Content uploaded by Anne E M Speckens
Author content
All content in this area was uploaded by Anne E M Speckens on Mar 28, 2019
Content may be subject to copyright.
Copyright 2016 American Medical Association. All rights reserved.
Efficacy of Mindfulness-Based Cognitive Therapy
in Prevention of Depressive Relapse
An Individual Patient Data Meta-analysis
From Randomized Trials
Willem Kuyken, PhD; Fiona C. Warren, PhD; Rod S.Taylor, PhD; Ben Whalley, PhD; Catherine Crane, PhD; Guido Bondolfi, MD, PhD;
Rachel Hayes, PhD; Marloes Huijbers, MSc; Helen Ma, PhD; Susanne Schweizer,PhD; Zindel Segal, PhD; Anne Speckens, MD;
John D. Teasdale, PhD; Kees Van Heeringen, PhD; Mark Williams, PhD; Sarah Byford, PhD; Richard Byng,PhD; Tim Dalgleish, PhD
IMPORTANCE Relapse prevention in recurrent depression is a significant public health
problem, and antidepressants are the current first-line treatment approach. Identifying an
equally efficacious nonpharmacological intervention would be an important development.
OBJECTIVE To conduct a meta-analysis on individual patient data to examine the efficacy of
mindfulness-based cognitive therapy (MBCT) compared with usual care and other active
treatments, including antidepressants, in treating those with recurrent depression.
DATA SOURCES English-language studies published or accepted for publication in
peer-reviewed journals identified from EMBASE, PubMed/Medline, PsycINFO, Web of
Science, Scopus, and the Cochrane Controlled Trials Register from the first available year to
November 22, 2014. Searches were conducted from November 2010 to November 2014.
STUDY SELECTION Randomized trials of manualized MBCT for relapse prevention in recurrent
depression in full or partial remission that compared MBCT with at least 1 non-MBCT
treatment, including usual care.
DATA EXTRACTION AND SYNTHESIS This was an update to a previous meta-analysis. We
screened 2555 new records after removing duplicates. Abstracts were screened for full-text
extraction (S.S.) and checked by another researcher (T.D.). There were no disagreements. Of
the original 2555 studies, 766 were evaluated against full study inclusion criteria, and we
acquired full text for 8. Of these, 4 studies were excluded, and the remaining 4 were
combined with the 6 studies identified from the previous meta-analysis, yielding 10 studies
for qualitative synthesis. Full patient data were not available for 1 of these studies, resulting in
9 studies with individual patient data, which were included in the quantitative synthesis.
RESULTS Of the 1258 patients included, the mean (SD) age was 47.1 (11.9) years, and 944
(75.0%) were female. A 2-stage random effects approach showed that patients receiving
MBCT had a reduced risk of depressive relapse within a 60-week follow-up period compared
with those who did not receive MBCT (hazard ratio, 0.69; 95% CI, 0.58-0.82). Furthermore,
comparisons with active treatments suggest a reduced risk of depressive relapse within a
60-week follow-up period (hazard ratio, 0.79; 95% CI, 0.64-0.97). Using a 1-stage approach,
sociodemographic (ie, age, sex, education, and relationship status) and psychiatric (ie, age at
onset and number of previous episodes of depression) variables showed no statistically
significant interaction with MBCT treatment. However, there was some evidence to suggest
that a greater severity of depressive symptoms prior to treatment was associated with a
larger effect of MBCT compared with other treatments.
CONCLUSIONS AND RELEVANCE Mindfulness-based cognitive therapy appears efficacious as a
treatment for relapse prevention for those with recurrent depression, particularly those with
more pronounced residual symptoms. Recommendations are made concerning how future
trials can address remaining uncertainties and improve the rigor of the field.
JAMA Psychiatry. doi:10.1001/jamapsychiatry.2016.0076
Published online April 27, 2016.
Editorial
Supplemental content at
jamapsychiatry.com
Author Affiliations: Author
affiliations are listed at the end of this
article.
Corresponding Author: Willem
Kuyken, PhD, Department of
Psychiatry,Warneford Hospital,
Prince of Wales International Centre,
University of Oxford, Oxford OX3
7JX, England (willem.kuyken
@psych.ox.ac.uk).
Research
Original Investigation |META-ANALYSIS
(Reprinted) E1
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
Although progress has been made in the treatment of
many psychiatric conditions, recurrent depressioncon-
tinues to cause significant disability and remains a high
cost to society.
1,2
Interventions that prevent depressive re-
lapse among people at high risk of recurrent episodes have
significant potential to reduce the disease’s burden.
3
Mind-
fulness-based cognitive therapy (MBCT), one such interven-
tion, teaches psychological skills that target cognitive mecha-
nisms implicated in depressive relapse to people with a history
of depression
4
by combining systematic mindfulness train-
ing with elements from cognitive therapy. A systematic re-
view and meta-analysis
5
of 6 randomized clinical trials (N = 593
patients) suggested that MBCT was associated with a signifi-
cant reduction in the rates of depressive relapse compared with
usual care or placebo, corresponding to a 34% relative risk re-
duction (risk ratio [RR], 0.66; 95% CI, 0.53-0.82).
While we have a growing body of evidence pointing to the
efficacy of MBCT in preventing depressive relapses, we do not
know whether MBCT is differentially efficacious for sub-
groups of people known to be at greater or lesser risk for de-
pressive relapse/recurrence.
6,7
Here, we present an analysis of individual patientdata (IPD)
compiled from 9 published randomized trials of MBCT iden-
tified through a systematic literature search. Unlike meta-
analyses of aggregate data at the trial level, IPD analyses per-
mit the investigation of patient-level characteristics that may
be potential moderators of treatment effects.
8
We examined
the efficacy of MBCT compared with usual care or active treat-
ment groups for patients from a rangeof soc iodemographic and
psychiatric backgrounds participating in studies conducted in
a number of countries in Europe and North America, taking
into account different periods of follow-up across studies.
Methods
The study was conducted in accordance with the Preferred Re-
porting Items for Systematic reviews and Meta-Analyses
statement
9
and the good practice guidelines of the Cochrane Col-
laboration IPD Methods Group
10
(eTable 1 in the Supplement).
Study Identification and Data Extraction
We searched for relevant publications from November 2010
(the searching end date of the previous meta-analysis,
5
which
performed searches from the first available date for each da-
tabase) to November 2014 (Figure 1) using the same a priori cri-
teria for study inclusion as the previous review, as follows: (1)
Study design: randomized trials of MBCT for the prevention
of relapse in patients with recurrent major depressive disor-
der currently in remission, reported in the English language,
and published or accepted for publication in peer-reviewed
journals; (2) Participants: participants aged 18 years or older,
diagnosed as having recurrent major depressive disorder in full
or partial remission according to a formal diagnostic classifi-
cation system (major depressive disorder was defined as a di-
agnosis based on the DSM-III, -III-R, -IV, or -IV-TR or the In-
ternational Statistical Classification of Diseases and Related
Health Problems, Tenth Revision [ICD-10]); (3) Intervention
group: MBCT delivered according to the treatment manual
11
;
(4) Control group: at least 1 non-MBCT treatment, including
usual care; and (5) Outcome measures: number of partici-
pants meeting the diagnostic criteria for a new major depres-
sive episode over the follow-up study period, according to
accepted clinical diagnostic criteria such as the ICD-10 or the
DSM-IV-TR.
