See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/277960365

Systematic review and meta-analysis of

transdiagnostic psychological treatments for
anxiety and...

Article in Clinical psychology review June 2015

DOI: 10.1016/j.cpr.2015.06.002

CITATIONS READS

31 192

5 authors, including:

Jill M Newby Anna Mckinnon

UNSW Australia Macquarie University
31 PUBLICATIONS 392 CITATIONS 20 PUBLICATIONS 137 CITATIONS

SEE PROFILE SEE PROFILE

Tim Dalgleish
Medical Research Council (UK)
226 PUBLICATIONS 10,795 CITATIONS

SEE PROFILE

All content following this page was uploaded by Jill M Newby on 29 June 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 Clinical Psychology Review 40 (2015) 91110

Contents lists available at ScienceDirect

Clinical Psychology Review

Systematic review and meta-analysis of transdiagnostic psychological

treatments for anxiety and depressive disorders in adulthood
Jill M. Newby a,, Anna McKinnon b,1, Willem Kuyken d, Simon Gilbody e, Tim Dalgleish b,c
a
Clinical Research Unit for Anxiety and Depression, School of Psychiatry, University of New South Wales at St Vincent's Hospital, 390 Victoria Street, Darlinghurst, NSW 2010, Australia
b
Medical Research Council Cognition and Brain Sciences Unit, 15 Chaucer Road, Cambridge CB2 7EF, United Kingdom
c
Cambridgeshire and Peterborough Mental Health Foundation Trust, Cambridge, United Kingdom
d
Department of Psychiatry, University of Oxford, Warneford Hospital, Warneford Lane, Oxford OX37JX, United Kingdom
e
Department of Health Sciences, The University of York, Seebohm Rowntree Building, Heslington, York YO105DD, United Kingdom

H I G H L I G H T S

Transdiagnostic (TD) treatments have large effects on anxiety and depression.

TD-CBT has been most widely evaluated, followed by mindfulness-based treatments.
Medium to large comparative group differences between TD treatments and control conditions
Type of treatment, delivery format, and type of control condition inuenced outcomes.
More comparisons are needed with TAU controls, and across treatment types.

a r t i c l e i n f o a b s t r a c t

Article history: A broad array of transdiagnostic psychological treatments for depressive and anxiety disorders have been evaluated,
Received 4 December 2014 but existing reviews of this literature are restricted to face-to-face cognitive behavioural therapy (CBT) protocols.
Received in revised form 2 June 2015 The current meta-analysis focused on studies evaluating clinician-guided internet/computerised or face-to-face
Accepted 4 June 2015
manualised transdiagnostic treatments, to examine their effects on anxiety, depression and quality of life (QOL).
Available online 6 June 2015
Results from 50 studies showed that transdiagnostic treatments are efcacious, with large overall mean uncon-
Keywords:
trolled effects (pre- to post-treatment) for anxiety and depression (gs = .85 and .91 respectively), and medium
Treatment outcome for QOL (g = .69). Uncontrolled effect sizes were stable at follow-up. Results from 24 RCTs that met inclusion criteria
Meta-analysis showed that transdiagnostic treatments outperformed control conditions on all outcome measures (controlled ESs:
Systematic review gs = .65, .80, and .46 for anxiety, depression and QOL respectively), with the smallest differences found compared to
Depression treatment-as-usual (TAU) control conditions. RCT quality was generally poor, and heterogeneity was high. Exami-
Anxiety nation of the high heterogeneity revealed that CBT protocols were more effective than mindfulness/acceptance pro-
Transdiagnostic tocols for anxiety (uncontrolled ESs: gs = .88 and .61 respectively), but not depression. Treatment delivery format
inuenced outcomes for anxiety (uncontrolled ESs: group: g = .70, individual: g = .97, computer/internet: g = .96)
and depression (uncontrolled ESs: group: g = .89, individual: g = .86, computer/internet: g = .96). Preliminary
evidence from 4 comparisons with disorder-specic treatments suggests that transdiagnostic treatments are as ef-
fective for reducing anxiety, and may be superior for reducing depression. These ndings show that transdiagnostic
psychological treatments are efcacious, but higher quality research studies are needed to explore the sources of
heterogeneity amongst treatment effects.
2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
1.1. Overview of existing transdiagnostic treatments for anxiety and depressive disorders . . . . . . . . . . . . . . . . . . . . . . . . 93
1.2. Transdiagnostic cognitive behaviour therapy (TD-CBT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Corresponding author at: Clinical Research Unit for Anxiety and Depression (CRUfAD), School of Psychiatry, University of New South Wales at St Vincent's Hospital, Level 4, The O'Brien
Centre St. Vincent's Hospital, 394-404 Victoria Street, Darlinghurst, NSW 2010, Sydney, Australia.
E-mail address: j.newby@unsw.edu.au (J.M. Newby).
1
Anna McKinnon is now at the Brain and Mind Research Institute, 94 Mallet Street, Camperdown, NSW 2050, Australia.

http://dx.doi.org/10.1016/j.cpr.2015.06.002
0272-7358/ 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
92 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110

1.3. Transdiagnostic mindfulness- and acceptance-based treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

1.4. Previous systematic reviews and meta-analyses of transdiagnostic psychological treatments and gaps in the existing literature . . . . . . . 94
2. Aim/objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.1. Protocol and registration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.2. Inclusion/eligibility criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.3. Excluded studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.4. Identication and selection of studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.5. Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.6. Data extraction and management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.7. Primary outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.8. Risk of bias in individual studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.9. Statistical analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.9.1. Calculation of effect sizes: changes on primary outcome measures between pre- and post-treatment, and pre-treatment and follow-up
96
3.9.2. Calculation of effect sizes: transdiagnostic treatments versus controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.10. Subgroup analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.11. Testing homogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.12. Risk of bias across studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.12.1. Testing for publication bias and dealing with publication bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.1. Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.2. Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.3. Pre-treatment diagnostic assessment and outcome measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.4. Risk of bias within randomised controlled trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.5. Synthesis of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.5.1. What are the effects of transdiagnostic psychological treatments on primary outcomes between pre- and post-treatment (uncontrolled effect
sizes)? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.6. Subgroup analyses of uncontrolled pre- to post-treatment effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.6.1. Is there a differential effect of transdiagnostic interventions on depression versus anxiety symptoms between pre- and post-treatment? 102
4.6.2. What is the impact of treatment delivery format on pre-to-post-treatment outcomes? . . . . . . . . . . . . . . . . . . 103
4.6.3. What is the impact of treatment type on pre-to-post-treatment outcomes? . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.7. Effects of transdiagnostic psychological treatments on primary outcomes between pre-treatment and follow-up (uncontrolled effect sizes) . . 103
4.8. Between-group effects of transdiagnostic psychological interventions versus control groups . . . . . . . . . . . . . . . . . . . . . . 104
4.8.1. Subgroup analyses: does the magnitude of effect depend on the type of control condition? . . . . . . . . . . . . . . . . . . . 104
4.9. Risk of bias across studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
4.10. Comparisons between transdiagnostic treatments versus disorder-specic treatments . . . . . . . . . . . . . . . . . . . . . . . . 105
5. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.1. Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Role of funding sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Conict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

