning, presumably by promot-
ing an adaptive form of self-focused attention that reduces rumination and emotional avoidance
and improves behavioral self-regulation (Baer, 2009; Kuyken et al., 2010).
Jon Kabat-Zinn developed the mindfulness-based stress reduction (MBSR) training in the
1970s to help patients suffering from chronic pain and stress cope with their complaints (Kabat-
Zinn, 1990). About two decades later, mindfulness-based cognitive therapy (MBCT) was devel-
oped by Teasdale and colleagues, combining elements of MBSR and cognitive therapy, aimed at
preventing recurrence in depressive disorder (Teasdale, Segal, & Williams, 1994; Teasdale et al.,
2000). Several randomized controlled studies conducted since then have shown that MBSR is
indeed effective in improving mental health in clinical as well as nonclinical populations (Gross-
man, Niemann, Schmidt, & Walach, 2004; Hofmann, Sawyer, Witt, & Oh, 2010; Fjorback,
Arendt, Ornbol, Fink, & Walach, 2011), while MBCT was found to be effective in preventing
recurrence in remitted depressed patients (Piet & Hougaard, 2011; Chiesa & Serretti, 2011; Van
Aalderen et al., 2012).

Please address correspondence to: Dr. Bos, Elisabeth, P.O. Box 30.001, Groningen 9700RB. E-mail:
[email protected], [email protected]

JOURNAL OF CLINICAL PSYCHOLOGY, Vol. 70(1), 60–71 (2014) 

C 2013 Wiley Periodicals, Inc.

Published online in Wiley Online Library (wileyonlinelibrary.com/journal/jclp). DOI: 10.1002/jclp.22008

 Mindfulness in a Heterogeneous Sample 61

Both in somatic medicine and psychiatry, mindfulness training (MBSR/MBCT) is being

used more and more in a wide range of diagnostic categories. A number of studies suggest
that mindfulness training may also have beneficial effects in currently symptomatic depression
(Chiesa & Serretti, 2011), eating disorder (Wanden-Berghe, Sanz-Valero, & Wanden-Berghe,
2011), sleeping disorder (Winbush, Gross, & Kreitzer, 2007), and psychosis (Langer, Cangas,
Salcedo, & Fuentes, 2012). Furthermore, some preliminary evidence has been found for positive
effects in bipolar disorder (Deckersbach et al., 2012; Stange et al., 2011; Williams et al., 2008),
generalized anxiety disorder (Hofmann et al., 2010; Kim et al., 2010a), panic disorder (Kim
et al., 2010b), social phobia (Piet, Hougaard, Hecksher, & Rosenberg, 2010), hypochondriasis
(Lovas & Barsky, 2010; Williams, McManus, Muse, & Williams, 2011; McManus, Surawy, Muse,
Vazquez-Montes, & Williams, 2012), and substance use disorder (Zgierska et al., 2009; Bowen
et al., 2009), although in these contexts the evidence is still rather weak.
Although mindfulness training is now being offered to and studied within a broad variety of
diagnostic categories, the treatment groups usually are rather homogeneous as regards diagnosis.
So far, few studies have been done on the feasibility and effectiveness of mindfulness training
in a heterogeneous patient group; we have found only two small studies in depressed/anxious
samples (Green & Bieling, 2012; Ree & Craigie, 2007). Therefore, little is known about the
applicability of offering mindfulness training to a mixed patient group. Moreover, knowledge
about the differential effectiveness of such training in different diagnostic groups is limited, as
no direct comparisons can be made between different diagnostic categories. The aim of this
naturalistic outcome study was therefore to examine outcome following mindfulness training in
a heterogeneous psychiatric outpatient population and to explore whether outcome differs in
patients from different diagnostic categories. We hypothesized that mindfulness training would
be associated with improvement in this heterogeneous patient group: We expected a reduction
in the number of psychological symptoms, improvement in quality of life, and an increase in
the level of mindfulness. We did not expect a differential outcome for patients from different
diagnostic categories.

Method
Participants
Outpatients attending mindfulness training at the Center for Integrative Psychiatry (CIP) in
Groningen, the Netherlands between 2006 and 2009 were subjects in this study. Inclusion criteria
for participation in the training were as follows: aged 18 years or older; a stable psychiatric
disorder (i.e. no current severe depressive episode, no current psychotic episode, and no current
[hypo]mania: sufficient stability for the patient to attend the sessions and do the homework
practices); no alcohol or drug dependence; reasonable expectations of the training; motivation
and willingness to do assigned homework; ability to participate in a group; and no practical or
physical impediments that would preclude following the training. On the basis of a clinical intake
interview, a psychiatrist made the Diagnostic and Statistical Manual of Mental Disorders Fourth
Edition (DSM-IV) diagnosis and assessed the other inclusion criteria by explicitly discussing
the training’s objective and evaluating potential obstacles for successful participation.
A total of 233 patients participated in the training between 2006 to 2009, 214 of whom
agreed to participate in the study and complete the pretraining questionnaires (92%). At the
end of the training, 143 patients completed the posttraining questionnaires (i.e. 67% of 214).
No systematic record was kept of how many of the 71 noncompleters (33%) did complete the
mindfulness training, but dropout percentages for this training at our center tend to be about
7.5%. The noncompleters did not differ from the completers with respect to age, t(212) = 0.41, p
= 0.684, illness duration, t(187) = −1.15, p = 0.250, gender, χ2 (1) = 0.11, p = 0.742, education,
χ2 (2) = 1.49, p = 0.475, having a partner, χ2 (1) = 1.10, p = 0.294, or diagnostic category,
χ2 (4) = 3.31, p = 0.508. The noncompleters and completers also did not differ with respect to
their pretreatment scores on the outcome measures, except for the Awareness subscale of the
Kentucky Inventory of Mindfulness Skills, t(179) = −2.05, p = 0.042: noncompleters scored a
little higher on this subscale, mean (standard deviation) = 27.8 (5.7) versus 26.0 (5.4).
62 Journal of Clinical Psychology, January 2014

Measures
Outcome data were collected at pretreatment and posttreatment assessments by means of ques-
tionnaires addressing symptomatology, quality of life, and mindfulness skills. Demographic data
were collected at the pretreatment assessment by means of a questionnaire.