Studies were identified from searches of titles, abstracts,
and keywords of electronic databases (EMBASE, PubMed/
Medline, PsycINFO, Web of Science, Scopus, and the Coch-
rane Controlled Trials Register) using the following search
string: (mindfulness-based cognitive therapy) OR (mindfulness
Figure 1. Preferred Reporting Items for Systematic Reviews
and Meta-analyses Flow Diagram From Record Identification
to Study Inclusion
1789 Reviews, qualitative studies, case
studies, dissertation abstracts,
study protocols, and non-English
articles were excluded
758 Records excluded (did not
meet inclusion criteria)
1Study excluded because individual
patient data were unavailable
7766 Records identified through
database searching
2555 Records after duplicates removed
766 Records screened against inclusion criteria
8Full-text articles assessed for eligibility
10 Studies included in qualitative synthesis
9Studies included in quantitative synthesis
4Full-text articles excluded
with reasons
6Studies identified in
previous meta-analysis
5
2Additional records identified
through other sources
Key Points
Question What is the efficacy of mindfulness-based cognitive
therapy compared with usual care and other treatments?
Findings This individual patient data meta-analysis included 9
trials, comprising 1329 participants. Patients receiving
mindfulness-based cognitive therapy had a significantly reduced
risk of depressive relapse within a 60-week follow-up period
compared with those who received usual care and had comparable
outcomes to those who received other active treatments.
Meaning Mindfulness-based cognitive therapy appears
efficacious as a treatment for relapse prevention for those who
have recurrent depression and provides an alternative choice to
other active treatments.
Research Original Investigation Efficacy of Mindfulness-Based Cognitive Therapy
E2 JAMA Psychiatry Published online April 27, 2016 (Reprinted) jamapsychiatry.com
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
based cognitive therapy) OR (MBCT) AND (depress*). No lan-
guage or other limitations were imposed at this stage. We also
checked reference lists of relevant studies and reviews for ad-
ditional references to potentially relevant studies. This pro-
cedure is summarized in Figure 1, and narrative text and an
example of a full search string are provided in eTable 2 in the
Supplement.
Individual patient data were obtained from the authors of
9 of the 10 trials meeting the inclusion criteria and collated into
1 data set (N = 1329). Overall, IPD integrity was found to be high.
The trials are summarized in Table 1, and data extraction and
cleaning are elaborated on in eTable 3 in the Supplement.Of
the 9 relevant trials, 2 had 3 arms and 7 had 2 arms. One trial
included a placebo pill arm
18
; this small arm (n = 30) was ex-
cluded from all analyses. The other 3-arm trial
23
had 2 non-
MBCT arms: one treatment as usual and the other treatment
as usual with cognitive psychological education. Forthe analy-
ses of MBCT vs non-MBCT, the 2 non-MBCT arms were com-
bined; for the analyses of MBCT vs an active comparator, the
treatment as usual arm was excluded. We used the Cochrane
risk of bias assessment tool.
24
While the risk of bias was gen-
erally low across all trials for most criteria (eTable 4 in the
Supplement), 2 of 9 trials did not blind assessors
17,19
and1of
these also had incomplete outcome data.
17
Primary Outcome
The primary outcome was relapse to depression within 60
weeks of follow-up, collected through a Structured Clinical Di-
agnostic Interview.
25
For studies with a follow-up beyond 60
weeks, follow-up was censored at 60 weeks. From the 9 trials
available, participants with data for relapse status and time to
relapse measured in weeks were included in all analyses; if re-
lapse did not occur, time to end of follow-upwas used. We also
reported adverse events.
Sociodemographic and Psychiatric Status Variables
We predefinedseveral soc iodemographic characteristics as po-
tential moderators of the effect of MBCT, ie, sex, age, educa-
tion, relationship status, race/ethnicity, socioeconomic sta-
tus, and employment status. These variables were standardized
across the 9 trials using available data to map each partici-
pant to the standardized category (eTable 3 in the Supple-
ment).
Psychiatric status variables included in the moderator
analyses were severity of depression symptoms at baseline
(measured with the Beck Depression Inventory–II or Inven-
tory of Depressive Symptomatology), baseline mindfulness
measured on 1 of several scales, age at onset of depression, and
number of previous major depressive episodes.
Statistical Methods
All statistical analyses were conducted according to partici-
pants’ randomized allocation in the primary studies. Only com-
plete case data were included for all trials in the main analy-
ses. In the event of substantive missing data (>10%) for an
individual trial, a sensitivity analysis was performed using im-
puted data based on 2 scenarios—one maximally favoring the
intervention group and the other maximally favoring the con-
trol group—for the 2-stage meta-analysis comparing MBCT with
non-MBCT only.All analyses were performed using Stata ver-
sion 14 (StataCorp LP).
Does MBCT Reduce Depressive Relapse/Recurrence
Compared With Control Conditions?
Meta-analyses of time-to-event data were used to evaluate the
effect of MBCT compared with non-MBCT on the primary out-
come. Both 2-stage and 1-stage meta-analysis methods were
used.
26
Two-stage methods involved calculating hazard ra-
tios (HRs) for depressive relapse (MBCT vs non-MBCT) for each
study individually
27,28
and using Cox proportional hazard mod-
els, and then combining these HRs in a meta-analysis. Hetero-
geneity was assessed within 2-stage models using the I
2
statistic.
29
A 95% CI for the I
2
statistic was calculated using the
test-based method.
30
Both fixed and random effect(s) mod-
els were applied.
31
Meta-analyses were performed on 3 pair-
wise comparisons: MBCT vs all non-MBCT treatments, MBCT
vs active treatments (antidepressantmedic ation [ADM] or cog-
nitive psychological education), and MBCT vs ADM only.
For 1-stage meta-analyses, both fixed and random ef-
fect(s) models were applied to the same 3 pairwise compari-
sons. Fixed effect models used the Cox proportional hazards
model to produce an HR; these models included each indi-
vidual study as a stratum with its own baseline hazard.
32
Where
the proportional hazards assumption was unsupported, MBCT
status interacting with log(time) was added to the model (and
to all subsequent models) to allow the effect of MBCT status
on risk of relapse to vary during the follow-up period. Ran-
dom effects 1-stage models used the Stata command stmixed,
33
included a study-level random effect on MBCT status, and
applied a flexible parametric survival model.