1. Introduction
Depressive and anxiety disorders (collectively referred to as emotional
disorders) represent one of the largest causes of disability worldwide,
with estimated lifetime prevalence rates of up to 29% for anxiety disorders
and 19% for depressive disorders (Kessler et al., 2005). The economic bur-
den is enormous, with depression and anxiety accounting for a third to
one half of the global cost of mental illness, currently estimated at $2.5 tril-
lion (2010), and projected to increase to over $6 trillion by 2030 (WHO
Global Burden of Disease: 2004 update, WHO, 2008). Depressive and anx-
iety disorders typically develop in late adolescence/early adulthood, and
are chronic and relapsing if they remain untreated and even after acute
treatment (Judd, 1997). The burden of these disorders is difcult to over-
state: they impair the quality of life of sufferers and their loved ones, re-
duce workforce participation, contribute to marked occupational
impairment and lost productivity (Birnbaum et al., 2010), and increase
risk for the development and morbidity associated with chronic physical
conditions (e.g., cardiovascular disease) (Katon, 2011).
Until relatively recently, the disorder-specic approach has domi-
nated the way in which depressive and anxiety disorders have been
conceptualised and researched, and has shaped the way treatments
have been developed and evaluated. While this approach to treatment
has demonstrable efcacy (Hofmann & Smits, 2008), there are also sig-
nicant drawbacks. First, there is substantial discrepancy between the
disorder-specic treatment approach with our current understanding
of depressive and anxiety disorders. There is strong evidence indicating
that similar aetiological and maintenance processes underlie depressive
and anxious psychopathology. For instance, anxiety and depressive dis-
orders share many similar genetic, familial, and environmental risk fac-
tors (Kendler, 1996; Kessler et al., 2005), with structural modelling
studies showing that a broad internalising liability or a common dis-
tress/negative affectivity component underlies anxiety and depressive
disorders (Andrews et al., 2009; Watson, 2005). These disorders also
share similar cognitive-affective, interpersonal, and behavioural main-
taining factors, with the similarities superseding any minor differences
between disorders (Harvey, Watkins, Mansell, & Shafran, 2004).
Second, disorder-specic interventions pay relatively limited attention
to comorbidity, despite evidence of high comorbidity rates up to 40
80% in both clinical and epidemiological studies (Brown, Campbell,
Lehman, Grisham, & Mancill, 2001; Kessler et al., 2005). High comorbid-
ity poses a signicant problem for both conceptualisation and treatment
decisions: how should the treating practitioner provide optimal
 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Abbreviations
ACT acceptance and commitment therapy
AM + VR anxiety management and virtual reality exposure
therapy
BA behavioural activation
CBT cognitive behaviour therapy
CCBT computerised cognitive behaviour therapy
GCBT group cognitive behaviour therapy
CG-ERP clinician guided exposure and response prevention
EBT evidence-based treatment
E-COM enhanced community treatment
F-SET false safety behaviour elimination therapy
GMBCT group mindfulness based cognitive therapy
GMBSR group mindfulness based stress reduction
GMBSM group mindfulness based stress management
iCBT internet-delivered cognitive behavioural therapy
iCBT (CO) internet cognitive behavioural therapy with coach
guidance
iCBT (CL) internet cognitive behavioural therapy with clinician
guidance
i-AFPP internet affect focused psychodynamic psychotherapy
IPT interpersonal psychotherapy
QOL quality of life
RT relaxation training
SC supportive counselling
SG-ERP self-guided exposure and response prevention
STPP short-term psychodynamic psychotherapy
T-CBT telephone-delivered cognitive behaviour therapy
TD transdiagnostic
TAU treatment-as-usual
WLC waiting list control
treatment to a patient presenting with multiple disorders, when the
majority of evidence-based treatments (EBTs) are tailored to specic
diagnoses?
Third, despite the clinical utility of diagnostic categories, there are
also some limitations to the reliability and validity of the diagnostic
classications that disorder-specic treatments have been based on:
there is considerable heterogeneity within diagnostic categories, poor
discrimination between supposedly distinct emotional disorders, and
high rates of not otherwise specied diagnoses (e.g., see Brown, Di
Nardo, Lehman, & Campbell, 2001). In addition, the ever-growing num-
bers of treatment manuals for different disorders (and multiple manuals
for the same disorder) represent a signicant barrier to implementation,
dissemination and training, and EBT manuals often share many com-
mon elements (e.g., cognitive restructuring) leading to signicant re-
dundancy across treatment protocols (Chorpita & Daleiden, 2009).
Driven by these concerns, there has been growing consensus amongst
international experts in the eld that a novel approach is needed in the
way we classify, formulate, treat, and prevent depression and anxiety
disorders (Barlow, Allen, & Choate, 2004). The move away from the
single-diagnosis approach towards a transdiagnostic conceptualisation
and treatment of depressive and anxiety disorders represents a signicant
paradigm shift (Craske, 2012), mirroring a similar shift in EBTs for eating
disorders (Fairburn, Cooper, & Shafran, 2003). The transdiagnostic
approach focuses on identifying the common and core maladaptive tem-
peramental, psychological, cognitive, emotional, interpersonal and behav-
ioural processes that underpin a broad array of diagnostic presentations
(Harvey et al., 2004) and targeting these factors in treatment (Barlow
et al., 2004). Transdiagnostic or unied treatments apply the same un-
derlying treatment principals across mental disorders, without tailoring
the protocol to specic diagnoses (p21, McEvoy, Nathan, & Norton,
2009), and as such operate outside the traditional diagnostic boundaries
93
of DSM (American Psychiatric Association, 2013) or ICD (World Health
Organization, 1992). The transdiagnostic approach is also compatible
with the Research Domain Criteria (RDoC), put forth by the U.S. National
Institute for Mental Health as an alternative to DSM or ICD, that focuses on
the underlying mechanisms (e.g., cognition, negative affect, arousal) that
cut across multiple disorders (see Cuthbert, 2014).
Transdiagnostic psychological treatments have been hailed as a prom-
ising new approach to overcome some of the pitfalls of disorder-specic
treatments (see Clark & Taylor, 2009; Craske, 2012 for a discussion of
the promise and pitfalls of the transdiagnostic approach). In theory,
transdiagnostic treatments should enable the treating practitioner to con-
ceptualise the common maintaining processes across presenting issues,
and deliver evidence-based treatment strategies within the one protocol,
increasing the efciency and efcacy of treatment, reducing the need for
multiple manuals, and increasing the ease of implementation (Chorpita,
Taylor, Francis, Moftt, & Austin, 2004). At present however, it is unclear
whether transdiagnostic treatments meet these high expectations. It re-
mains uncertain how efcacious transdiagnostic treatments are in reduc-
ing anxiety and depression symptoms, and in improving quality of life. In
addition, although one of the supposed strengths of transdiagnostic treat-
ment approach is that it reduces the need for multiple treatment proto-
cols, an increasing number of transdiagnostic treatments are being
developed and evaluated around the world. These protocols differ with
respect to the diagnostic combinations and symptom proles they are
designed to target (e.g., multiple anxiety disorders or anxiety and depres-
sion), the type of treatment techniques they use (e.g., cognitive behav-
ioural therapy [CBT] versus mindfulness-based treatments), and their
delivery format (e.g., group-based versus internet-delivered treatments),
but it is unknown whether these differences affect outcomes. In addition,
many of the existing evaluations have been conducted with small sam-
ples, and therefore may have produced biased estimates of effect sizes,
and lacked power to detect potentially important differences be-
tween transdiagnostic treatments and control conditions. To ad-
dress these uncertainties, we conducted a systematic review and
meta-analysis to synthesise the evidence from studies evaluating
transdiagnostic psychological treatments for adult depressive and
anxiety disorders.
1.1. Overview of existing transdiagnostic treatments for anxiety and
depressive disorders
The transdiagnostic treatments that have been evaluated to-date for
anxiety and depressive disorders tend to fall into two broad approaches.
The rst approach, which has been most researched, is broad-spectrum
transdiagnostic CBT (TD-CBT) interventions which target either hetero-
geneous anxiety disorders (e.g., Norton, 2008), or anxiety and depres-
sion (e.g., McEvoy & Nathan, 2007). TD-CBT protocols are informed by
cognitive and behavioural models of emotional disorders (Barlow,
2000; Beck, 1979; Norton, 2006) and apply a core set of generic CBT-
based treatment principles and techniques (e.g., graded exposure and
cognitive restructuring) to target the common processes underlying
anxiety and depression, rather than targeting the symptoms of specic
disorders. In contrast to this traditional CBT-based transdiagnostic
approach, there has also been increasing interest in third wave or
new wave behavioural and cognitive behavioural therapies such as
acceptance- and mindfulness-based interventions for the transdiagnostic
treatment of anxiety and depressive disorders (for a review of terminolo-
gy see Hofmann, Sawyer, Witt, & Oh, 2010). Unlike traditional CBT, which
aims to modify dysfunctional cognitions, behaviours and emotions,
acceptance- and mindfulness-based interventions aim to change a
person's perspective on, and relationship with their cognitions and emo-
tions. This process is facilitated through mindfulness, non-judgemental
awareness, and acceptance of psychological experiences.
94 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
1.2. Transdiagnostic cognitive behaviour therapy (TD-CBT)
TD-CBT interventions have now been evaluated in group (anxiety
and depression: McEvoy & Nathan, 2007; anxiety: Norton, 2008), indi-
vidual (Arch et al., 2012), computerised (anxiety and depression:
Proudfoot et al., 2004) and online delivery formats (mixed anxiety
disorders: Johnston, Titov, Andrews, Spence, & Dear, 2011; anxiety
and depression: Titov et al., 2011). While it is still unclear which deliv-
ery format leads to superior outcomes, these studies have found moder-
ate to large uncontrolled effect sizes (ESs) for reductions in anxiety
symptoms (e.g., d = 1.68), depression and functional impairment
(e.g., Norton, 2008). In randomised controlled trials (RCTs) TD-CBT
has been shown to be more effective compared to waitlist control
(WLC) in group-format (controlled between-group ES: d = .50)
(Erickson, Janeck, & Tallman, 2007) and online delivery (Johnston
et al., 2011). However, there have been few RCTs comparing TD-
CBT with attention control conditions or other psychological treatments,
and those conducted give a more modest account of the impact of TD-
CBT. For example, although Norton (2012) found lower drop-out rates
during group TD-CBT, there were no signicant differences in anxiety
severity at post-treatment compared to relaxation training.
In addition, there is also discrepancy in the literature of existing RCTs
regarding how effective TD-CBT treatments are compared to disorder-
specic CBT treatments. For example, in one RCT, Craske et al. (2007)
showed that a disorder-specic programme of group plus individual
CBT sessions that was targeted solely at a patient's primary panic disorder
was more effective than a CBT protocol that targeted both the primary
panic disorder and comorbid disorders (e.g., depression). In contrast, in
another RCT, Norton and Barrera (2012) found no signicant differences
between group-based TD-CBT and disorder-specic group CBT for anxiety
disorders. The latter study was likely to be underpowered to detect statis-
tically signicant differences between the two treatment approaches. Our
aim was to use meta-analytic techniques to combine the results across
multiple RCTs comparing transdiagnostic to disorder-specic treatments.
By combining the results of individual trials, meta-analysis has the poten-
tial to increase the power to detect differences across these treatment
approaches.
In addition to broad-spectrum TD-CBT, other variations to TD-CBT
have been developed and evaluated. For example, Barlow and col-
leagues' Unied Protocol for Transdiagnostic Treatment of Emotional Dis-
orders (UP) (Ellard, Fairholme, Boisseau, Farchione, & Barlow, 2010) is
a transdiagnostic emotion-focused CBT-based intervention which aims
to improve emotional awareness and emotion regulation skills, facili-
tate cognitive and emotional exibility, and reduce avoidance and
maladaptive emotion-driven behaviours. The UP has been demonstrated
to lead to robust effects on anxiety and depression symptoms in two un-
controlled trials (Ellard et al., 2010; Wilamowska et al., 2010), and is more
effective than a wait list control (WLC) in reducing anxiety (controlled
between-group ESs: 0.56 on the BAI, and 1.39 on the clinician severity rat-
ings of (co) principal diagnosis), and depression (controlled between-
group ESs: 1.11 on the BDI-II) (Farchione et al., 2012). Another variation
of TD-CBT is group-based false safety behaviour elimination therapy
(F-SET), which focuses solely on the reduction of transdiagnostic safety
seeking behaviours and safety signals. F-SET has been shown to have
large effects in the reduction of both self-reported and clinician-rated
anxiety severity (uncontrolled pre- to post-treatment ESs: d = 0.81.1),
moderate effects for depression (uncontrolled ES: d = .73), and medium
to large between group effect sizes compared to WLC on anxiety out-
comes at post-treatment (controlled between-group ES: d = .77), with
gains maintained up to 6 months (Schmidt et al., 2012).
1.3. Transdiagnostic mindfulness- and acceptance-based treatments
While there have been fewer evaluations of transdiagnostic
mindfulness- and acceptance-based treatments, results suggest that
they provide a viable and effective treatment option for emotional
disorders. For example, Kabat-Zinn and colleagues (Kabat-Zinn et al.,
1992; Miller, Fletcher, & Kabat-Zinn, 1995), and more recently, Arch and
colleagues (Arch et al., 2013) have evaluated the impact of mindfulness-
based stress reduction (MBSR) with mixed samples of participants with
emotional disorders. MBSR is a group-based treatment that combines
psychoeducation with yoga and intensive guided mindfulness-based
meditation practices (e.g., body scan, formal sitting meditation). Kabat-
Zinn et al. (1992) demonstrated large uncontrolled ESs (0.88 on the
BAI) for reduction in anxiety symptoms following MBSR, that were main-
tained three years post-treatment (Miller et al., 1995). More recently,
Arch et al. (2013) compared MBSR with TD-CBT in an RCT, and found
that both MBSR and TD-CBT led to large reductions in the severity of pri-
mary anxiety diagnoses, worry, anxious arousal and depression symp-
toms. Although there were no signicant overall differences between
MBSR and TD-CBT, there were small effect sizes in favour of MBSR for
the severity of primary diagnoses and worry frequency at follow-up (con-
trolled between-group ES: d = .29) and in favour of TD-CBT (controlled
between-group ES: d = .31) for reducing anxious arousal. These results
indicate that MBSR provides an efcacious treatment option for people
with multiple emotional disorders (Arch et al., 2013), and may have com-
parable effects to TD-CBT.
In another recent RCT, Arch et al. (Arch et al., 2012) compared the ef-
cacy of individual TD-CBT to individual acceptance and commitment
therapy (ACT). ACT (Hayes, Strosahl, & Wilson, 1999) is a psychological
therapy that combines psychoeducation and experiential exercises that
encourage mental exibility, cognitive defusion, mindfulness and ac-
ceptance of experiences and the reduction of experiential avoidance.
ACT interventions also aim to reduce experiential avoidance by promot-
ing active engagement in activities and behaviours that are done in pur-
suit of ones goals and personal values. The authors did not nd any
signicant differences between individual TD-CBT compared to ACT at
post-treatment, likely due to lack of power. However, there was a
small to medium effect in favour of TD-CBT for improvements in quality
of life (d = .42), and a large effect for lower primary disorder severity
ratings at 12 months post-treatment for the ACT group (d = 1.10, for
the completer sample only) that did not reach signicance. Unfortu-
nately, there were too few participants to detect whether CBT conferred
signicant benet over ACT or vice versa. Our aim was to combine the
results across individual trials using meta-analysis to shed light on
whether the type of transdiagnostic treatment inuences outcomes.
1.4. Previous systematic reviews and meta-analyses of transdiagnostic
psychological treatments and gaps in the existing literature
To date, there have been no previous attempts to systematically re-
view the full set of transdiagnostic psychological treatments for anxiety
and depressive disorders, beyond CBT delivered face-to-face. There has
been one narrative review of transdiagnostic treatments for depression
and anxiety disorders (McEvoy et al., 2009), and two meta-analyses of
transdiagnostic interventions, both restricting their focus to face-to-
face CBT protocols for anxiety disorders (Norton & Philipp, 2008;
Reinholt & Krogh, 2014). The rst meta-analytic review published by
Norton and Philipp (2008) found large mean uncontrolled effect sizes
for pre- to post-treatment reductions in anxiety (d = 1.29, 95%CI:
0.661.93). However, they only examined the uncontrolled effect sizes
because of the lack of available evaluations that included comparison
control conditions at the time of review. In a more recent evaluation,
Reinholt and Krogh (2014) carried out a meta-analysis of 11 RCTs
of TD-CBT protocols for anxiety disorders that had been published
prior to June 2013, and found that TD-CBT outperformed WLC and
treatment-as-usual (TAU) control conditions, with a medium overall
difference between groups in anxiety severity at post-treatment
(d = .68, 95%CI: .45.90). The results suggested that RCTs using WLC
conditions had larger effect sizes (controlled ES: d = 1.00) compared
to those that used TAU control conditions (controlled ES: d = .28).
 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Together these meta-analyses provide promising evidence for the
efcacy of TD-CBT protocols in reducing anxiety severity, but there are
questions that await further investigation. Because Reinholt and Krogh
(2014) included anxiety severity as their sole outcome measure, it re-
mains unclear whether transdiagnostic treatments have differential ef-
fects on anxiety compared to depression, and what impact they have on
quality of life. Measurement of quality of life is particularly important be-
cause quality of life is a standard measure used not only for the calculation
of QALYs (quality-adjusted-life-years), but in health economic analyses
(e.g., cost utility analyses). The degree to which gains are maintained
following transdiagnostic treatments also awaits further evaluation. In
addition, both of the meta-analyses restricted their analysis to CBT inter-
ventions that were delivered face-to-face. A more inclusive systematic re-
view and meta-analysis is now needed that comprehensively reviews:
(i) the full set of transdiagnostic treatments for emotional disorders, in-
cluding CBT as well as third wave ACT and mindfulness-based interven-
tions, and (ii) treatments that are delivered across different formats,
including face-to-face group, individual, and computerised treatments.
Understanding whether the type of transdiagnostic treatment and the de-
livery format have an impact on outcomes has the potential to inform
both clinical practice and future treatment developments in this area. In
addition, Reinholt and Krogh (2014) only explored how transdiagnostic
treatments compare to WLC and TAU control conditions. Further research
is therefore needed to explore how transdiagnostic treatments perform