Short Symptom List (SSL). The SSL (Dutch: Korte Klachten Lijst, KKL) is a self-
report questionnaire of 13 items about the degree to which respondents suffer from common
psychological symptoms, like anxiety, depression, sleeping problems, and addiction (Lange &
Appelo, 2007). For this study, an extra item was included regarding (hypo)manic symptoms.
Scores for each item can range from 0 (not at all) to 4 (very much). The total SSL score therefore
varies from 0 to 56. The reliability and validity of the SSL are satisfactory to good (Lange &
Appelo, 2007). The SSL is a proper short alternative for the Symptom Checklist-90; total scores
on these measures are highly correlated (Appelo & Lange, 2007).

World Health Organisation Quality of Life-Bref (WHOQOL-Bref). The

WHOQOL-Bref aims to measure quality of life (De Vries & Van Heck, 1996; De Vries & Van
Heck, 2003; Trompenaars, Masthoff, Heck, Hodiamont, & Vries, 2005). It contains 26 items,
24 of which can be classified into four domains: (a) Physical health, (b) Psychological health,
(c) Social relationships and (d) Environment. The other two items, regarding overall quality of
life and general health, are combined in a facet (Overall Quality of Life and General Health
[Overall QoL & GH]). Item scores range from 1 to 5. Total scores on the different domains and
the facet range from 4 to 20, with higher scores denoting higher quality of life. The reliability
and validity of the WHOQOL-Bref are satisfactory to good (Trompenaars et al., 2005).

Kentucky Inventory of Mindfulness Skills (KIMS-39). The KIMS-39 measures the

degree of mindfulness (Baer, Smith, & Allen, 2004; Baer et al., 2008) with 39 questions in
four subscales. The Observe subscale questions the ability to perceive or notice internal stimuli
(feelings, thoughts, physical sensations) and external stimuli (e.g., noises). The Describe subscale
investigates the ability to describe or label internal and external stimuli. The Act with awareness
subscale measures how well the respondent can focus on and pay attention to the present
moment. The Accept without judgment subscale measures whether the patient is able to accept
present moment experiences without judging them. Item scores range from 1 (never true) to 5
(always true). Ranges for total scores for the different subscales are as follows: 12–60 (Observe),
8–40 (Describe), 10–50 (Act With Awareness), and 9–45 (Accept Without Judgment). The
psychometric properties of the KIMS-39 are good (Baer et al., 2004; Baer et al., 2008).

Training
The mindfulness training comprised eight weekly sessions of 2 1/2 hours each, and a silent
retreat session held between sessions 6 and 7. Each group comprised 12 to 16 participants. The
training followed the format of the MBSR program as described by Jon Kabat-Zinn (Kabat-Zinn,
1990). Because a substantial proportion of the participants had been diagnosed with (recurrent)
depression, the MBSR program in our center was combined with a number of core elements
from the MBCT program (Segal, Williams, & Teasdale, 2002). This was done because MBCT
elements are more specific in addressing skills relevant to depression, like disengaging from
dysphoria-induced depressogenic thinking and dealing constructively with negative cognitions
and self-critical thoughts (Teasdale et al., 2000; Segal et al., 2002). Some of the MBCT elements
were integrated in the program as standard elements, while other ones were offered optionally in
the handouts and homework assignments; participants were encouraged to choose from these
optional elements according to their own needs and preferences.
All sessions included the following: one or more formal mindfulness exercises (body scan,
sitting meditation, yoga); a group-based discussion of patients’ experience of these practices;
psychoeducation related to body-mind interaction, automatic versus mindful responses to stress,
and communication; interactive group evaluation of homework assignments; and instructions
 Mindfulness in a Heterogeneous Sample 63

regarding homework assignments for the following week. MBCT-specific elements in the core
program included 3-Minute Breathing Space (standard and coping), Automatic Thoughts Ques-
tionnaire, and Developing an action plan (i.e., strategies for an adequate response to early warn-
ing signals). A full description of the structure and elements of each session of the program is
outlined in a treatment protocol available from the authors upon request.
Before starting the training, participants were instructed that the program would entail do-
ing homework for 45 minutes to 1 hour every day. During the training, the trainers repeatedly
reminded the participants how important this homework was and motivated them to do the as-
signments, in an encouraging, noncoercive way. The homework assignments comprised guided
(audiotaped) mindfulness exercises (e.g. body scan, yoga exercises, sitting meditation, mountain
meditation), informal exercises to integrate mindfulness skills in daily routines (e.g., eating,
brushing teeth), and other MBSR exercises (e.g. Pleasant and unpleasant events calendars,
Identification of nourishing activities, Stress diary), supplemented with selected MBCT exer-
cises (e.g., Automatic Thoughts Questionnaire, Developing an action plan, 3-Minute Breathing
Space).
All participants received a workbook including a general instruction to the training, home-
work instructions for each session, a form for registering (experiences with) homework, and
handouts related to the respective session themes selected from Dutch-language versions of the
handbooks by Kabat-Zinn (1990) and Segal et al. (2002). A central theme in the exercises was
nonjudgmental moment-to-moment awareness of cognitive, emotional, and bodily experiences,
fostering a decentered stance and a more adaptive response to negative thoughts and feelings
(Kabat-Zinn, 1982; Teasdale et al., 2000).