34
Are the Effects of MBCT on Outcomes Moderated
By Demographic or Depression-Related Variables?
For our primary outcome of depressive relapse, the use of
1-stage meta-analysis models facilitated inclusion of our so-
ciodemographic and depression-related covariates to inves-
tigate moderation.
35
The choice of whether to use a fixed ef-
fect or random effects approach would be informed by the
degree of heterogeneity between studies evident from the
2-stage and 1-stage models comparing MBCT with non-
MBCT; in the event of very low heterogeneity, a fixed effect
model would be used. A series of multivariable models were
created, initially including only the MBCT status of the par-
ticipant and 1 additional covariate (the interaction between
MBCT and log[time] was included if appropriate). As a fur-
ther check, all covariates were included in 1 overall model to
establish which were significantly associated with depres-
sive relapse in the presence of all other covariates. Individual
covariates that were found to be statistically significant at the
P< .10 threshold in a model including MBCT status only or in
a model with all covariates combined were then included in a
further model. Covariates that did not achieve significance at
the P< .05 level were removed individually from this new
model until the most parsimonious model had been ascer-
tained. Each covariate within this model was individually
investigated for interaction with MBCT status (ie, each model
Efficacy of Mindfulness-Based Cognitive Therapy Original Investigation Research
jamapsychiatry.com (Reprinted) JAMA Psychiatry Published online April 27, 2016 E3
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
Table 1. Description of the 9 Primary Studies of MBCT vs Non-MBCT Treatment
Source; Study Population
and Country
Arm: No. of
Participants
Randomized
Patients With
60-wk Relapse
Status and Time
to Relapse, No.
(%)
Person-wk
Contributed to
Unadjusted
Analyses
Patients With
Final BDI
(Closest
Available to 60
wk)
a
Baseline
BDI
b
,No.,
Mean (SD)
Mindfulness
Measure
Used
Patients With
Pre- and
Posttreatment
Mindfulness
Score
a
SAEs/SARs
Teasdale et al,
7
2000;
Community adults with
history of depression,
currently in full remission,
not receiving ADM at
assessment;
United Kingdom and
Canada
TAU: 69;
MBCT: 76
Non-MBCT:
66 (96);
MBCT: 70 (92);
total % missing: 6
Non-MBCT:
2363;
MBCT: 3093
Non-MBCT: 65;
MBCT: 65
132,
11.4 (7.9)
EQ
12
Non-MBCT:
12;
MBCT: 14
Not formally
recorded
Ma and Teasdale,
6
2004;
Community adults with
history of depression,
currently in full remission,
not receiving ADM at
assessment;
United Kingdom
TAU: 38;
MBCT: 37
Non-MBCT:
37 (97);
MBCT: 36 (97);
total % missing: 3
Non-MBCT:
1237;
MBCT: 1770
Non-MBCT: 33;
MBCT: 34
73,
13.9 (8.4)
EQ
12
Non-MBCT:
31;
MBCT: 32
Not formally
recorded
Kuyken et al,
13
2008;
Community adults with
history of ≥3 episodes of
depression, currently in
remission, receiving ADM;
United Kingdom
ADM: 62;
MBCT: 61
Non-MBCT:
62 (100);
MBCT: 61 (100)
Non-MBCT:
2271;
MBCT: 2592
Non-MBCT: 58;
MBCT: 59
123,
19.3 (11.9)
KIMS
14
Non-MBCT:
58;
MBCT: 55
No SARs in either
arm
Bondolfi et al,
15
2010;
Community adults with
history of ≥3episodes of
depression, currently in
remission, not receiving
ADM at assessment;
Switzerland
TAU: 29;
MBCT: 31
Non-MBCT:
29 (100);
MBCT: 31 (100)
Non-MBCT:
1205;
MBCT: 1386
Non-MBCT: 26;
MBCT: 26
60,
9.9 (9.0)
MAAS
16
Non-MBCT:
29;
MBCT: 28
Not formally
recorded; author
communication
that none was
recorded
Godfrin and van
Heeringen,
17
2010;
Community adults with
history of ≥3episodes of
depression, currently in
remission, both receiving
and not receiving ADM at
assessment;
Belgium
TAU: 54;
MBCT: 52
Non-MBCT:
47 (87);
MBCT: 40 (77);
total % missing:
18
Non-MBCT:
1690;
MBCT: 1964
Non-MBCT: 40;
MBCT: 35
86,
19.9 (12.2)
MAAS
16
Non-MBCT:
47;
MBCT: 37
Not formally
recorded
Segal et al,
18
2010
c
;At
point of randomization to
MBCT, community adults
with history of depression,
currently in full remission
after8moof
algorithm-informed ADM in
an earlier study phase;
Canada
Maintenance
ADM: 28;
MBCT after
discontinued
ADM: 26
Non-MBCT:
28 (100);
MBCT: 26 (100)
Non-MBCT:
1002;
MBCT: 1007
Non-MBCT: 7;
MBCT: 11
51,
4.0 (3.9)
MAAS
16
Non-MBCT:
10;
MBCT: 15
1 SAE in acute
phase (ADM arm)
and in the
follow-up phase;
0 SAEs in either
arm of the trial
Huijbers et al,
39
2012, and
Huijbers et al,
19
2015
d
;
Community adults with
history of ≥3 episodes of
depression, currently in
remission, receiving ADM;
The
Netherlands
Maintenance
ADM: 35;
MBCT and
ADM: 33
Maintenance
ADM: 35 (100);
MBCT and ADM:
33 (100)
Non-MBCT:
1342;
MBCT: 1433
Non-MBCT: 28;
MBCT: 28
68,
12.1 (9.6)
FFMQ
20
Non-MBCT:
27;
MBCT: 26
No SARs in either
arm
Kuyken et al,
21
2015;
Community adults with
history of ≥3 episodes of
depression, currently in
remission, receiving ADM;
United Kingdom
Maintenance
ADM: 212;
MBCT and
discontinued
ADM: 212
Maintenance
ADM: 202 (95);
MBCT and
discontinued
ADM: 200 (94);
total % missing: 5
Non-MBCT:
8882;
MBCT: 9471
Non-MBCT:
157;
MBCT: 167
396,
14.1 (10.2)
FFMQ
22
Non-MBCT:
169;
MBCT: 173
10 SAEs (5 MBCT
arm; 5 in ADM
arm), none of
which were
judged as SAR
Williams et al,
23
2014;
Community adults with
history of ≥3 episodes of
depression, currently in
remission, receiving ADM;
United Kingdom
TAU: 56;
CPE: 110;
MBCT: 108
TAU: 53 (95);
CPE: 103 (94);
MBCT: 99 (92);
total % missing: 7
Non-MBCT:
6022;
MBCT: 4199
Non-MBCT:
135;
MBCT: 88
255,
8.0 (7.8)
FFMQ
22
Non-MBCT:
138;
MBCT: 87
15 SAEs (5 MBCT
arm; 10 in CPE
arm), of which
1 (CPE) was
judged as SAR
Abbreviations: ADM, antidepressant medication; BDI, Beck Depression
Inventory; CPE, cognitive psychological education; EQ, Experiences
Questionnaire; FFMQ, Five Facet Mindfulness Questionnaire; KIMS, Kentucky
Inventory of Mindfulness Scale; MAAS, Mindful Attention Awareness Scale;
MBCT,mindfulness-based cognitive therapy; SAEs, serious adverse events;
SARs, serious adverse reactions; TAU,treatment as usual.
a
Primary outcome data available.
b
Includes all participants irrespective of trial arm.
c
Placebo arm excluded.
d
Huijbers et al
19
used Inventory of Depressive Symptomatology
(Clinician-Rated).