on average relative to no-treatment/WLC controls versus TAU, as well as
against other attention control conditions (such as relaxation training or
psychological placebo), and disorder-specic treatments.
2. Aim/objectives
The aim of this review was to provide a comprehensive review of the
published studies evaluating transdiagnostic psychological treatments
for adults with depression and/or anxiety disorders. In conducting this
review, we aimed to answer three main questions: (1) what are the
overall effects of transdiagnostic treatments on depression, anxiety
and quality of life?; (2) what is the relative efcacy of transdiagnostic
treatments versus various control conditions (WLC, TAU, attention con-
trol and disorder-specic treatments)?; and (3) what is the impact of
potential moderators of treatment effect including type of treatment
and delivery format?. The nal aim of this review was to provide recom-
mendations for future research and treatment development.
3. Methods
3.1. Protocol and registration
We followed the PRISMA guidelines for reporting of this systematic
review (Moher, Liberati, Tetzlaff, Altman, & The, 2009). This review pro-
tocol was developed following the procedures outlined in the Cochrane
Handbook for systematic reviews (Higgins & Green, 2011), and the pro-
tocol was registered with PROSPERO2 [CRD42014010469].
3.2. Inclusion/eligibility criteria
(i) Types of participants: We included studies in which participants
were adults (18 years and older), who met DSM or ICD criteria
for a primary diagnosis of an anxiety or depressive disorder or
mixed anxiety and depression, established by a formal validated
diagnostic interview. Because the search protocol was initially de-
veloped with DSM-IV-TR diagnostic categories in mind (American
Psychiatric Association, 2000), we included posttraumatic stress
disorder (PTSD) and acute stress disorder (ASD) as anxiety
2
http://www.crd.york.ac.uk/PROSPERO/index.asp.
95
disorders, although they are now classied as trauma and
stressor-related disorders in DSM-5 (American Psychiatric Associ-
ation, 2013). Depressive disorders included major depressive dis-
order (MDD), dysthymic disorder, and minor depression. For the
purposes of determining whether studies met our inclusion
criteria, depressive disorders were classied as one disorder cate-
gory (e.g., dysthymic disorder and major depression were grouped
together under the category of depressive disorder/depression).
(ii) Types of interventions: We included studies of manualised psycho-
logical treatments for adults that lasted for 2 or more sessions and
targeted: (i) two or more anxiety disorders; (ii) at least one
anxiety disorder together with a depressive disorder/depression.
Face-to-face clinician-delivered treatments were included.
Internet- or computer-delivered treatments were included
only if they were clinician-guided. Clinician-guided internet- or
computer-delivered treatments were dened as treatments that
involved having some form of email and/or telephone contact
with a therapist, clinician or support person during the treatment
or intervention period; that is, the clinician contact was not just re-
stricted to pre-treatment assessment interviews.
(iii) Types of comparisons and outcomes: Studies were included that
reported at least one validated self-report measure of anxiety or
depression at both baseline and post-treatment, enabling us to cal-
culate at minimum, the average uncontrolled effect size on prima-
ry outcome measures. Given we were interested in examining the
maintenance of gains following treatment; we were also in-
terested in examining outcomes at short-term follow-up (de-
ned as up to 6 months post-treatment). We also included
randomised trials in which the effects of transdiagnostic
treatment were compared with either: (a) a no-treatment
(WLC) condition; (b) a care-as-usual, or treatment-as-usual
control condition (TAU); (c) another control psychological
treatment (e.g., relaxation) or attention control condition;
or (d) a disorder-specic psychological treatment. This enabled
us to calculate between-groups effect sizes based on comparisons
between conditions at post-treatment. Diagnostic status was not
used as an outcome variable because there were too few studies
that reported this variable at post-treatment or follow-up.
(iv) Types of study design: We included studies published in English in a
peer-reviewed journal up to February 28, 2014. We included all
uncontrolled (open) trials which involved a prepost-study de-
sign, or randomised controlled trials comparing transdiagnostic in-
terventions to a control condition. RCTs with multiple control
conditions were also included as long as all of the other inclusion
criteria were met. One of the purposes of this study was to conduct
a comprehensive overview and examination of the available liter-
ature. Therefore, we chose to include uncontrolled trials as well as
RCTs, rather than restricting our analysis to the available RCTs.
3.3. Excluded studies
We excluded studies in which the psychological treatment was not
manualised, studies in which there was insufcient data reported to cal-
culate effect sizes and where we could not obtain those data, and studies
that focused on populations under age 18. Case studies and case series
were also excluded. Finally, studies with mixed samples with psychotic
disorders, personality disorders, and substance use disorders were ex-
cluded. Treatment protocols that involved combinations of pharmacolog-
ical and psychological interventions were excluded. Entirely self-guided
computerised or internet-delivered treatments were also excluded.
96 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
3.4. Identication and selection of studies
To identify studies for possible inclusion, we conducted comprehen-
sive systematic searches of the electronic databases PSYCInfo, and
PubMed up to 28 February 2014 (see Appendix A for our electronic
search strategy for PSYCInfo). To broaden the search criteria and maxi-
mise the sensitivity of our search, we combined terms indicative of anx-
iety, depression, mixed anxiety and depression, anxious depression, as
well as various anxiety and depressive disorders with the terms
transdiagnostic and a large range of psychotherapies such as vari-
ous forms of CBT, stress management, mindfulness-based therapies
(e.g., MBSR), ACT, attention training, metacognitive therapy (MCT)
and psychotherapy. We searched for a range of study types includ-
ing uncontrolled open trials, effectiveness studies and randomised
trials. We also searched the references lists of relevant manuscripts
and previous reviews of this literature, to identify additional studies
that met our inclusion criteria.
3.5. Study selection
The rst author (JN) initially screened all of the titles and abstracts
for all studies to determine their relevance to this study. Studies that
could be immediately excluded on the basis of the title and abstract
were discarded. For the remaining references, full text manuscripts
were reviewed for more comprehensive evaluation of the study inclu-
sion criteria. Both JN and AM independently read each full text to assess
eligibility for inclusion, and disagreements were resolved through dis-
cussion, and consultation with TD.
3.6. Data extraction and management
Data regarding methodology and outcome measures were extracted
into a Microsoft Excel spread sheet by JN (which was independently
checked by a research assistant), before being transferred to Com-
prehensive Meta-Analysis (version 2.0; Biostat, Inc.) for the meta-
analysis. The following information from each study was extracted: au-
thors, year of publication, setting, diagnostic assessment measure, mean
age, gender (proportion female), type of intervention (e.g., CBT, ACT),
delivery format (face-to-face individual, internet/computer, and face-
to-face group), duration of intervention, and attrition rates. Outcome
data for the primary and secondary outcome measures assessing anxi-
ety, depression, and quality of life were also extracted. For multiple re-
ports of the same study, we combined the outcome measures and
considered them as one study.
3.7. Primary outcomes
1. Symptoms of anxiety, according to continuous symptom measures
such as the Beck Anxiety Inventory (BAI) (Beck & Steer, 1996), Gen-
eralised Anxiety Disorder 7-item scale (GAD-7) (Spitzer, Kroenke,
Williams, & Lowe, 2006), or the Anxiety subscale of the Depression
Anxiety Stress Scale (DASS-21) anxiety subscale (Lovibond &
Lovibond, 1995).
2. Symptoms of depression, according to continuous symptom measures
such as the Beck Depression Inventory (BDI, Beck, Ward, Mendelson,
Mock, & Erbaugh, 1961), Beck Depression Inventory Second edition
(BDI-II, Beck, Steer, & Brown, 1996), the nine-item Patient Health
Questionnaire (PHQ-9: Kroenke, Spitzer, & Williams, 2001) or the
Depression subscale of the DASS-21 (Lovibond & Lovibond, 1995).
3. Quality of life according to general measures of quality of life or func-
tional impairment including the Euro-Qol (Euroqol Group, 1990) or
the SF-12 (Ware, Kosinski, & Keller, 1996).
We extracted and analysed outcome data on self-reported symp-
toms of depression, anxiety and quality of life at post-treatment and
follow-up. There was not enough information reported in the studies
to analyse diagnostic status following treatment. Because many of the
studies used more than one instrument to measure anxiety (or depres-
sion) we used the primary outcome as reported by the study investiga-
tors, or if absent, we used the most frequently used measures across
studies (BAI, GAD-7, and BDI, PHQ-9, DASS-21 depression subscale).
All included studies reported means and standard deviations at post-
intervention, allowing us to calculate the effect size directly, but there
were only 24 studies that reported at least one outcome measure at 3-
to 6-months follow-up.
3.8. Risk of bias in individual studies
We assessed the quality and risk of bias of included RCTs using the
Risk of Bias tool, developed by the Cochrane Collaboration (Higgins &
Green, 2011). This tool allowed us to assess possible sources of risk in
RCTs, including: (1) allocation sequence (the method used to generate
the allocation sequence is provided in sufcient detail to allow an as-
sessment of whether it should produce comparable groups), (2) alloca-
tion concealment (the method used to conceal the allocation sequence
is given in sufcient detail to determine whether intervention alloca-
tions could have been foreseen in advance of, or during, enrolment),
(3) blinding of participants, personnel and outcome assessors (all
measures used to blind outcome assessors from knowledge of which in-
tervention a participant received), (4) incomplete outcome data (assess-
ment of the completeness of outcome data from each main outcome
including attrition and exclusions, and whether all randomised partici-
pants were included in the analyses). The assessment of study quality
was conducted by two independent reviewers (JN and AM) and disagree-
ments were resolved via discussion. See Table 1 for quality ratings of in-
cluded controlled trials.
3.9. Statistical analyses
3.9.1. Calculation of effect sizes: changes on primary outcome measures
between pre- and post-treatment, and pre-treatment and follow-up
To examine the within-group effect (uncontrolled effect size) of
transdiagnostic psychological treatments, for each treatment we calcu-
lated the effect size referring to the difference between baseline and
post-treatment (or between baseline and follow-up), divided by pooled
standard deviation on each primary outcome measure, and the 95% con-
dence intervals around the effect sizes. Effect sizes were also adjusted
to address small sample sizes, according to the procedures outlined in
Hedges and Olin (1985, or Hedges g). Because the baseline and post-
treatment values are not independent from each other, the correlation
between time points was required, but most studies did not include
the correlation within the body of the manuscript. In the absence of
the correlation between time-points, we used a conservative value of
0.50 (Balk, Earley, Patel, Trikalinos, & Dahabreh, 2012).3
3.9.2. Calculation of effect sizes: transdiagnostic treatments versus controls
For each comparison between a transdiagnostic psychological treat-
ment and a control condition, we calculated the effect size (Hedges g)
referring to the difference between the two groups at post-treatment
(standardised mean difference) and the 95% condence intervals
around the effect sizes. Effect sizes were calculated by subtracting the
average score of the transdiagnostic treatment at post-treatment from
the average score for the control group, and dividing the result by the
pooled standard deviation of the two groups. Effect sizes of 0.2, 0.5
and 0.8 refer to small, moderate and large effect sizes respectively
(Cohen, 1988). Effect sizes were also adjusted to address small sample
3
We conducted a sensitivity analysis by recalculating the estimated effect using corre-
lation values of 0.25 and 0.75 to explore whether changing this imputed value affected the
overall effect size estimate. There were no substantial differences in the effect size esti-
mates when we used within-group correlations of 0.25 or 0.75 instead of 0.5, with the
largest difference between estimates being .02.
 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
sizes, according to the procedures outlined in Hedges and Olin (1985, or
Hedges g).
To calculate pooled mean effect sizes, we used the programme Com-
prehensive Meta-Analysis (version 2.0, Biostat Inc.). Given that Reinholt
and Krogh (2014) found evidence of heterogeneity, we similarly ex-
pected substantial heterogeneity amongst the interventions, and there-
fore calculated the mean effect sizes using a random effects model. The
random effects model assumes that the true effect size varies from one
study to the next, and that the studies in our analysis represent a ran-
dom sample of effect sizes that could have been observed. The summary
effect is our estimate of the mean of these effects (Borenstein, Hedges,
Higgins, & Rothstein, 2009).
3.10. Subgroup analyses
To assess the differences between subgroups (treatment type: CBT
vs. mindfulness/acceptance-based treatments; treatment format: indi-
vidual, group and computer/internet interventions), and the differences
in outcomes on anxiety versus depression measures, we conducted sub-
group analyses using the mixed effect model approach with the dataset
that included all studies with pre- to post-treatment data. In this model,
a random-effects model is used to pool studies within subgroups
(e.g., individual treatment), and we test for signicant differences be-
tween subgroups using a xed-effects model.
3.11. Testing homogeneity
To test the homogeneity of effect sizes, we calculated the I2 statistic,
which is an indicator of heterogeneity across effect sizes, and is provid-
ed as a score in percentages. A value of 0% indicates no heterogeneity,
whereas scores of 25%, 50% and 75% indicate low, moderate, and high
heterogeneity, respectively.
3.12. Risk of bias across studies
3.12.1. Testing for publication bias and dealing with publication bias
To test for publication bias, we inspected the funnel plot on the
primary outcome measures (for depression, anxiety and quality of life
measures respectively) (Egger, Davey Smith, Schneider, & Minder,
1997). In addition, we also conducted Duval and Tweedie's Trim and
Fill procedure (Duval & Tweedie, 2000a, 2000b) within Comprehensive
Meta-Analysis, which yields an adjusted effect size that takes into
account the publication bias observed within the funnel plot. This pro-
cedure corrects for the variance of the effects and provides a best esti-
mate of the unbiased effect size.
4. Results
4.1. Study selection
Fig. 1 presents the owchart describing the inclusion of studies. A
total of 10958 titles and abstracts were examined. 10589 were rejected
at title and abstract, and a further 323 were rejected after the entire ar-
ticle was reviewed. This left a total of 47 studies (from 46 articles) that
were analysed. Of these, 31 were RCTs, 15 were uncontrolled trials, and
one was a non-randomised trial.
4.2. Study characteristics
Characteristics of the included studies are found in Table 1. Because
some studies reported evaluations of either separate groups undergoing
the same treatment (e.g., results for immediate and delayed treatment
undergoing internet-delivered CBT [iCBT], Titov, Andrews, Johnston,
Robinson, & Spence, 2010), or different transdiagnostic treatments
(e.g., results for ACT and CBT, see Arch et al., 2012), we were able to in-
clude 50 studies in which 1865 patients participated to calculate the
97
average uncontrolled effects of transdiagnostic treatments between
pre- and post-treatment. The majority of these examined TD-CBT or
variants of TD-CBT such as behavioural activation combined with expo-
sure therapy, and anxiety management with virtual reality exposure
(n = 40). Seven studied mindfulness/acceptance-based interventions
(of which one was ACT) and three evaluated other forms of psychother-
apy (e.g., short term psychodynamic psychotherapy). Five studies eval-
uated TD-CBT in older adult samples only (N55 years, mean age =
69 years), and the remainder included working age adults over
18 years of age (mean age = 39 years). The samples were predominant-
ly comprised of females (mean proportion of females = 63%). With re-
gard to treatment format, 17 evaluated computer/internet, 18 face-to-
face group, and 17 face-to-face individual treatment format. In most
studies of face-to-face therapies, the treatments had 12 or fewer treat-
ment sessions (range 640 sessions). Rates of attrition varied across
studies from 0% to 56% (see Table 1).
Of the 31 RCTs, there were 27 studies (reporting 29 comparisons) in-
cluded in the meta-analysis comparing transdiagnostic treatments
(n = 1109) to control conditions (n = 992). The studies ranged in sam-
ple size from 10 to 121 per condition. For the control conditions, 12 used
WLC condition, 4 used treatment-as-usual (TAU) or (enhanced) usual
care control conditions, two used discussion forums, one online clinical
support, two used relaxation training, one supportive counselling, three
used psychoeducation as a control condition, and four used disorder-
specic treatments. In addition, there were two studies (with 233
participants altogether) which compared two alternative types of
transdiagnostic interventions (TD-CBT versus ACT, group MBSR versus
group TD-CBT). There was also one randomised comparison between
computerised versus face-to-face TD-CBT, and another study that used
antidepressant medications and pill placebo as their control condition.4
4.3. Pre-treatment diagnostic assessment and outcome measures
The majority of the studies administered the Structured Clinical
Interview for DSM-IV Axis I Disorders (SCID-I) (First, Spitzer, Gibbons,
& Williams, 1996), the Anxiety Disorders Interview Schedule (ADIS-
IV) (Brown, Dinardo, & Barlow, 1994) or the Mini International Neuro-
psychiatric Interview version 5.0.0 (MINI) (Sheehan et al., 1998) to ob-
tain a diagnostic assessment at pre-treatment. A wide range of outcome
measures were used to assess self-reported anxiety and depressive
symptoms and quality of life, with the most common the BAI, the
BDI (or BDI-II), and the QOLI. Only 39 out of 50 studies that reported
pre- to post-treatment effects incorporated quality of life assess-
ment measures.
4.4. Risk of bias within randomised controlled trials
The methodological quality of the studies reporting RCTs varied
widely (see Table 1). Twenty one (68%) reported adequate generation
of random sequencing, 14 (45%) reported adequately concealing
group allocation, and 21 (68%) reported appropriate blinding of out-
come assessments. Twenty four (78%) studies were coded as at low
risk of bias for incomplete outcome data, and 30 studies (97%) were
found to have low risk of bias for selective reporting of outcomes. Over-
all, only seven studies were classied by reviewers as at low risk of bias
on all ve measures of risk of bias. Thirteen studies were low risk on
four, 5 on three, 4 on two, and 2 were low risk of bias on either none
or only one measure.
4
These studies were not included in the calculation of effect sizes comparing
transdiagnostic treatments versus control groups, as they did not fall within the control
group categories that were the focus of this review.
 98
Table 1
Characteristics of studies included in the meta-analysis evaluating transdiagnostic treatments for anxiety and/or depression.