Trainers
Each of the mindfulness groups was led by one of two qualified mindfulness trainers. Both were
fully trained in MBSR and MBCT and certified by the Dutch Association for Mindfulness-based
Trainers (VMBN). Both had extensive meditation experience (15 years and 30 years, respectively)
and were engaged in ongoing meditation practices, in personal as well as professional contexts
(e.g., as teachers/supervisors at the Dutch Institute for Mindfulness and as leaders of retreats).
One of these trainers was a psychiatrist and psychotherapist, while the other was a specialized
psychiatric nurse and Vipassana meditation teacher who had spent 6 years in Asia as a Buddhist
monk. They had offered MBSR and MBCT in mental health care settings from the very first
introduction of these programs in Dutch mental health care (2005) and are considered two of
the Dutch pioneers in this regard. Both trainers were trained and supervised by experienced
teachers from the United States and the United Kingdom and had ongoing intervision with
other senior mindfulness trainers in the Netherlands.
In about half of the training groups, the mindfulness trainers were assisted by one of three co-
trainers (a nurse practitioner, a specialized psychiatric nurse, and a psychomotor therapist). All
co-trainers had many years of experience as mental health care workers and had received a basic
training in MBSR/MBCT for mental health care professionals. The co-trainers were supervised
by the two main trainers and had regular intervision with each other under supervision of one of
the main trainers. During these supervision and intervision meetings, adherence to the treatment
protocol was monitored and problems were discussed.

Procedure
Prior to the training, patients were invited to participate in the study by means of a letter.
After replying in the affirmative, they signed an informed consent and completed the pretraining
questionnaires at home. Immediately after the training they were given the same questionnaires
(posttreatment assessment) to be filled out at home and sent back by mail. Distribution and
collection of the questionnaires was done by the mindfulness trainers, as part of a routine
outcome measurement procedure. For this reason, no particular effort was made to increase
response percentages.
64 Journal of Clinical Psychology, January 2014

Statistical Analysis
Primary analyses were performed on data of the completers sample (N = 143). KIMS data were
available for only 100 patients; the KIMS questionnaire was not yet included in the study battery
during the first year of the study. Differences between completers and noncompleters on baseline
demographic and clinical variables were tested using independent t tests for continuous variables
and chi-square tests for categorical variables. Because a completers analysis may produce biased
estimates, sensitivity analyses were done in the intention-to-treat (ITT) sample (N = 214), using
the last observation carried forward (LOCF) approach. LOCF is usually applied only for subjects
who completed at least one follow-up measurement (Julious & Mullee, 2008), but because our
study had only one follow-up measurement, all patients with missing data would have to be
excluded, unless the baseline was included in the LOCF analysis (BOCF). Undertaking a BOCF
analysis is counterintuitive because baseline scores cannot be influenced by treatment, and
natural improvement due to regression to the mean and spontaneous recovery is not accounted
for in baseline scores (Julious & Mullee, 2008). So, BOCF is likely to produce rather conservative
results. In spite of this, we performed this sensitivity analysis to present a more balanced picture
of the results.
Based on their primary diagnosis, the patients were classified in five diagnostic categories:
depressive disorder, bipolar disorder, anxiety disorder, adjustment disorder, and other. Classifi-
cation of the groups was guided by the number of patients having a particular diagnosis; every
diagnosis was considered a separate category, except diagnoses that were too infrequent, which
were grouped into the category “Other”.
All continuous variables were tested for normality with the Kolmogorov-Smirnov test and by
inspection of the distribution plots. Non-normally distributed variables were logtransformed.
This was the case only for the variable “illness duration.” The effectiveness of the mindfulness
training was tested using repeated measures analysis of variance (ANOVA), in which the change
from pretreatment to posttreatment on the outcome measures was investigated in the entire
sample (using models with Time as predictor) as well as in the different diagnostic categories
(using models with Time, Diagnostic category, and the interaction Time × Diagnostic category as
predictors). In the latter model, a significant interaction between Time and Diagnostic category
implies that the effect of the training is different for different diagnostic categories. Post hoc
tests for effectiveness within diagnostic categories were performed only in case of a significant
interaction effect.
In a secondary series of analyses, we investigated whether observed differences in effectiveness
could be explained by baseline differences between the diagnostic categories. To analyze whether
the various diagnostic groups differed on baseline demographic and clinical variables, we used
independent t tests for continuous variables and chi-square tests for categorical variables. Vari-
ables that were unevenly distributed over the different diagnostic categories were subsequently
included in the repeated measures ANOVAs to investigate whether they reduced the significance
of the interaction Time × Diagnostic category. Analyses were done in SPSS 18. A two-tailed
alpha level of 0.05 was used.