Research Original Investigation Efficacy of Mindfulness-Based Cognitive Therapy
E4 JAMAPsychiatry Published online April 27, 2016 (Reprinted) jamapsychiatry.com
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
included only 1 interaction term), and any that were not found
to be a significant predictor of time to relapse were individu-
ally included in the model with all other significant predic-
tors to investigate potential interaction with MBCT status. In
addition, moderation effects between each MBCT status and
each individual covariate were investigatedin a series of mod-
els including only MBCT status, the specified covariate, and
their interaction.
Results
Description of Primary Studies
The 9 included studies are described fully in the original trial
reports and are summarized here in Table 1. We defined loss
to follow-up as a lack of data on relapse status after 60
weeks (or closest available time) of follow-up. Of the 1329
randomized participants from the 9 trials with available IPD,
data on relapse status and time to relapse (or end of
follow-up with no relapse) were available for 1258 partici-
pants (94.7%). Across the sample of 1258 participants, the
mean (SD) age was 47.1 (11.9) years (median, 47 years; inter-
quartile range, 39-56), and 944 (75.0%) were female. Of 1230
participants with data available, 509 (41.4%) had at least
degree-level qualifications, 636 (51.7%) had qualifications
below degree level, and 85 (6.9%) had no qualifications. Of
1239 participants, 726 (58.6%) were married or had a part-
ner, 234 (18.9%) were single, and 279 (22.5%) were divorced,
separated, or widowed. Among 1234 participants, the mean
(SD) age at onset of depressions was 26.0 (12.2) years (me-
dian, 23 years; interquartile range, 17-34), and of 1200 par-
ticipants, 694 (57.8%) had 5 or more past depressive epi-
sodes. Within individual studies, the proportion of
participants lost to follow-up ranged from 0% to 18%
(Table 1). Of 596 participants who received MBCT, 229 (38%)
had a depressive relapse within 60 weeks’ follow-up,
whereas 327 of 662 participants (49%) who did not receive
MBCT had a depressive relapse within 60 weeks.
Does MBCT Reduce Depressive Relapse Compared
With Control Conditions?
Owing to clinical heterogeneity across the 9 studies, the re-
sults of the random effects models are reported; because of very
low heterogeneity of treatment effects between studies, the
results of equivalent fixed effect analyses were very similar.
A forest plot of the 2-stage meta-analysis of time to relapse of
depression compared MBCT with all non-MBCT treatments
(HR, 0.69; 95% CI, 0.58-0.82; I
2
, 1.7%; 95% CI, 0-20)
(Figure 2A). The funnel plot associated with this analysis in-
dicated some asymmetry, with an absence of smaller studies
that showed an increased risk of relapse with MBCT treat-
ment (eFigure 1 in the Supplement). The associated Egger test
produced a Pvalue of .18, although we recognize the limited
power of this test with only 9 studies. A sensitivity analysis
whereby missing outcome data from Godfrin and van
Heeringen
17
were imputed favoring the MBCT group pro-
duced an HR of 0.63 (95% CI, 0.49-0.82); using imputed data
that favored the non-MBCT group produced an HR of 0.74 (95%
CI, 0.63-0.88). An equivalent analysis comparing MBCT with
all active treatments was conducted using data from 5
studies
13,18,19,21,23
(HR, 0.79; 95% CI, 0.64-0.97; I
2
, 0%)
(Figure 2B). An analysis comparing MBCT with ADM treat-
ment was conducted using data from 4 studies
13,18,19,21
(HR,
0.77; 95% CI, 0.60-0.98; I
2
, 0%) (Figure 2C). For the latter 2
meta-analyses, the I
2
value was 0% in both cases, the lower
boundary of the 95% CI was 0%, and the upper boundaries
were 43% and 65%, respectively.
An unadjusted 1-stage fixed effect meta-analysis com-
pared MBCT with non-MBCT treatment (1248 patients, 554 de-
pressive relapses within 60 weeks; HR, 0.69; 95% CI, 0.58-
0.82) (Table 2, model A). However, evidence indicated that the
proportional hazards assumption was not valid (eFigure 2 in
the Supplement shows the log-log plots comparing the MBCT
and non-MBCT groups for each of the 9 included studies).
Owing to the lack of proportional hazards, the interaction be-
tween MBCT status and log(time) was added, allowing the
effects of MBCT to vary with log(time). This model (Table 2,
model B) yielded an HR for MBCT of 0.34 (95% CI, 0.19-0.60),
and for the interaction of MBCT with log(time) of 1.28 (95%
CI, 1.06-1.55), this model indicated a reduction in the preven-
tive effect of MBCT on depressive relapse as time progressed
during the follow-up period.
A 1-stage fixed effect model using 5 studies
13,18,19,21,23
com-
pared MBCT with active treatments only (892partic ipants and
385 relapses; HR, 0.78; 95% CI, 0.64-0.96) (Table 2, model C)
and was very similar to the 2-stage random effects model,
which provided little evidence to indicate lack of propor-
tional hazards. The equivalent analysis comparing MBCT with
ADM treatment used 4 studies
13,18,19,21
(637 participants and
266 relapses; HR, 0.77; 95% CI, 0.60-0.98) (Table 2, model D)
and was identical to the results of the 2-stage random effects
model, also with little evidence to support lack of propor-
tional hazards.
The 1-stage random effects model compared MBCT with
all non-MBCT treatments using a flexible parametric model
with 2 df (HR, 0.68; 95% CI, 0.58-0.81; between-study SD,
0.0008) (Table2, model E). A further model comparing MBCT
with non-MBCT was created by adding the interaction be-
tween MBCT status and the restricted cubic splines derived
from the previous model (HR, 0.63; 95% CI, 0.53-0.76; be-
tween-study SD, 0.0007)(Table 2, model F); the global Pvalue
for the interaction between MBCT status and each restricted
cubic spline was .04, consistent with a significant time-
varying effect of MBCT status observed in the fixed effect
model.
Equivalent analyses comparing MBCT with all active treat-
ments and comparing MBCT with ADM, with or without a
time-varying effect on MBCT status, failed to converge,
almost certainly owing to very low heterogeneity between
studies.