Study Country Participants Sample Design Intervention Clinician Diagnostic Measures Attrition Qual
measure

Anderson USA DSM-IV Panic/Ag or Community, mean Open trial: AM + VRE: 12 AM + 4 VR Psychologist SCID-IV PRCS, SSPS-pos, None n/a
et al. SocPhob, with public age not reported, 80% AM + VRE: 10 (individual) SSPS-neg, PRCA
(2005) speaking main fear female
Andersson Sweden DSM-IV anxiety Community, mean Open trial: iCBT: 27 iCBT: 10 modules over 10 weeks Clinical psychology SCID-IV CORE-OM, MADRS-S, 65% completed n/a
et al. disorder age: 38 years, 85% trainees BAI, QOLI b100% of
(2011) female modules
Arch et al. USA DSM-IV Panic/Ag, OCD, Community, mean RCT: GMBSR: 9 90 min (group) + 1 Clinical MINI CSR, PSWQ, MASQ-AA, GMBSR: 56% ?
(2013) SocPhob, GAD, PTSD age: 46 years, 17% GMBSR: 45 3 h (retreat) psychologist, BDI-II GCBT: 43%
female GCBT: 60 GCBT: 10 90 min (group) psychiatric nurse
Arch et al. USA DSM-IV anxiety Community, mean RCT: 12 60 min (individual) Doctoral level ADIS-IV ADIS CSR, ASI, PSWQ, FQ, CBT: 32% +