Results
Patient Characteristics
Mean age of the completers sample was 45.5 years (standard deviation [SD] 10.5, range 21–
66). The majority of the participants were female (n = 99, 69%). Educational level of the
sample was rather high; almost half of the sample had a higher education (higher vocational
education/university; n = 68, 47.6%). The participants differed widely in illness duration: for
some participants, the mindfulness training was the first contact they had with a mental health
care institute; on the other hand, for one participant the first contact had been 34 years ago.
Median illness duration was 8 years. A small majority of the participants had a partner (n = 88,
61.5%).
 Mindfulness in a Heterogeneous Sample 65

Table 1
Overall Results for the Mixed-Patient Group (N = 143)

Pretreatment Posttreatment F-test for change

Outcome measure M (SD) M (SD) n F p Cohen’s d

SSL 15.2 (7.1) 11.8 (7.0) 139 34.9 <0.001 0.48

WHOQOL-Bref
Physical health 12.9 (2.6) 14.1 (2.8) 143 37.7 <0.001 0.46
Psychological health 12.3 (2.4) 13.1 (2.3) 142 25.7 <0.001 0.33
Social relationships 13.1 (3.5) 13.5 (3.3) 142 4.6 0.034 0.11
Environment 14.9 (2.4) 15.4 (2.5) 142 10.5 0.002 0.21
Overall QoL and GH 12.3 (3.0) 13.6 (3.3) 143 27.3 <0.001 0.43
KIMS
Observe 39.2 (7.6) 43.5 (6.0) 100 49.7 <0.001 0.57
Describe 26.1 (7.0) 28.4 (5.8) 99 28.5 <0.001 0.33
Act with awareness 25.6 (5.4) 28.9 (5.2) 99 31.4 <0.001 0.61
Accept without judgment 26.6 (6.5) 30.4 (7.6) 100 37.8 <0.001 0.58

Note. M = mean; SD = standard deviation; SSL = Short Symptom List; WHOQOL-Bref = World Health
Organization Quality of Life Bref; Overall QoL and GH = Overall Quality of life and General Health;
KIMS = Kentucky Inventory of Mindfulness Skills.

The most frequently occurring primary diagnosis in the completers sample was depressive dis-
order (single episode or recurrent; n = 45, 31.5%), followed by bipolar disorder (type I or II; n =
42, 29.4%), anxiety disorder (generalized anxiety disorder, panic disorder, obsessive-compulsive
disorder, posttraumatic stress disorder, social phobia, or anxiety disorder not otherwise speci-
fied; n = 17, 11.9%), and adjustment disorder (n = 14, 9.8%). The remaining participants (n =
25, 17.5%) had a primary diagnosis of another category (including psychotic disorder, relation-
ship problems, dysthymic disorder, personality disorder, attention deficit hyperactivity disorder,
identity problem, hypochondriasis, or occupational problem). A hundred and one patients were
diagnosed with one disorder, 35 patients had two disorders, and seven patients had three or more
disorders. Most frequently co-occurring diagnosis was a personality disorder (n = 18), followed
by anxiety disorder (n = 7), depressive disorder (n = 5), and alcohol abuse or dependency (in
remission) (n = 5).

Overall Outcome
We first examined outcome after the mindfulness training in the entire patient group. Table 1
presents the results. On average, the participants showed improvement on all outcome measures.
The repeated measures ANOVAs showed that these improvements were all significant. Effect
sizes (Cohen’s d), calculated by dividing the mean difference between the pretreatment and post-
treatment scores by the standard deviation of the pretreatment scores, were small to moderate.
Smallest effect size was observed for the Social Relationship subscale of the WHOQOL-Bref (d
= 0.11). Largest effect size was for the Act With Awareness subscale of the KIMS (d = 0.61).
As a sensitivity analysis, we repeated the analyses in the ITT sample (N = 214), with the
baseline observation carried forward for the subjects with missing values at the posttreatment
assessment. As could be expected, effect sizes were lower in the ITT analyses, though p-values
were essentially the same, presumably due to an increase in power. Significant improvement was
still observed on all outcome measures: SSL, F(1,209) = 32.2, p < 0.001; WHOQOL Physical
health, F(1,210) = 34.8, p < 0.001; WHOQOL Psychological health, F(1,212) = 24.3, p < 0.001;
WHOQOL Social relationships, F(1,212) = 4.5, p = 0.034; WHOQOL Environment, F(1,212)
= 10.2, p = 0.002; Overall QOL & GH, F(1,213) = 25.6, p < 0.001; KIMS Observe, F(1,180)
= 40.7, p < 0.001; KIMS Describe, F(1,186) = 25.2, p < 0.001; KIMS Act with awareness,
F(1,180) = 27.5, p < 0.001; and KIMS Accept without judgment, F(1,180) = 32.4, p < 0.001.
66 Journal of Clinical Psychology, January 2014

Table 2
Effect Sizes by Diagnostic Category

Pretreatment to posttreatment change (Cohen’s d)

Depression Bipolar Anxiety Adjustment Other

Outcome measure n = 45 n = 42 n = 17 n = 14 n = 25

SSL 0.55 0.13 0.76 0.59 0.70

WHOQOL
Physical health 0.40 0.30 0.55 0.62 0.64
Psychological health 0.36 0.16 0.52 0.75 0.52
Social relationships 0.29 0.02 0.15 0.00 0.17
Environment 0.29 0.01 0.29 0.48 0.03
Overall QoL and GH 0.40 0.06 0.62 0.87 0.91
KIMS n = 39 n = 15 n = 15 n = 11 n = 20
Observe 0.58 0.54 0.74 0.50 0.52
Describe 0.16 0.47 0.42 0.45 0.55
Act with awareness 0.47 0.16 0.86 0.97 0.72
Accept without judgment 0.56 0.27 1.00 1.65 0.35

Note. SSL = Short Symptom List; WHOQOL-Bref = World Health Organization Quality of Life Bref;
Overall QoL and GH = Overall Quality of life and General Health; KIMS = Kentucky Inventory of
Mindfulness Skills.