Are the Effects of MBCT on Outcomes Moderated
By Demographic and Depression-Related Variables?
In view of the low heterogeneity between studies, fixed ef-
fect 1-stage models were used for the moderation analyses. In-
dividually,5 soc iodemographic and psychiatric variables were
Efficacy of Mindfulness-Based Cognitive Therapy Original Investigation Research
jamapsychiatry.com (Reprinted) JAMA Psychiatry Published online April 27, 2016 E5
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
found to be significantly associated with risk of relapse within
60 weeks: baseline depression zscore, baseline mindfulness
zscore, age at onset, number of previous episodes, and
marital status (all P< .10). With the exception of marital sta-
tus, all of these covariates were also significantly associated
with time to relapse when included in a model with MBCT
status and its interaction with log(time) and all other covari-
ates. When included in a model with MBCT status and its
interaction with log(time), only 4 remained statistically sig-
nificant: baseline depression zscore, baseline mindfulness
zscore, age at onset, and number of previous episodes.
However, on including these 4 covariates, the interaction
between MBCT status and log(time) was no longer signifi-
cant (P= .052), so it was removed from the model. Thus, the
significant predictors of depressive relapse were baseline
depression zscore, baseline mindfulness zscore, age at
onset, and number of previous episodes. When including
the interaction with MBCT and each predictor in turn into
this model, baseline depression zscore had a significant
interaction with MBCT status (Table 2, model G; Figure 3);
patients with a higher baseline depression zscore received
greater benefit from MBCT therapy compared with all non-
MBCT treatments. Of the remaining significant covariates,
only baseline mindfulness zscore had a significant interac-
tion with MBCT status both in a model with no other covari-
ates and in a model with all other significant covariates.
However, these interactions became nonsignificant when
the interaction between MBCT status and baseline depres-
sion zscore was added to the model. No other covariates
were found to have a significant interaction with MBCT sta-
tus when included in a model with all other significant
covariates or in a model with only the respective covariate,
MBCT status, their interaction, and the interaction between
MBCT status and log(time).
Figure 2. Random Effects Meta-analyses Comparing Mindfulness-Based Cognitive Therapy(MBCT)
With Other Variables
Weight, %Study
Hazard Ratio
(95% CI)
12.8Teasdale et al,7 2000 0.64 (0.40-1.03)
6.5Ma and Teasdale,6 2004 0.45 (0.23-0.88)
11.6Kuyken et al,13 2008 0.66 (0.40-1.08)
3.6Bondolfi et al,15 2010 0.77 (0.31-1.90)
6.6Godfrin and van Heeringen,17 2010 0.34 (0.17-0.66)
5.6Segal et al,18 2010 0.74 (0.36-1.52)
4.5Huijbers et al,19 2015 0.80 (0.36-1.78)
28.4Kuyken et al,21 2015 0.81 (0.59-1.10)
20.4Williams et al,23 2014 0.82 (0.57-1.20)
100Overall (I2
=
1.7%, P
=
.42) 0.69 (0.58-0.82)
MBCT vs no MBCT
A
Weight, %Study
Hazard Ratio
(95% CI)
17.4Kuyken et al,13 2008 0.66 (0.40-1.08)
6.4Segal et al,18 2010 0.80 (0.35-1.82)
6.7Huijbers et al,19 2015 0.80 (0.36-1.78)
43.7Kuyken et al,21 2015 0.81 (0.59-1.11)
25.8Williams et al,23 2014 0.85 (0.56-1.28)
100Overall (I2
=
0.0%, P
=
.96) 0.79 (0.64-0.97)
MBCT vs any active treatment
B
Weight, %Study
Hazard Ratio
(95% CI)
23.4Kuyken et al,13 2008 0.66 (0.40-1.08)
8.6Segal et al,18 2010 0.80 (0.35-1.82)
9.1Huijbers et al,19 2015 0.80 (0.36-1.78)
58.9Kuyken et al,21 2015 0.81 (0.59-1.11)
100Overall (I2
=
0.0%, P
=
.92) 0.77 (0.60-0.98)
MBCT vs antidepressants
C
1.0 2.00.1 0.5
0.5
0.5
Hazard Ratio (95% CI)
1.0 2.00.1
Hazard Ratio (95% CI)
1.0 2.00.1
Hazard Ratio (95% CI)
Forest plot of 2-stage meta-analysis
of aggregate data on hazard ratio
scale comparing (A) risk of relapse of
depression in participants receiving
MBCT with participants not receiving
MBCT;(B) risk of relapse of
depression in participants receiving
MBCT with participants receiving an
alternative active therapy; and (C)
risk of relapse of depression in
participants receiving MBCT with
participants receiving antidepressant
medication. Weights are from
random effects analyses.
Research Original Investigation Efficacy of Mindfulness-Based Cognitive Therapy
E6 JAMAPsychiatry Published online April 27, 2016 (Reprinted) jamapsychiatry.com
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
Discussion
Summary of Results
Replicating previous work, we found clear evidence that MBCT
was associated with a significant reduction in the risk of de-
pressive relapse/recurrence over 60 weeks compared with
usual care. Extending previous work, we found that MBCT re-
duces the risk of depressive relapse/recurrence compared with
the current mainstay approach, maintenance antidepres-
sants. We further showed that there is no support for MBCT
having differential effects for patients based on their sex, age,
education, or relationship status, suggesting the interven-
tion’sgeneralizability across these characteristics. Different re-
search groups conducted the 9 randomized clinical trials and
used different clinicians across a range of European and North
American countries. The lack of heterogeneity between stud-
ies in effects on time to depressive relapse suggests that the
effects of MBCT are similar in these different contexts.
Mindfulness-based cognitive therapy was developed for
patients in remission but at high risk for depressive relapse/
recurrence. Our analyses suggest that the treatment effect of
MBCT on the risk of depressive relapse/recurrence is larger in
participants with higher levels of depression symptoms at base-
line compared with non-MBCT treatments, suggesting that
MBCT may be particularly helpful to those who still have sig-
nificant depressive symptoms. This is consistent with several
recent trials that suggest MBCT may be more effective for
people whose depressive symptoms fluctuate
18
and/or who re-
port a history of early adversity.
21,23
Adverse events were for-
mally recorded in 6 of 9 studies, but none were attributed to
MBCT.