J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110

(2012) disorder age: 38 years, 52% CBT: 71 therapists QOLI, AAQ, ACT: 35%
female ACT: 57
Barlow et al. USA DSM-III panic disorder Community, mean RCT: CBT: 18 sessions (individual) Doctoral level ADIS-III STAI, BDI, CSR, PSC CBT: 0 ??+
(1984) or GAD age: 38 years, 35% CBT/biofeedback/RT: therapists,
female 10 graduate students
WLC: 10
Barrowclough UK Age over 55 and Community, mean RCT: CBT: 812 60 min (individual) CBT: doctoral level SCID-IV BAI, STAI-T, HARS-A, BDI, CBT: 11% ?
et al. DSM-IV Panic/Ag, age: 72 years, 77% CBT: 27 SC: 812 60 min sessions psychologist GDS SC: 4%
(2001) SocPhob, GAD, ADNOS female SC: 28 (individual) SC: counsellor
Berger et al. Switzerland, DSM-IV GAD, SocPhob, Community, mean RCT:TD- TD-CBT and DS-CBT: 8 sessions Master of Science SCID-IV BAI, BDI-II, BSI, SPS, SIAS, TD-iCBT: 9.1% ?
(2014) Germany, Panic/Ag age: 35 years, 56% iCBT: 44,DS-iCBT, over 8 weeks students and MI, PSWQ, BSQ, CAQ disorder specic
Austria female WLC: 44 psychologist iCBT: 11.3%,
WLC: 9.1%
Brenes et al. USA Age over 60, DSM-IV Community, mean RCT: T-CBT: 8 chapters over 8 sessions Doctoral level SCID-IV PSWQ, STAI-T, ASI, BDI, T-CBT: 10% ??
(2012) GAD, Panic/Ag, ADNOS age: 69 years, 83% T-CBT: 30 + 4 booster sessions students and HARS-A, SF-36 Psychoeducation:
female Psychoeducation: 30 masters level 3%
therapist
Bressi et al. Italy DSM-IV TR depressive Community, mean RCT: STPP: 40 45 minute sessions Psychiatrists SCID-IV SCL-90-R, CGI, IIP STPP: 20%
(2010) or anxiety disorder age: 37 years, 77% STPP: 30 TAU: 20%
female TAU: 30
Carlbring Sweden DSM-IV anxiety Community, mean RCT: iCBT: 610 modules over 10 Clinical psychology SCID-IV CORE-OM, MADRS-S, M = 7.96
et al. disorder (including age: 39 years, 76% iCBT: 27 weeks students BAI, QOLI modules
(2011) ADNOS) female Online forum: 27 completed
Craske et al. USA DSM-IV panic disorder Community, mean RCT:TD- DS-CBT (panic focus): 12 120 Doctoral students ADIS-IV ASI, FQ, CSR, BSI, SSS, BATs DS-CBT: 18% +
(2007) age: 39 years, 60% CBT: 33DS- min (group + 6 60 min and postdoctoral TD-CBT: 13%
female CBT: 32 individual) fellows
TD-CBT: 12 120 min (group:
panic focus) + 6 60 min
(individual: comorbid disorder
focus)
Cyranowski USA DSM-IV MDD + Community, mean Open trial: IPT: 1624 individual sessions Not reported SCID-IV HRSD, HARS, BDI, BAI, WLESQ IPT: 28% n/a
et al. Panic/Ag age: 37 years, 83% IPT: 18
(2005) female
Dear et al. Australia DSM-IV MDD, GAD, Community, mean Open trial: iCBT: 32 iCBT: 5 lessons over 8 weeks Clinical MINI DASS-21, PHQ-9, PSWQ, iCBT: 19% n/a
(2011) Panic/Ag, or SocPhob age: 44 years, 78% psychologists PDSS-SR, SP-12, GAD-7, K-10,
female SDS, NEO
Ellard et al. USA Primary DSM-IV Community, mean Open trial: Study 1: 815 60 min individual Doctoral students ADIS-IV BDI-II, BAI, PANAS-PA, Study 1: 8% n/a
(2010) anxiety disorder age: 30 years, 56% Study 1: 24 sessions and doctoral level PANAS-NA, OCI-R, PDSS-SR, Study 2: 17%
female Study 2: 18 Study 2: 1218 60 min psychologist PSWQ, SIAS, WSAS
individual sessions
Erickson et al. Canada DSM-IV PTSD, GAD, Community, mean RCT: 11 120 minute sessions Doctoral students SCID-IV GAF, BAI, BDI-II, ASI CBT: 41% +++++
(2007) Panic/Ag, OCD, age: 41 years, 64% CBT: 73
SocPhob female WLC: 79
Farchione USA DSM-IV anxiety Community, mean RCT: 18 60 minute sessions Doctoral students ADIS-IV HARS, HRSD, SIGH-A, SIGH-D, CBT: 15% ??
et al. disorder age: 30 years, 59% CBT: 26 and doctoral level BDI-II, BAI, PANAS-PA, PANAS-NA,
(2012) female WLC: 11 therapists PSWQ, SIAS, PDSS-SR, YBOCS, WSAS
Jakupcak USA DSM-IV PTSD and Community, mean Open trial: 8 60 minute sessions Not reported CAPS, SCID-IV PCL-M, BDI-II, QOLI 29% n/a
et al. MDD age: 28 years, 0% BA: 7
(2010) female
Johansson Sweden MDD + comorbid Community, mean RCT: iCBT (tailored): Masters level SCID-IV BDI-II, MADRS-S, BAI, QOLI iCBT (tailored)
et al. symptoms age: 45 years, 71% iCBT (tailored): 8 modules over 10 weeks psychologists 19%DS-
(2012) female 36DS- iCBT (DS): 810 chapters over 10 iCBT: 23% did not
iCBT: 37 weeks complete all
Online discussion modules
forum: 42
Johansson Sweden DSM-IV MDD, Community, mean RCT: I-AFPP: 50, I-AFPP: 8 modules over 10 weeks Masters level MINI GAD-7, PHQ-9 I-AFPP: 16%,
et al. SocPhob, Panic, GAD, age: 45 years, 82% online support: 50 students (telephone) Online Support:
(2013) DDNOS, ADNOS female 0%
Johnston Australia DSM-IV GAD, SocPhob, Community, mean RCT: iCBT-CO and iCBT-CL: 8 lessons iCBT-CO: MINI GAD-7, DASS-21, PSWQ, iCBT-CO: 26% +
et al. Panic/Ag age: 42 years, 59% iCBT-CO: 47 over 10 weeks registered (telephone) SIAS-6/SPS-6, PDSS-SR, PHQ-9, SDS iCBT-CL: 24%
(2011) female iCBT-CL: 46 psychologist
WLC: 46 iCBT-CL: doctoral
level psychologist
Kabat-Zinn USA DSM-III-R GAD, Community, mean Open trial: 8 120 min group sessions (+7.5 Not reported SCID-III HARS-A, HARS-D, BAI, BDI, MI, FSS 0% n/a