Differential Outcome for Different Diagnostic Categories?

Next, we investigated whether there was evidence for differential outcome for participants from
different diagnostic categories. The interaction Time × Diagnostic category was significant in the
models for SSL, F(4,134) = 2.8, p = 0.028, WHOQOL Environment, F(4,137) = 2.6, p = 0.040,
and Overall QoL and GH, F(4,138) = 3.3, p = 0.013. This implies that improvement with respect
to these subscales was different for the different diagnostic categories. Post hoc tests showed
that the bipolar category did not improve significantly on the SSL, WHOQOL Environment,
and Overall QoL and GH. The other diagnostic categories did improve significantly on these
outcome measures, with one exception: The patients from the Anxiety category did not improve
significantly on the Environment subscale of the WHOQOL.
Table 2 presents the effect sizes for the change from pretreatment to posttreatment for each
diagnostic category separately. This table shows substantial differences in effect sizes, also for
scales for which the interaction Time × Diagnostic category was not significant. However, we
did not perform post hoc tests for the latter scales because the absence of a significant interaction
between time and diagnostic category did not justify doing so. The absence of such an interaction
effect implies that there is no evidence for differential outcome with respect to these outcome
measures.

Potential Covariates Explaining the Difference in Outcome

Given the finding that participants of some diagnostic categories, particularly the bipolar, showed
less improvement than participants of other categories, we investigated whether there were
differences in demographic or clinical characteristics between the diagnostic categories that could
explain these differences in improvement. To this end, we first tested whether participants from
different diagnostic categories differed from each other with respect to age, gender, educational
level, having a partner, illness duration, or baseline severity on the outcome measures. There was
a clear and significant difference in illness duration, one-way ANOVA; overall F(4,126) = 5.47, p
< 0.001: Illness duration was higher in the bipolar category compared with the other diagnostic
categories. Further, we found a difference in the baseline scores on the SSL, overall F(4,135)
= 3.05, p = 0.019. Bipolar patients had lower SSL scores than the other diagnostic categories.
 Mindfulness in a Heterogeneous Sample 67

Thus, the bipolar category diverged from the other categories in having a longer illness duration
and a lower baseline symptom severity.
To investigate whether these differences could explain the differences in improvement after
the training, we included the variables for illness duration and baseline SSL scores as covariates
in the repeated measures ANOVA models to see whether this would reduce the interaction
Time × Diagnostic category. This was indeed the case. In all models, the interaction effect
was reduced and turned into nonsignificance after inclusion of the covariates. Illness duration
appeared to account for the largest part of this reduction. A significant interaction between
Illness duration and Time was found in the models for SSL, F(1,95) = 8.50, p = 0.004, and
WHOQOL Environment, F(1,95) = 6.15, p = 0.015, and a trend was found in the model for
Overall QoL and GH, F(1,95) = 3.41, p = 0.068. In the model for SSL, the baseline SSL score
also accounted for a substantial reduction in the interaction Time × Diagnostic category. A
significant interaction between Baseline SSL score and Time was found in this model, F(1,95)
= 38.0, p < 0.001. Inspection of the nature of these interaction effects revealed that participants
with longer illness duration showed less improvement on the SSL, WHOQOL Environment,
and Overall QoL and GH. Lower baseline SSL scores were related to less improvement on the
SSL. Thus, the lower effectiveness of the mindfulness training in the bipolar category could be
explained by the fact that participants from this category had a longer illness duration and a
lower baseline symptom severity.

Discussion
We examined outcome following mindfulness training in a heterogeneous psychiatric outpatient
population. The results showed that this mixed patient group improved on all outcome measures;
i.e., significant reductions in psychological symptoms were found, as well as improvement in five
domains of quality of life and in four different mindfulness skills. Differential improvement for
different diagnostic groups was observed for psychological symptoms and environmental and
overall quality of life. Bipolar patients showed less improvement on these outcomes compared
with patients from other diagnostic categories. These differences appeared to be attributable
to differences in illness duration and baseline symptom severity; in the bipolar patient group,
illness duration was longer and baseline severity was lower.

Mindfulness Training in a Heterogeneous Patient Group

Our study lends support to the notion that mindfulness training delivered in a heterogeneous
patient group can be effective. To our knowledge, only two other studies have reported previously
on the effectiveness of mindfulness training in a mixed-patient group (Green & Bieling, 2012; Ree
& Craigie, 2007). The sample sizes of these studies, however, were low (23 and 26, respectively),
so no direct comparisons could be made between diagnostic groups. Moreover, the diagnoses
included in these studies were all in the depression and anxiety spectrum. Our study group
was truly heterogeneous; it included patients with various types of diagnoses. The results are
in line with expectations of our mindfulness trainers, in whose clinical experience delivering
mindfulness training in such a heterogeneous group is both effective and feasible. In their
opinion, the combination of MBSR and MBCT elements in our program has added value in
such a mixed group because it allows for a flexible approach so that details of the program can
be adapted to the various patient needs.
The trainers also report some advantages of administering mindfulness in a heterogeneous
group; by hearing other patients talk about their complaints, participants tend to disengage
from their own complaints. As a result, the training is lifted to a more transdiagnostic level in
which “living life as a human being” rather than “living with this or that disorder” is the issue.
While the training may not be suitable for all psychiatric patients – patients with a current severe
depressive episode, psychotic episode, (hypo)mania, or substance dependence are excluded from
the training in our center – the results do support the notion that the scope of mindfulness
training may be expanded to a wider variety of patients and that group formats do not need to
be homogeneous.
68 Journal of Clinical Psychology, January 2014