Strengths and Limitations of the Study
To address the question of whether treatment effects are in-
fluenced by individual patient characteristics, a study needs
to be adequately powered and use appropriate statistical ap-
proaches. Within the constraints of the constituentstudies, our
Table 2. CoxPropor tional Hazards RegressionModels and Flexible Parametric Survival Models
Covariate No. HR (95% CI) PValue
Model A
a
Participants 1248 NA NA
Depressive relapses 554 NA NA
MBCT status (reference: non-MBCT) NA 0.69 (0.58-0.82) <.001
Model B
a
Participants 1248 NA NA
Depressive relapses 554 NA NA
MBCT status (reference: non-MBCT) NA 0.34 (0.19-0.60) <.001
MBCT by log(time)
b
NA 1.28 (1.06-1.55) .01
Model C
a
Participants 892 NA NA
Depressive relapses 385 NA NA
MBCT status (reference: active treatments) NA 0.78 (0.64-0.96) .02
Model D
a
Participants 637 NA NA
Depressive relapses 266 NA NA
MBCT status (reference: antidepressant medication) NA 0.77 (0.60-0.98) .03
Model E
c
Participants 1248 NA NA
Depressive relapses 554 NA NA
Between-study SD 0.0008 NA NA
MBCT status (reference: non-MBCT) NA 0.68 (0.58-0.81) <.001
Model F
d
Participants 1248 NA NA
Depressive relapses 554 NA NA
Between-study SD 0.0007 NA NA
MBCT status (reference: non-MBCT) NA 0.63 (0.53-0.76) <.001
Model G
a,e
Participants 1022 NA NA
Depressive relapses 443 NA NA
MBCT status (reference: non-MBCT) NA 0.74 (0.61-0.90) .003
Baseline depression score NA 1.40 (1.24-1.58) <.001
Baseline depression score by MBCT status NA 0.80 (0.66-0.97) .02
Abbreviations: HR, hazard ratio;
MBCT,mindfulness-based cognitive
therapy; NA, not applicable.
a
Cox proportional hazards regression
model stratified by individual study.
b
Time measured in weeks.
c
Flexible parametric model with 2 df
and random treatment effects.
d
Based on model E, with the
inclusion of interaction between
MBCT status and restricted cubic
splines to account for the
time-varying effect of MBCT
(P= .04).
e
Model adjusted for baseline
mindfulness zscore, age at onset of
depression, and number of past
episodes of depression (5 or more/
4 or fewer).
Efficacy of Mindfulness-Based Cognitive Therapy Original Investigation Research
jamapsychiatry.com (Reprinted) JAMA Psychiatry Published online April 27, 2016 E7
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
IPD approach provided an opportunity to answer these ques-
tions. Risk of bias was low, suggesting confidence in these find-
ings. Combining a series of modest-sized trials with effects in
the predicted direction but missing significance individually
yields a significant combined estimate of effect.
We did observe asymmetry in the funnel plot with an ab-
sence of smaller studies that showed an increased risk of re-
lapse with MBCT treatment. It is possible that there are un-
published studies that we are not aware of, and we welcome
investigators of any such studies to bring them to our atten-
tion so that their data can be included in future updates. The
unavailability of the Meadows et al
36
study data represents an
impediment to IPD, transparency, and external scrutiny. Fund-
ing bodies, ethics committees, and sponsors should work to
a consensus position. Finally, allegiance effects can influ-
ence effect sizes in psychological therapy trials,
37
and the
constituent trials were largely conducted by proponents of
MBCT. Therefore, we analyzed risk of bias (eTable 4 in the
Supplement).
There were a number of limitations resulting from avail-
ability of data within the constituent studies. For example, we
were not able to obtain information about important poten-
tial moderators such as race/ethnicity and employment. Trials
also vary in the way data are collected. For example, age at
first onset of depression was collected in some trials by
simple self-report and in others through standardized Struc-
tured Clinical Interview. Number of prior episodes was also
gathered inconsistently. Adverse events were not systemati-
cally recorded or reported. As with all meta-analyses, there
may be trials published in other languages or unpublished
trials we were not able to access. Moderator analyses were
not formally powered, exploratory, or relatively large in num-
ber, increasing the risk of type I errors. Future studies should
plan and power for moderator analyses.
Conclusions
While previous research has shown the superiority of MBCT
compared with usual care,
5
this study provides important new
evidence that MBCT is also effective compared with other
active treatments and that its effects are not restricted to
particular groups defined by age, educational level, marital
status, or sex. A recent meta-analysis
38
of the effectiveness
of all psychological interventions to prevent recurrence
compared with usual care and antidepressants suggests that
the protective effects of MBCT are comparable with those
for cognitive therapy (vs usual care: RR, 0.68; 95% CI, 0.54-
0.87; vs ADM: RR, 0.08; 95% CI, 0.61-1.02) and interper-
sonal therapy (vs usual care: RR, 0.41; 95% CI, 0.27-0.63; vs
ADM: RR, 0.08; 95% CI, 0.50-1.38). However, MBCT
addresses a particular clinical problem, namely teaching
skills to stay well to people currently well but at high risk of
depressive relapse. There is a reduction in protective effects
over time. The finding that MBCT may be most helpful for
patients with higher levels of depressive symptoms adds to
an emerging consensus that the greater the risk for depres-
sive relapse/recurrence, the more benefit MBCT offers.
Patients with lower baseline scores appeared to receive less
benefit but were not disadvantaged by MBCT.
We recommend that future trials consider an active con-
trol group, use comparable primary and secondary outcomes
(Structured Clinical Interview for DSM for depressive
relapse), use longer follow-ups, report treatment fidelity, col-
lect key background variables (eg, race/ethnicity and employ-
ment), take care to ensure generalizability, conduct cost-
effectiveness analyses, put in place ethical and data
management procedures that enable data sharing, consider
mechanisms of action, and systematically record and report
adverse events.
ARTICLE INFORMATION
Submitted for Publication: September 15, 2015;
final revision received January 13, 2016; accepted
January 15, 2016.
Published Online: April 27, 2016.
doi:10.1001/jamapsychiatry.2016.0076.
Author Affiliations: Department of Psychiatry,
University of Oxford, Prince of Wales International
Centre, Warneford Hospital, Oxford, England
(Kuyken, Crane, Ma, Williams); Institute of Health
Research, Primary Care Research Group, Exeter
Medical School, Exeter,England (Warren, Taylor);
School of Psychology,Faculty of Health and Human
Sciences, University of Plymouth, Plymouth,
England (Whalley); Department of Psychiatry,
University Medical Centre, University of Geneva,
Geneva, Switzerland (Bondolfi); Institute of Health
Research, Child Health Group, Exeter Medical
School, Exeter,England (Hayes); Department of
Psychiatry,Radboud University Nijmegen Medical
Centre, Radboud University Nijmegen, Nijmegen,
The Netherlands (Huijbers, Speckens); Hong Kong
Centre for Mindfulness, Hong Kong (Ma, Dalgleish);
Medical Research Council Cognition and Brain
Sciences Unit, Cambridge, England (Schweizer,
Teasdale, Dalgleish);Depar tment of Psychology,
University of TorontoScarborough, Toronto,
Ontario, Canada (Segal); University Department of
Psychiatry,University Hospital, Gent, Belgium (Van
Heeringen); King’s Health Economics, King’s
College London, London, England (Byford);
Peninsula School of Medicine, Plymouth University,
Plymouth, England (Byng); Cambridgeshire and
Peterborough NHS Foundation Trust, Cambridge,
England (Dalgleish).