J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110

et al. Panic/Ag age: 42 years, 59% MBSR: 22 hour intensive meditation
(1992) female retreat)
Kenwright UK ICD-10 phobia or Community, mean Non-randomised 12 week ERP with clinician Nurse WHO (1992) FQ, WSAS, BDI 37% n/a
et al. Panic/Ag age 46 years, 41% trial: assistance interview
(2004) female CCBT: 17 checklist for
iCBT: 10 ICD-10
Kim et al. Republic of DSM-IV GAD or Community, mean RCT: MBCT: 8 90 m group sessions Psychiatrist SCID-IV HAM-A, BAI, SCL-90-R, HAM-D, BDI RCT: ++
(2009) Korea Panic/Ag age: 39 years, 26% MBCT: 24 Psychoed: 8 60 min group MBCT: 12.5%
female Psychoed: 22 sessions Psychoed: 9.1%
Lee et al. South Korea DSM-IV GAD, or Community, mean RCT: RCT: MBSM: Psychiatrist SCID-IV HAM-A STAI, HAM-D BDI, SCL-90-R MBSM: 12% ??+
(2007) Panic/Ag age: 38 years, 35% MBSM: 24 MBSM: 8 60 min session and meditation Psychoeducation:
female Psychoeducation: 22 Psychoeducation: 8 60 min specialist 9%
session Psychoeducation:
psychiatrist
Liu et al. China Primary ICD-10 Hospital clinic, mean RCT PST: 6 sessions over 16 weeks PST: psychologist, CISR CISR, HRSD, SF-36, PST: 42% +
(2007) depression, GAD, age: 43 years, 81% PST: 84 CL: 16 weeks social worker, CL: 20%
Panic/Ag, SocPhob, female CL: 85 TAU: 16 weeks psychiatric nurse TAU: 19%
OCD, MADD TAU: 85 CL: psychiatrist
Marks et al. UK ICD-10 phobia or Community, mean RCT: RCT: Nurse and WHO (1992) FQ, WSAS CCBT: 43% +
(2004) Panic/Ag age: 38 years, 69% CCBT: 37 6 60 min individual sessions psychologist interview CBT: 26%
female CBT: 39 over 10 weeks checklist for RT: 6%
RT: 17 ICD-10
McEvoy and Australia DSM-IV Anxiety Community, mean Open trial: 11 120 min group sessions Masters or doctoral MINI BDI-II, BAI M = 8.38 n/a
Nathan disorder or affective age: 35 years, 59% GCBT: 143 level psychologists sessions attended
(2007) disorder female
Newby et al. Australia DSM-IV GAD, MDD, or Community, mean RCT: iCBT: 6 lessons over 10 weeks Doctoral level MINI PHQ-9, GAD-7, K-10, BDI-II, PSWQ, 6.5% +
(2013) mixed anxiety and age: 44 years, 78% iCBT: 46 psychologist (telephone) WHODAS-II
depression female WLC: 53
Nixon and Australia DSM-IV MDD and Community, mean Open trial: 1216 60 min sessions Postgraduate CAPS and MINI CAPS, PDS, DASS-D, PTCI 30% n/a
Nearmy PTSD age: 45 years, 85% BA + ERP: 20 clinical
(2011) female psychologists
trainee
Norton and USA DSM-IV Panic/Ag, Community, mean RCT:TD- 12 120 m sessions Doctoral level ADIS-IV ADIS CSR, STAI, PDSS, SPDQ, TD-GCBT: +++
Barrera SocPhob, GAD age: 31 years, 50% GCBT graduate students GADQ-IV, BDI 22%DS-
(2012) female (transdiagnostic): GCBT: 39%
23
GCBT
(disorder-spec): 23
Norton USA DSM-IV anxiety Community, mean Open trial: 12 120 m sessions Doctoral level ADIS-IV STAI-S, M = 7 sessions n/a
(2008) disorder age: 31 years, 56% GCBT: 52 graduate students attended
female

(continued on next page)

99
 100
J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Table 1 (continued)

Study Country Participants Sample Design Intervention Clinician Diagnostic Measures Attrition Qual
measure

Norton USA DSM-IV anxiety Community, mean RCT:TD- TD-GCBT: 12 120 m sessions Doctoral level ADIS-IV ADIS CSR, CGI, ADDQ, BAI, PDSS, RCT:TD- +++
(2012) disorder age: 33 years, 62% GCBT: 65 RT: 12 120 m sessions graduate students SPDQ, GAD-Q-IV, STAI-S GCBT: 30%
female RT: 22 RT: 57%
Norton et al. USA DSM-IV anxiety Community, mean RCT: 12 150 m (group) Doctoral level ADIS-IV DASS-42, MASQ GCBT: 25% ++++
(2004)a disorder age: 42 years, 61% GCBT: 12 graduate students
female WLC: 11
Patel et al. India ICD-10 common Community, mean RCT: CBT: up to 6 sessions Trained therapists CISR CISR, BDQ CBT: 12%
(2003) mental disorders age: 49 years, 81% CBT: 150 Medications: 11%
female Medications: 150 Placebo: 5%
Placebo: 150
Proudfoot UK ICD-10 depression, Community, mean RCT: CCBT: 9 sessions with nurse Nurses CISR BDI-II, BAI, WSAS CCBT: 33%
et al. mixed age: 44 years, 74% CCBT: 89 guidance (computerised) TAU: 30%
(2004) anxiety/depression or female TAU: 78
anxiety disorder
Proudfoot UK ICD-10 depression, Community, mean RCT: CCBT: 9 sessions with nurse Nurses CISR BDI-II, BAI, WSAS CCBT: 27%
et al. mixed age: 44 years, 74% CCBT: 146 guidance (computerised) TAU: 24%
(2003) anxiety/depression or female TAU: 128
anxiety disorder
Radley et al. UK DSM-IV anxiety Community elderly Open trial: GCBT: 7 90 min Psychiatrist, HADS N 10 HAD-A, HAM-A, GAS, STAI, FI, PSI, GCBT: 30% n/a
(1997) disorders patients, mean age: GCBT: 9 clinical CAQ, ELI
71 years, 100% female psychologist and
psychiatric nurse
Ree and Australia DSM-IV mood or Private clinic, mean Open trial: 8 150 min Clinical SCID BDI, DASS-42 MBCT: 11% n/a
Craigie anxiety disorders age: 40 years, 77% MBCT: 26 psychologist and
(2007) female clinical psychology
trainee
Schmidt et al. UK Primary DSM-IV Community, mean RCT: F-SET: 10 120 min s Masters or doctoral SCID-IV ASI, BDI-II, MI, SDS, SPRAS, CGI F-SET = 7% +
(2012) Panic/Ag, GAD, age: 36 years, 72% F-SET: 57 level therapists
SocPhob female WLC: 39
Titov et al. Australia DSM-IV GAD, SocPhob, Community, mean RCT: 8 lessons over 10 weeks Masters or doctoral MINI PSWQ, SPSQ, PDSS-SR, GAD-7, iCBT: 19% +
(2011) Panic/Ag age: 44 years, 73% iCBT: 37 level clinical (telephone) PHQ-9, SDS, K-10, DASS-21, NEO
female WLC: 38 psychologist
Titov et al. Australia DSM-IV GAD, Panic/Ag, Community, mean RCT: 6 lessons over 10 weeks Masters or doctoral MINI PHQ-9, GAD-7, Social Phobia-12, iCBT: 25% +
(2010) SocPhob, or MDD age: 39 years, 68% iCBT: 40 level clinical (telephone) PDSS-SR, SDSK-10, BDI-II, PSWQ,
female WLC: 38 psychologist WHODAS-II
 Vollestadet al. Norway DSM-IV anxiety Community, mean RCT: MBSR: 8 150 minute sessions Not reported MINI BAI, PSWQ, STAI, BDI-II, SCL-90 MBSR: 18% ??
(2011) disorder age: 46 years, 17% MBSR: 39 WLC:
female WLC: 37
Westra et al. Canada DSM-IV anxiety Community hospital Open trial: 10 120 minute sessions Allied health SCID-IV BDI-II, BAI 15% n/a
(2007) disorders unit, mean age 41 GCBT: 115 professionals
years, 63% female
Wetherell Australia Age N 60 years and Community, mean RCT: CBT: 12 60 min Masters and ADIS-IV HARS, PSWQ, BDI-II, SF-36 Modular CBT: ?+
et al. DSM-IV GAD or age: 72 years, 84% Modular CBT: 15 E-COM: 12 60 minute sessions doctoral level 20%
(2009) ADNOS female E-COM: 15 clinicians E-COM: 20%
Wuthrich and Australia Age N 60 and DSM-IV Community, mean RCT 12 120 min Clinical ADIS-IV GDS, CES-D, GAI, PSWQ, SF-12 GCBT: 12% ?
Rapee primary mood or age: 67 years, 65% GCBT: 25 psychologist and WLC: 3%
(2013) anxiety disorder female WLC: 35 graduate students
Zou et al. Australia Age N 59 and DSM-IV Community, mean Open trial: iCBT: 22 5 lessons over 8 weeks Clinical MINI GAD-7, PHQ-9, SDS 0% n/a
(2012) primary mood or age: 66 years, 68% psychologist
anxiety disorder female