Effectiveness in Different Diagnostic Groups

Our study showed significant improvements in patients from all diagnostic categories. Evidence
for the effectiveness of mindfulness training in depressed patients is not new; convincing results
have been reported in several studies (Fjorback et al., 2011; Piet & Hougaard, 2011; Van Aalderen
et al., 2012; Chiesa & Serretti, 2011), although these results mainly concerned remitted patients.
There are also some reports on the effectiveness of mindfulness training in patients with anxiety
disorders (Hofmann et al., 2010; Kim et al., 2010a, 2010b; Lovas & Barsky, 2010; Williams
et al., 2011), although this evidence is moderate or the studies had small sample sizes. Our results
add to this evidence, showing improvements after the training in the anxious patient group. We
have not found any effect studies of mindfulness training in patients with an adjustment disorder,
despite the fact that this disorder is rather common in both adolescents and adults and may
have severe consequences (Casey, 2009; Semprini, Fava, & Sonino, 2010). Casey argues that
underinvestigation of adjustment disorder treatment may be because of the fact that there are
problems with the diagnostic criteria. Our results show that patients diagnosed with adjustment
disorder do improve after mindfulness training.
Patients with bipolar disorder appeared to be atypical in our study. They seemed to derive the
least benefit from the training. Although they did improve on some of the measures, for example,
psychological QoL and mindfulness skills, effect sizes for this category were generally lower than
for the other diagnostic categories. On three of the outcome measures the improvements of the
bipolar patients were not significant at all. Our covariance analyses showed that this patient
group diverged from the other groups also in other aspects: Illness duration was relatively high
and baseline symptom severity was relatively low. These differences seemed to account for the
differences in outcome. So, our results do not necessarily imply that mindfulness training is
unsuitable for patients with bipolar disorder; the training might have led to improvements in
bipolar patients with a shorter history of mental health problems and/or more severe problems
at baseline.
Also, the value of the training for bipolar patients may lie in relapse prevention rather than
symptom reduction; clinical experience at our center suggests this may be the case, but the
follow-up time of our study was too short to substantiate this. The literature does present some
evidence for the effectiveness of mindfulness training in bipolar patients (Williams et al., 2008;
Deckersbach et al., 2012; Stange et al., 2011). None of these studies, however, had a no-treatment
control condition and sample sizes were low.
Overall effect sizes for the changes observed in our study were small to medium, depending
on the outcome measure (e.g., effect sizes were typically low for social relationships, which is
often the case in outcome studies because improvements in interpersonal skills take time to
produce an effect; Howard, Lueger, Maling, & Martinovich, 1993). Effect sizes calculated for
the different diagnostic categories were rather diverse, ranging from 0 to 1.65. We doubt whether
these differences are solely because of diagnostic differences. Our results suggest that illness
characteristics such as disease history and symptom severity are more important in predicting
the success of the training. Presumably, other baseline demographic and clinical characteristics
as well as differences in personality and motivation (e.g., willingness to perform daily home
practices) play a role (e.g., Bowen & Kurz, 2012; Shapiro, Oman, Thoresen, Plante, & Flinders,
2008; Vettese, Toneatto, Stea, Nguyen, & Wang, 2009). Therefore, allocation of patients to
mindfulness training should probably be guided by other indices than DSM-IV diagnoses. More
research is needed to substantiate this supposition.

Limitations and Strengths

A number of limitations of the current study should be mentioned, the absence of a control
group being the most obvious one. Because of this, we cannot be certain whether the observed
changes can be ascribed to the mindfulness training. However, for the comparison between the
different diagnostic categories, this point is less of a problem. Another limitation is the lack of
a follow-up assessment, which precludes conclusions about the sustainability of the observed
changes. A further weakness is that medication use was not controlled for.
 Mindfulness in a Heterogeneous Sample 69

Finally, it should be mentioned that there was a large number of noncompleters (33%). This
high attrition rate can largely be attributed to the fact that the study was performed in the context
of a routine outcome monitoring procedure. There was no special research team in charge of the
data collection and therefore no specific efforts were made to increase the number of responses.
In terms of routine outcome monitoring, the response percentage is actually rather high (e.g.
Happell, 2008). The fact that no differences were found between noncompleters and completers
and that the observed changes were still significant in ITT analyses with the baseline observation
carried forward strengthens our confidence in the validity of our results. A strong feature of the
study is its naturalistic design. All patients who were given the training were eligible for the
study and no restrictions were made with regard to other therapies or medication. Therefore,
generalizability of the results to real clinical practice is very high. Another strength is the fact
that in our study a direct comparison could be made between diagnostic categories. To our
knowledge, this is the first study in which this has been possible.
Clearly, our results need replication. It would be interesting to see future studies comparing
a mixed patient group with a homogeneous patient group, preferably in a randomized design.
This may yield better insight into whether mindfulness training in a heterogeneous patient
group is as effective as it is in a single diagnosis group. Further, future studies comparing
effectiveness in different diagnostic categories may benefit from a better match of patients
on characteristics like illness duration and baseline severity, and from including long-term
assessments.