Figure 3. Interactive Effect Between Mindfulness-Based Cognitive
Therapy (MBCT) Status and Baseline Depression With Regard
to the Relative Hazard of Depressive Relapse
3.0
2.5
2.0
1.5
1.0
0.5
0
–2.5 2.52.00 1.0 1.5
Relative Hazard of Depressive Relapse
Depression z Score
–2.0 –1.5 –1.0
No MBCT
MBCT
Predictive margins for the relative hazard of depressive relapse comparing
participants receiving MBCT with those not receiving MBCT at baseline
depression zscores, derived from a model including MBCT status, baseline
depression zscore, the interaction between MBCT status and baseline
depression zscore, baseline mindfulness zscore, age at onset of depression,
and number of past episodes of depression (5 or more/4 or fewer).
Research Original Investigation Efficacy of Mindfulness-Based Cognitive Therapy
E8 JAMAPsychiatry Published online April 27, 2016 (Reprinted) jamapsychiatry.com
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
Author Contributions: Dr Kuyken had full access to
all of the data in the study and takes responsibility
for the integrity of the data and the accuracy of the
data analysis.
Study concept and design: Kuyken, Warren, Taylor,
Whalley,Dalgleish.
Acquisition, analysis, or interpretation of data:
Dalgleish, Schweizer,Warren.
Drafting of the manuscript: Kuyken, Warren, Taylor,
Whalley,Crane, Schweizer, Dalgleish.
Critical revision of the manuscript for important
intellectual content: All authors.
Study supervision: Taylor.
Conflict of Interest Disclosures: All authors with
the exception of Drs Warren (independent
statistician) and Schweizer (independent
systematic reviewer) were investigators on 1 or
more of the original randomized clinical trials that
contributed data to the individual patient data and
secured grant funding for these trials. Dr Williams
founded the Oxford Mindfulness Centre and was its
director until 2013. Dr Kuyken is its current director.
Dr Speckens is founder and clinical director of the
Radboud UMC Centre for Mindfulness and Dr Ma is
director of the Centre for Mindfulness, Hong Kong.
Dr Crane and Ms Huijbers are affiliated with the
Oxford and Radboud University–based mindfulness
centers, respectively. Drs Teasdale, Williams, and
Segal receive royalties for books on mindfulness-
based cognitive therapy that they have coauthored.
Drs Williams, Kuyken, Speckens, Ma, and Segal
additionally receive payments for training
workshops and presentations related to
mindfulness-based cognitive therapy. Dr Kuyken
donates all such fees to the Oxford Mindfulness
Foundation, a charitable trust that supports the
work of the Oxford Mindfulness Centre, as does Dr
Speckens to the Radboud UMC. Dr Segal is a
member of the scientific advisory board for Mindful
Noggin, which is part of NogginLabs, a private
company specializing in customized web-based
learning. Dr Kuyken was an unpaid director of the
Mindfulness Network Community Interest
Company until 2015. Drs Byng, Kuyken,and
Williams gave evidence to the UK Mindfulness All
Party Parliamentary Group. No other disclosures
were reported.
Funding/Support: This work was supported by
Wellcome Trustgrants GR067797 and 104908/Z/
14/Z. Drs Kuyken, Taylor, Byford, and Byng were
partially supported by the National Institute for
Health Research Healthy Technology Assessment
program. Drs Taylorand Byng have also been
supported by the National Institute for Health
Research Collaboration for Leadership in Applied
Health Research and Care South West Peninsula at
the Royal Devon and Exeter National Health Service
Foundation Trust.Drs Schweizer and Dalgleish were
supported by the Medical Research Council.
Role of the Funder/Sponsor:The funders had no
role in the design and conduct of the study;
collection, management, analysis, and
interpretation of the data; preparation, review, or
approval of the manuscript; and decision to submit
the manuscript for publication.
Disclaimer: The views and opinions expressed in
this article are those of the authors and not
necessarily those of the National Health Service,
the National Institute for Health Research, or the
UK Department of Health.
Additional Contributions: We are grateful to the
trial teams of the constituent trials in this individual
patient data analysis and to Daniel Brett, MSc
(Department of Psychiatry, Universityof Oxford,
Prince of Wales International Centre, Warneford
Hospital, Oxford, England), for administrative
assistance. Mr Brett did not receive additional
compensation for his contributions.
REFERENCES
1. Collins PY, Patel V, JoestlSS, et al; Scientif ic
Advisory Board and the Executive Committee of
the Grand Challenges on Global Mental Health.
Grand challenges in global mental health. Nature.
2011;475(7354):27-30.
2. Layard R. The case for psychological treatment
centres. BMJ. 2006;332(7548):1030-1032.
3. Cuijpers P, van Straten A, Smit F,Mihalopoulos C,
Beekman A. Preventing the onset of depressive
disorders: a meta-analytic review of psychological
interventions. Am J Psychiatry. 2008;165(10):1272-
1280.
4. Segal Z V, Williams JMG, Teasdale JD.
Mindfulness-Based Cognitive Therapy for
Depression: A New Approach to Preventing Relapse.
New York, NY: Guilford Press; 2002.
5. Pie t J, Hougaard E. The effect of
mindfulness-based cognitive therapy for
prevention of relapse in recurrent major depressive
disorder: a systematic review and meta-analysis.
Clin Psychol Rev. 2011;31(6):1032-1040.
6. Ma SH, Teasdale JD. Mindfulness-based
cognitive therapy for depression: replication and
exploration of differential relapse prevention
effects. J Consult Clin Psychol. 2004;72(1):31-40.
7. Teasdale JD, Segal ZV, Williams JMG, Ridgeway
VA, Soulsby JM, Lau MA. Prevention of
relapse/recurrence in major depression by
mindfulness-based cognitive therapy. J Consult Clin
Psychol. 2000;68(4):615-623.
8. Riley RD, Lambert PC, Abo-Zaid G. Meta-analysis
of individual participant data: rationale, conduct,
and reporting. BMJ. 2010;340:c221.
9. Liberati A , Altman DG, Tetzlaff J, et al. The
PRISMA statement for reporting systematic reviews
and meta-analyses of studies that evaluate health
care interventions: explanation and elaboration.
Ann Intern Med. 2009;151(4):W65-W94.
10. Stewart LA , ClarkeM, Rovers M, et al;
PRISMA-IPD Development Group. Preferred
reporting items for systematic review and
meta-analyses of individual participant data: the
PRISMA-IPD statement. JAMA. 2015;313(16):1657-
1665.
11. Segal ZV, Williams JMG, Teasdale JD.
Mindfulness-Based Cognitive Therapy forD epression.
2nd ed. New York, NY: Guilford Press; 2013.