Note. Countries: USA = United States of America, UK = United Kingdom. Treatments: ACT = acceptance and commitment therapy, CBT = cognitive behaviour therapy, CG-ERP = clinician guided exposure and response prevention, DS = disorder-
specic, E-COM = enhanced community treatment, F-SET = false safety behaviour elimination therapy, GCBT = group cognitive behaviour therapy, GMBSR = group mindfulness based stress reduction, iCBT = internet-delivered cognitive behav-
ioural therapy, iCBT (CO) = internet cognitive behavioural therapy with coach guidance, iCBT (CL) = internet cognitive behavioural therapy with clinician guidance. RT = relaxation training, SC = supportive counselling, SG-ERP = self-guided
exposure and response prevention, STPP: short-term psychodynamic psychotherapy, T-CBT = telephone-delivered cognitive behaviour therapy, TAU = treatment as usual, TD = transdiagnostic WLC = waiting list control. Participants: GAD =
generalised anxiety disorder, MADD = mixed anxiety and depressive disorder, MDD = major depressive disorder, OCD = obsessive compulsive disorder, Panic/Ag = panic disorder and/or agoraphobia, PTSD = posttraumatic stress disorder,

J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110

SocPhob = social phobia or social anxiety disorder. Trial types: RCT = randomised controlled trial. Measures: ADIS-IV: Anxiety Disorders Interview Schedule for DSM-IV; ADIS-R: Anxiety Disorders Interview Schedule-Revised; AKUADS: Aga
Khan University Anxiety and Depression Scale; ASI: Anxiety Sensitivity Index, BAI: Beck Anxiety Inventory; BAT: Behavioural Activation Test; BDI-II: Beck Depression Inventory, second edition; BSI : Brief Symptom Inventory; CAQ: Cognitive Anxiety
Questionnaire; CGI: Clinical Global Improvement-Patient Rating, CID: Clinical Interview for Depression; CORE-OM: Clinical Outcome's in Routine Evaluation; DASS-21: Depression Anxiety and Stress Scales (DASS) 21-item version; DASS-42: Depres-
sion Anxiety and Stress Scales (DASS) 42-item version; EFI: Effects on Life Inventory; FI: Fear Inventory; FNE: Fear of Negative Evaluation Scale; FQ: Fear Questionnaire; FQAD: Fear Questionnaire AnxietyDepression Subscale; FQSP: Fear Question-
naire Social Phobia Subscale; FSS: Fear Survey Schedule; GAD-7: Generalised Anxiety Disorder 7-item scale; GAS: Generalised Anxiety Scale from the Guys/Age Concerned Survey; HADS-A: Hospital Anxiety and Depression Scale Anxiety Subscale;
HADS-D: Hospital Anxiety and Depression Scale Depression Subscale; HAI: Health Anxiety Inventory; HARS: Hamilton Anxiety Rating Scale; HRSD: Hamilton Rating Scale for Depression; K-10: Kessler 10-item; LSAS-SR: Liebowitz Social Anxiety
Scale self-report version; MADRS-S: MontgomeryAsberg Depression Rating Scale self rated version; MASQ: Mood and Symptoms Questionnaire; MIA: Mobility Inventory for Agoraphobia; NEO-N: NEO-Five Factor InventoryNeuroticism Sub-
scale; OCI-R: ObsessiveCompulsive InventoryRevised Version; PANAS: Positive and negative affect scale; PAS: Panic and Agoraphobia Scale; PCL-C Posttraumatic Stress Disorder Checklist Civilian; PSC: Psychosomatic Symptom Checklist; PDSS-
SR: Panic Disorder Severity Scale Self report; PRCA: Personal Report of Communication Apprehension; PRCS: Personal Report of Condence as a Speaker; PSI: Physical Symptoms inventory; PSWQ: Penn State Worry Questionnaire; QLESQ: Quality
of Life Enjoyment and Satisfaction Scale; QOLI: Quality of Life Inventory; SAD: Social Avoidance and Distress Scale; SDS: Sheehan Disability Scale; SIAS: Social Interaction Anxiety Scale; SIAS/SPS6 composite: Social Interaction Anxiety Scale and Social
Phobia Scale 6-item composite; SIGH: Structured Interview Guide for the Hamilton Anxiety Rating Scale (A = Anxiety, D = Depression); SPSQ: Social Phobia Screening Questionnaire; SQ: Kellner's Symptom Questionnaire; SSPS: Self-Statements
During Public Speaking (positive and negative subscales); SSS: Subjective Symptoms Scale (measure of interference with daily functioning) STAI: State-Trait Anxiety Inventory (T = trait, S = state); WSAS: Work and Social Adjustment Scale; YBOCS:
Yale Brown Obsessive Compulsive Scale; Zung SRSD: Zung Self-Rating Scale for Depression. Qual = risk of bias coding where - = low risk of bias, + = high risk of bias, and ? = unclear risk of bias on the following indices: random sequencing,
allocation concealment, blinding of outcome assessment, incomplete outcome data, and selective reporting.
a
The data from this study was analysed together with Norton and Hope (2005) who reported depression-related outcomes.

101
102 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Fig. 1. Study ow chart.
4.5. Synthesis of results
4.5.1. What are the effects of transdiagnostic psychological treatments on
primary outcomes between pre- and post-treatment (uncontrolled effect
sizes)?
The analysis of studies that reported pre- to post-treatment effects
showed a mean effect size (Hedges g) of 0.86 for anxiety (n = 50)
(95%CI: .75.96, p b .001), 0.91 for depression (n = 41) (95%CI: .78
1.04, p b .001), and .69 for quality of life (QOL) (n = 29) (95%CI:
.59.78, p b .001). Hedges g values for transdiagnostic treatments, at
post-treatment on anxiety and depression are presented in Figs. 2 and
3 respectively (and see Fig. 4 for QOL outcomes). Heterogeneity was
moderate to high, and signicant amongst these studies (anxiety:
I2 = 72, depression: I2 = 78, QOL: I2 = 43). Heterogeneity remained
signicant even after removal of studies investigating older samples,
after removal of studies with high proportion of male participants, and
after removal of potential outliers (ps b .001).
4.6. Subgroup analyses of uncontrolled pre- to post-treatment effects
4.6.1. Is there a differential effect of transdiagnostic interventions on depres-
sion versus anxiety symptoms between pre- and post-treatment?
For the rst subgroup analysis, we selected only the studies that re-
ported both depression and anxiety (n = 39) outcomes to evaluate the
magnitude of the uncontrolled (before and after treatment) effects for
depression and anxiety symptoms when both are targeted as outcomes
and to compare these effects. Transdiagnostic interventions had large
effects on both anxiety (g = .85) and depression (g = .92), but overall
had a larger effect on depression symptoms compared to anxiety symp-
toms (Q = 3.94, df = 1, p = .047).
5
For the purposes of the subgroup analyses, we coded Craske et al. (2007) as group
based treatment because the majority of treatment was provided within the group setting,
and Brenes et al., 2012 as individual, as the telephone-based CBT was administered
individually.
 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Fig. 2. Forest plot of within-group effect of transdiagnostic treatment on self-reported anxiety (uncontrolled prepost-effect sizes).
4.6.2. What is the impact of treatment delivery format on pre-to-post-
treatment outcomes?
In the next subgroup analyses (see Table 2), we compared the mean
uncontrolled effect for subgroups based on treatment delivery format
(individual, group and computer/internet interventions), and found sig-
nicant differences for anxiety symptoms (Q = 31.75, df = 2, p b .001).
While we were unable to compare between specic formats, obser-
vation of the means suggests that group-based treatments yielded
lower effects (n = 18: g = .70) than individual (n = 15: g = .97)
and computer/internet treatments (n = 17: g = .96).5 There were
also signicant group differences in the size of the uncontrolled ef-
fects for depression symptoms (Q = 12.41, df = 2, p = .002), with
observation of the means indicating that the highest effect sizes
were found in computerised/internet treatments (n = 16: g =
.96), followed by group-based face-to-face treatments (n = 12:
g = .89) and individual treatments (n = 13: g = .86). However,
there were no signicant differences between studies which used
computer, group, or individual treatment for quality of life measures
(individual, n = 10: g = .68, computer/internet, n = 15: g = .74,
group, n = 4: g = .65, Q = 3.5, df = 2, p = .17).
103
4.6.3. What is the impact of treatment type on pre-to-post-treatment
outcomes?
In the next subgroup analysis (see Table 2), we compared the mean
uncontrolled effect for subgroups based on treatment type (CBT versus
mindfulness/acceptance based interventions), and showed that there
was a signicant difference favouring CBT compared with mindful-
ness/acceptance based interventions for reducing anxiety symptoms
(Q = 7.95, df = 1, p = .005) (CBT, n = 40: g = .88, mindfulness/
acceptance, n = 7: g = .61). This difference remained signicant even
after removing an outlier (Arch et al., 2013). There were no signicant
differences across treatment type for depression symptoms (Q = .78,
df = 1, p = .38), although observation of the uncontrolled effect esti-
mate suggests that it was slightly higher (albeit not signicant) in the
mindfulness/acceptance interventions (mindfulness/acceptance, n =
6: g = .92, CBT, n = 31: g = .84). Quality of life was not assessed be-
cause only one mindfulness/acceptance study included an appropriate
measure.
4.7. Effects of transdiagnostic psychological treatments on primary outcomes
between pre-treatment and follow-up (uncontrolled effect sizes)
There were 24 studies that reported follow-up data (15 reporting
3-month follow-up and 9 studies reported 6-month follow-up data).
Pre-treatment to follow-up uncontrolled effects were large on average
104 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Fig. 3. Forest plot of the within-group effect of transdiagnostic treatment on self-reported depression (uncontrolled prepost-effect sizes).
for anxiety (g = 0.87, n = 24, 95%CI: .731.01, p b .001), depression
(g = 0.91, n = 21, 95%CI: .701.12, p b .001), and medium to large for
quality of life (g = 0.75, n = 16, 95%CI: .59.91, p b .001). Heterogeneity
was moderate to high across all measures (anxiety: I2 = 72, depression:
I2 = 88, QOL: I2 = 73).
4.8. Between-group effects of transdiagnostic psychological interventions
versus control groups
We compared the effects of transdiagnostic treatments with control
groups in 24 studies (see Figs. 5, 6, and 7 for forest plots of anxiety,
depression, and quality of life outcomes respectively). These analyses
excluded studies that compared transdiagnostic treatments with alter-
native transdiagnostic interventions, with disorder-specic interven-
tions, and comparisons between alternative versions of the same
programme. The overall effect was medium for anxiety severity (n =
24, g = .65, 95%CI: .51.79), with moderate and signicant heterogene-
ity (I2 = 51), large for depression (n = 22, g = .80, 95%CI: .62.98), with
moderate heterogeneity (I2 = 66), and medium for quality of life
measures (n = 13, g = .46, 95%CI: .34.57).
4.8.1. Subgroup analyses: does the magnitude of effect depend on the type
of control condition?
Next, we conducted a subgroup analysis to explore whether the
magnitude of between-group controlled effect differs according to the
type of control condition (WLC: n = 13, TAU: n = 4, attention control:
n = 8) and found that there was a signicant overall effect of control
condition on anxiety outcomes (Q = 19.56, df = 2, p b .001).
Comparisons between transdiagnostic treatments versus TAU control
conditions had the smallest differences (g = .24, 95%CI: .05.43,
p b .01), whereas there were large differences in comparisons between
transdiagnostic treatments and attention control conditions (g = .80,
95%CI: .521.08, p b .001) and WLC conditions (g = .70, 95%CI:
.56.84, p b .001).
For depression, there was also a signicant difference in the magni-
tude of effect depending on the type of control condition (Q = 7.08,
df = 2, p = .029), with comparisons with TAU control conditions
again showing the smallest difference (g = .57, n = 4, 95%CI: .38.76,
p b .001), followed by attention controls (g = .69, n = 8, 95%CI:
.46.91, p b .001) and WLC having the largest difference compared to
transdiagnostic treatments (g = 1.0, n = 12, 95%CI: .691.30,
p b .001). The results of quality of life measures did not reach conven-
tional signicance (Q = 1.40, df = 2, p = .49) (WLC: n = 7, g = .53,
95%CI: .34.71, p b .001, TAU: n = 2, g = .37, 95%CI: .16.58, p b .001,
and attention control conditions: n = 4, g = .42, 95%CI: .16.68,
p b .001), however there were only two comparisons with TAU control
conditions.
4.9. Risk of bias across studies
There was some evidence of publication bias as demonstrated by in-
spection of the funnel plot (see Appendix C), and using Duval and
Tweedie's Trim and Fill procedure. After adjusting for publication bias
using the Trim and Fill procedure, the estimate of the mean effect size
comparing transdiagnostic interventions to controls on anxiety reduced
 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110 105