References
Appelo, M., & Lange, A. (2007). Meten van klachten en veerkracht in vierentwintig vragen. Directieve
Therapie, 27(3), 197–204.
Baer, R. A. (2009). Self-focused attention and mechanisms of change in mindfulness-based treatment.
Cognitive Behaviour Therapy, 38, S1, 15–20.
Baer, R. A., Smith, G. T., & Allen, K. B. (2004). Assessment of mindfulness by self-report – The Kentucky
Inventory of Mindfulness Skills. Assessment, 11(3), 191–206.
Baer, R. A., Smith, G. T., Lykins, E., Button, D., Krietemeyer, J., Sauer, S., . . . Williams, J. M. (2008).
Construct validity of the five-facet mindfulness questionnaire in meditating and nonmeditating samples.
Assessment, 15(3), 329–342.
Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., Carmody, J., . . . Devins, G. (2004).
Mindfulness: A proposed operational definition. Clinical Psychology-Science and Practice, 11(3), 230–
241.
Bowen, S., Chawla, N., Collins, S. E., Witkiewitz, K., Hsu, S., Grow, J., . . . Marlatt, A. (2009). Mindfulness-
based relapse prevention for substance use disorders: a pilot efficacy trial. Substance Abuse, 30, 295–305.
Bowen, S., & Kurz, A. S. (2012). Between-session practice and therapeutic alliance as predictors of mind-
fulness after mindfulness-based relapse prevention. Journal of Clinical Psychology, 68(3), 236–245.
Casey, P. (2009). Adjustment disorder epidemiology, diagnosis and treatment. Cns Drugs, 23(11), 927–938.
Chiesa, A., & Serretti, A. (2011). Mindfulness based cognitive therapy for psychiatric disorders: A systematic
review and meta-analysis. Psychiatry Research, 187(3), 441–453.
De Vries, J., & Van Heck, G. L. (1996). De Nederlandse versie van de WHOQOL-bref. Tilburg: Universiteit
van Tilburg.
De Vries, J., & Van Heck, G. L. (Eds.). (2003). Nederlandse handleiding van de WHOQOL [Dutch Manual
of the WHOQOL]. Tilburg: Universiteit van Tilburg.
Deckersbach, T., Hoelzel, B. K., Eisner, L. R., Stange, J. P., Peckham, A. D., Dougherty, D. D., . . .
Nierenberg, A. A. (2012). Mindfulness-based cognitive therapy for nonremitted patients with bipolar
disorder. Cns Neuroscience & Therapeutics, 18(2), 133–141.
Fjorback, L. O., Arendt, M., Ornbol, E., Fink, P., & Walach, H. (2011). Mindfulness-based stress reduction
and mindfulness-based cognitive therapy–A systematic review of randomized controlled trials. Acta
Psychiatrica Scandinavica, 124(2), 102–119.
Green, S. M., & Bieling, P. J. (2012). Expanding the scope of mindfulness-based cognitive therapy: Evidence
for effectiveness in a heterogeneous psychiatric sample. Cognitive and Behavioral Practice, 19(1), 174–
180.
70 Journal of Clinical Psychology, January 2014

Grossman, P., Niemann, L., Schmidt, S., & Walach, H. (2004). Mindfulness-based stress reduction and
health benefits: A meta-analysis. Journal of Psychosomatic Research, 57, 35–43.
Happell, B. (2008). The value of routine outcome measurement for consumers of mental health services:
Master or servant? International Journal of Social Psychiatry, 54(4), 317–327.
Hofmann, S. G., Sawyer, A. T., Witt, A. A., & Oh, D. (2010). The effect of mindfulness-based therapy on
anxiety and depression: A meta-analytic review. Journal of Consulting and Clinical Psychology, 78(2),
169–183.
Howard, K. I., Lueger, R. J., Maling, M. S., & Martinovich, Z. (1993). A phase model of psychother-
apy outcome: Causal mediation of change. Journal of Consulting and Clinical Psychology, 61,
678–685.
Julious, S. A., & Mullee, M. A. (2008). Issues with using baseline in last observation carried forward analysis.
Pharmaceutical Statistics, 7, 142–146.
Kabat-Zinn J. (1982). An outpatient program in behavioral medicine for chronic pain patients based on
the practice of mindfulness meditation: Theoretical considerations and preliminary results. General
Hospital Psychiatry, 4, 33–47.
Kabat-Zinn, J. (1990). Full catastrophe living: Using the wisdom of your body and mind to face stress, pain
and illness. New York: Delacourt. (Dutch version: Handboek meditatief ontspannen [2005]. Altamira-
Becht, Haarlem.)
Kabat-Zinn, J. (2003). Mindfulness-based interventions in context: Past, present, and future. Clinical Psy-
chology: Science and Practice, 10, 144–156.
Kim, Y. W., Lee, S. H., Choi, T. K., Suh, S. Y., Kim, B., Kim, C. M., . . . Yook, K. H. (2010a). Effectiveness of
mindfulness-based cognitive therapy as an adjuvant to pharmacotherapy in patients with panic disorder
or generalized anxiety disorder. Depression & Anxiety, 26, 601–606.
Kim, B., Lee, S., Kim, Y. W., Choi, T. K., Yook, K., Suh, S. Y., . . . Yook, K. H. (2010b). Effectiveness of a
mindfulness-based cognitive therapy program as an adjunct to pharmacotherapy in patients with panic
disorder. Journal of Anxiety Disorders, 24(6), 590–595.
Kuyken, W., Watkins, E., Holden, E., White, K., Taylor, R. S., Byford, S., . . . Dalgleish, T. (2010).
How does mindfulness-based cognitive therapy work? Behaviour Research and Therapy, 48(11),
1105–1112.
Lange, A., & Appelo, M. (2007). De Korte Klachten Lijst handleiding [The Short Symptom List Users
Manual]. Houten: Bohn Stafleu Van Loghem.
Langer, A. I., Cangas, A. J., Salcedo, E., & Fuentes, B. (2012). Applying mindfulness therapy in a group of
psychotic individuals: A controlled study. Behavioural and Cognitive Psychotherapy, 40(1), 105–109.
Lovas, D. A., & Barsky, A. J. (2010). Mindfulness-based cognitive therapy for hypochondriasis, or severe
health anxiety: A pilot study. Journal of Anxiety Disorders, 24(8), 931–935.
McManus, F., Surawy, C., Muse, K., Vazquez-Montes, M., & Williams, M. J. (2012). A randomized clinical
trial of mindfulness-based cognitive therapy versus unrestricted services for health anxiety (hypochon-
driasis). Journal of Consulting and Clinical Psychology, 80, 817–828.
Piet, J., & Hougaard, E. (2011). The effect of mindfulness-based cognitive therapy for prevention of relapse in
recurrent major depressive disorder: A systematic review and meta-analysis. Clinical Psychology Review,
31(6), 1032–1040.
Piet, J., Hougaard, E., Hecksher, M. S., & Rosenberg, N. K. (2010). A randomized pilot study of mindfulness-
based cognitive therapy and group cognitive-behavioral therapy for young adults with social phobia.
Scandinavian Journal of Psychology, 51, 403–410.
Ree, M. J., & Craigie, M. A. (2007). Outcomes following mindfulness-based cognitive therapy in a hetero-
geneous sample of adult outpatients. Behaviour Change, 24(2), 70–86.
Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002). Mindfulness-based cognitive therapy for depression.
New York, NY: Guildford Press. (Dutch version: Aandachtgerichte cognitieve therapie [2004]. Uitgeverij
Nieuwezijds, Amsterdam.)
Semprini, F., Fava, G. A., & Sonino, N. (2010). The spectrum of adjustment disorders: Too broad to be
clinically helpful. CNS Spectrums, 15(6), 382–388.
Shapiro, S. L., Oman, D., Thoresen, C. E., Plante, T. G., & Flinders, T. (2008). Cultivating mindfulness:
Effects on well-being. Journal of Clinical Psychology, 64(7), 840–862.
Stange, J. P., Eisner, L. R., Hoelzel, B. K., Peckham, A. D., Dougherty, D. D., Rauch, S. L., . . . Dechersbach,
T. (2011). Mindfulness-based cognitive therapy for bipolar disorder: Effects on cognitive functioning.
Journal of Psychiatric Practice, 17(6), 410–419. CNS Spectrums.
 Mindfulness in a Heterogeneous Sample 71