12. Fresco DM, Moore MT, van Dulmen MHM, et al.
Initial psychometric properties of the Experiences
Questionnaire: validation of a self-report measure
of decentering. Behav Ther. 2007;38(3):234-246.
13. Kuyken W, Byford S, Taylor RS,e t al.
Mindfulness-based cognitive therapy to prevent
relapse in recurrent depression. J Consult Clin Psychol.
2008;76(6):966-978.
14. Baer RA, Smith GT, Allen KB. Assessment of
mindfulness by self-report: the Kentucky Inventory
of Mindfulness Skills. Assessment. 2004;11(3):191-206.
15. Bondolf i G, Jermann F, der Linden MV, et al.
Depression relapse prophylaxis with
mindfulness-based cognitive therapy: replication
and extension in the Swiss health care system.
J Affect Disord. 2010;122(3):224-231.
16. Brown KW, Ryan RM. The benefits of being
present: mindfulness and its role in psychological
well-being. J Pers Soc Psychol. 2003;84(4):822-848.
17. Godfrin KA , vanHeeringen C. The effects of
mindfulness-based cognitive therapy on recurrence
of depressive episodes, mental health and quality
of life: a randomized controlled study. Behav Res Ther.
2010;48(8):738-746.
18. Segal Z V, Bieling P, YoungT, et al.
Antidepressant monotherapy vs sequential
pharmacotherapy and mindfulness-based cognitive
therapy,or placebo, for relapse prophylaxis in
recurrent depression. Arch Gen Psychiatry. 2010;67
(12):1256-1264.
19. Huijbers MJ, Spinhoven P, Spijker J, et al.
Adding mindfulness-based cognitive therapy to
maintenance antidepressant medication for
prevention of relapse/recurrence in major
depressive disorder: randomised controlled trial.
J Affect Disord. 2015;187:54-61.
20. VeehofMM, Ten Klooster PM, Taal E,
Westerhof GJ, Bohlmeijer ET. Psychometric
properties of the Dutch Five Facet Mindfulness
Questionnaire (FFMQ) in patients with
fibromyalgia. Clin Rheumatol. 2011;30(8):1045-1054.
21. Kuyken W, Hayes R, Barrett B, et al.
Effectiveness and cost-effectiveness of
mindfulness-based cognitive therapy compared
with maintenance antidepressant treatment in the
prevention of depressive relapse or recurrence
(PREVENT): a randomised controlled trial. Lancet.
2015;386(9988):63-73.
22. Baer RA , Smith GT, Hopkins J, Krietemeyer J,
Toney L. Using self-report assessment methods to
explore facets of mindfulness. Assessment. 2006;13
(1):27-45.
23. Williams JM, Crane C, Barnhofer T, et al.
Mindfulness-based cognitive therapy for
preventing relapse in recurrent depression:
a randomized dismantling trial. J Consult Clin Psychol.
2014;82(2):275-286.
24. Higgins JPT, Altman DG, Gotzsche P C, et al.
The Cochrane Collaboration's tool for assessing risk
of bias in randomised trials. BMJ. 2011;343:d5928.
25. First MB, Spitzer RL, Gibbon M, Williams JBW.
The Structured Clinical Interview for DSM-IV Axis I
Disorder With Psychotic Screen. New York,NY: New
York PsychiatricInstitute; 1995.
26. Borenstein M, Hedges LV, Higgins JPT,
Rothstein HR. Introduction to Meta-analysis.
Chichester, England: John Wiley & Sons, Ltd;2009.
27. Tudur C, Williamson PR, Khan S, Best LY. The
value of the aggregate data approach in
meta-analysis with time-to-event outcomes. JRoy
Stat Soc A Sta. 2001;164(2):357-370. doi:10.1111
/1467-985X.00207.
28. Williamson PR , Smith CT, Hutton JL, Marson
AG. Aggregate data meta-analysis with
time-to-event outcomes. Stat Med. 2002;21(22):
3337-3351.
29. Hig gins JPT, Thompson SG, Deeks JJ, Altman
DG. Measuring inconsistency in meta-analyses. BMJ.
2003;327(7414):557-560.
Efficacy of Mindfulness-Based Cognitive Therapy Original Investigation Research
jamapsychiatry.com (Reprinted) JAMA Psychiatry Published online April 27, 2016 E9
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
Copyright 2016 American Medical Association. All rights reserved.
30. Hig gins JPT, Thompson SG. Quantifying
heterogeneity in a meta-analysis. Stat Med. 2002;
21(11):1539-1558.
31. Riley RD,Hig gins JP, Deeks JJ. Interpretation of
random effects meta-analyses. BMJ. 2011;342:d549.
32. Smith CT, Williamson PR, Marson AG.
Investigating heterogeneity in an individual patient
data meta-analysis of time to event outcomes. Stat
Med. 2005;24(9):1307-1319.
33. Crow ther MJ, Look MP, Riley RD. Multilevel
mixed effects parametric survival models using
adaptive Gauss-Hermite quadrature with
application to recurrent events and individual
participant data meta-analysis. Stat Med. 2014;33
(22):3844-3858.
34. Royston P, Parmar MKB. Flexible parametric
proportional-hazards and proportional-odds
models for censored survival data, with application
to prognostic modelling and estimation of
treatment effects. Stat Med. 2002;21(15):2175-2197.
35. Kraemer HC, Wilson GT, Fairburn CG, AgrasW S.
Mediators and moderators of treatment effects in
randomized clinical trials. Arch Gen Psychiatry.
2002;59(10):877-883.
36. Meadows GN, Shawyer F, Enticott JC, et al.
Mindfulness-based cognitive therapy for recurrent
depression: a translational research study with
2-year follow-up.AustNZJPsychiatry. 2014;48(8):
743-755.
37. Driessen E, Hollon SD, Bockting CLH, Cuijpers P,
Turner EH. Does publication bias inflate the
apparent efficacy of psychological treatment for
major depressive disorder? a systematic review and
meta-analysis of US National Institutes of
Health-funded trials. PLoS One. 2015;10(9):
e0137864.
38. Bie sheuvel-Leliefeld KE, Kok GD, Bockting CL,
et al. Effectiveness of psychological interventions in
preventing recurrence of depressive disorder:
meta-analysis and meta-regression. J Affect Disord.
2015;174:400-410.
39. Huijbers MJ, SpijkerJ, Donders AR, et al.
Preventing relapse in recurrent depression using
mindfulness-based cognitive therapy,
antidepressant medication or the combination: trial
design and protocol of the MOMENT Study. BMC
Psychiatry. 2012;12:125.
Research Original Investigation Efficacy of Mindfulness-Based Cognitive Therapy
E10 JAMA Psychiatry Published online April 27, 2016 (Reprinted) jamapsychiatry.com
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: http://archpsyc.jamanetwork.com/ by a Radboud University Nijmegen User on 04/29/2016