Teasdale, J. D., Segal, Z., & Williams, J. M. G. (1994). How does cognitive therapy prevent depressive relapse
and why should attentional control (mindfulness) training help? Behaviour Research and Therapy, 33(1),
25–39.
Teasdale, J. D., Segal, Z. V., Williams, J. M. G., Ridgeway, V. A., Soulsby, J. M., & Lau, M. A. (2000).
Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. Journal
of Consulting and Clinical Psychology, 68(4), 615–623.
Trompenaars, F., Masthoff, E., Heck, G., Hodiamont, P., & Vries, J. (2005). Content validity, construct
validity, and reliability of the WHOQOL-bref in a population of dutch adult psychiatric outpatients.
Quality of Life Research, 14(1), 151–160.
Van Aalderen, J. R., Donders, A. R. T., Giommi, F., Spinhoven, P., Barendregt, H. P., & Speckens, A.
E. M. (2012). The efficacy of mindfulness-based cognitive therapy in recurrent depressed patients with
and without a current depressive episode: A randomized controlled trial. Psychological Medicine, 42(5),
989–1001.
Vettese, L., Toneatto, T., Stea, J., Nguyen, L., & Wang, J. (2009). Do mindfulness meditation participants do
their homework? And does it make a difference? A review of the empirical evidence. Journal of Cognitive
Psychotherapy, 23, 198–225.
Wanden-Berghe, R. G., Sanz-Valero, J., & Wanden-Berghe, C. (2011). The application of mindfulness to
eating disorders treatment: A systematic review. Eating Disorders, 19(1), 34–48.
Williams, J. M. G., Alatiq, Y., Crane, C., Barnhofer, T., Fennell, M. J. V., Duggan, D. S., . . . Goodwin,
G. M. (2008). Mindfulness-based cognitive therapy (MBCT) in bipolar disorder: Preliminary evaluation
of immediate effects on between-episode functioning. Journal of Affective Disorders, 107(1–3), 275–279.
Williams, M. J., McManus, F., Muse, K., & Williams, J. M. G. (2011). Mindfulness-based cognitive therapy
for severe health anxiety (hypochondriasis): An interpretative phenomenological analysis of patients’
experiences. British Journal of Clinical Psychology, 50, 379–397.
Winbush, N. Y., Gross, C. R., & Kreitzer, M. J. (2007). The effects of mindfulness-based stress reduction on
sleep disturbance: A systematic review. Explore-the Journal of Science and Healing, 3(6), 585–591.
Zgierska, A., Rabago, D., Chawla, N., Kushner, K., Koehler, R., & Marlatt, A. (2009). Mindfulness
meditation for substance use disorders: A systematic review. Substance Abuse: Official Publication of
the Association for Medical Education and Research in Substance Abuse, 30(4), 266–294.
Copyright of Journal of Clinical Psychology is the property of John Wiley & Sons, Inc. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.