Fig. 4. Forest plot of the within-group effect of transdiagnostic treatment on self-reported quality of life (uncontrolled prepost-effect sizes).

from g = .66 to g = .46 (n = 9 studies removed). There was no evidence 5. Discussion

of publication bias for depression outcomes, nor for quality of life.
4.10. Comparisons between transdiagnostic treatments versus disorder-
specic treatments
Only four studies compared transdiagnostic interventions to
disorder-specic treatment control conditions. The analysis showed
that there were no signicant differences between transdiagnostic
and disorder-specic treatments for anxiety (n = 4, g = .15,
95%CI: .09.38, p = .22, I2 = 0, p N .05), but there were signicant
differences for depression outcomes, with the results in favour of
the transdiagnostic treatments (n = 3, g = .58, 95%CI: .0031.16,
p = 0.05). Notably, there was high heterogeneity (I2 = 76, p b .05)
amongst these effects.
In the current study, we systematically reviewed the existing litera-
ture on transdiagnostic psychological treatments for depression and
anxiety disorders in adults. We examined their overall effect on symp-
toms of depression, anxiety, and quality of life, as well as the relative
efcacy of transdiagnostic treatments compared to waitlist controls
(WLC), treatment-as-usual conditions (TAU), attention control con-
ditions and disorder-specic treatments. We identied 47 studies
(including 31 RCTs) with a total of 3705 participants evaluating
transdiagnostic treatments. The majority of included studies inves-
tigated CBT protocols or variants of CBT, and a variety of treatment
delivery formats were evaluated, with face-to-face group treatment
the most commonly studied.
Analysis of the results across all studies that reported baseline and
post-treatment data showed that transdiagnostic treatments lead to
large and signicant reductions in both anxiety and depression, and

Table 2
Subgroup analyses.

Subgroup analyses Measure N g 95%CI I2

Treatment format Individual Anxiety 15 .97 .731.21 75.7

Computer/internet 17 .96 .851.07 30.3
Group 18 .70 .53.87 72.4
Individual Depression 13 .86 .661.05 57.0
Computer/internet 16 .96 .761.16 77.6
Group 12 .89 .621.17 85.5
Individual QOL 10 .68 .46.90 67.7
Computer/internet 15 .74 .65.83 0
Group 4 .65 .42.89 27.2
Treatment type CBT Anxiety 40 .88 .771.0 71.6
Mindfulness/acceptance 7 .61 .37.86 66.7
CBT Depression 31 .84 .71.95 73.1
Mindfulness/acceptance 6 .92 .441.39 86.3

Note. CBT = cognitive behavioural therapy; CI = condence interval; N = number of studies; QOL = quality of life.
106 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110

Fig. 5. Forest plot of controlled between-group effect sizes for comparisons between transdiagnostic treatments and control conditions on self-reported anxiety.

moderate improvements in quality of life. Results at short-term follow-up
(three to six months after treatment) also suggest that these positive out-
comes are maintained following treatment. However, the evidence of
high heterogeneity suggests that there were signicant differences in
treatment effects across studies. Importantly transdiagnostic treatments
seem to have a large effect on both depression and anxiety symptoms
in studies that targeted both outcomes, providing important support for
their transdiagnostic utility. There was also some preliminary evidence
that in studies that measured both outcomes, the effects were larger for
depression than anxiety. Because the majority of studies only examined
general measures of anxiety, the impact on specic mechanisms and
outcomes that characterise different fears (e.g., fears of social situations,
or fears of physical sensations) still remains unclear. Nevertheless, these
positive effects of transdiagnostic treatments are consistent with the
ndings of previous meta-analyses (Norton & Philipp, 2008) and a narra-
tive review (McEvoy et al., 2009). Notably, our effect estimates for anxiety
were lower than a previous meta-analysis of TD-CBT (Norton & Philipp,
2008 who found d = 1.29). This may be because we evaluated a wider
range of treatment protocols, or because of methodological differences
between studies. For example, we only included studies which employed
more rigorous inclusion criteria (e.g., used structured diagnostic in-
terviews). We also used Hedges g rather than Cohen's d, which provides a
more conservative estimate of effect sizes by adjusting for small sample
sizes.
The majority of trials compared transdiagnostic treatments to WLC,
but transdiagnostic treatments were also compared to a range of atten-
tion control conditions that included psychoeducation, online discus-
sion forums, and supportive counselling. There was only one direct

Fig. 6. Forest plot of controlled between-group effect sizes for comparisons between transdiagnostic treatments and control conditions on self-reported depression.
 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Fig. 7. Forest plot of controlled between-group effect sizes for comparisons between transdiagnostic treatments and control conditions on self-reported quality of life.
comparison with medication, highlighting the need for future RCTs to
compare the efcacy of transdiagnostic psychological treatments to
pharmacological interventions for depression and anxiety. Analysis of
the pooled results from the RCTs demonstrated that transdiagnostic
treatments outperformed controls on all three outcome measures. Sim-
ilar to Reinholt and Krogh (2014), we found moderate differences be-
tween transdiagnostic treatments and control conditions in anxiety
severity at post-treatment. We also demonstrated large overall differ-
ences between transdiagnostic treatments and control condition in de-
pression severity, and moderate differences in quality of life at post-
treatment. This is the rst meta-analysis to demonstrate the positive
impact of transdiagnostic treatments on improving quality of life. Com-
pared to the effects found in a recent meta-analysis of CBT for anxiety
disorders on quality of life outcomes, our uncontrolled effects were
slightly higher (g = .69, versus g = .54 in Hofmann, Wu, & Boettcher,
2014), but the controlled effects were slightly lower (g = .46 versus
g = .56 averaged across group, individual and internet treatments in
Hofmann et al., 2014). Heterogeneity was also moderate to high for all
outcome measures, suggesting there was signicant variability amongst
these effects. We also found some evidence of publication bias for anx-
iety outcomes, which suggests these effect estimates may be inated. In
addition, the quality assessments revealed that 24 out of 31 RCTs were
at high risk of nding biased estimates of effects, based on commonly
used risk of bias indices (Higgins & Green, 2011).
Our preliminary results suggest that the nature of the control con-
dition inuenced the size of treatment effects in this study, which may
in part explain some of the heterogeneity we observed. We found
large differences between transdiagnostic treatments compared to
both WLC and attention control conditions (e.g., psychoeducation, on-
line discussion forums, relaxation training). In contrast, we found only
small overall differences in studies that compared transdiagnostic treat-
ments to TAU or usual care control conditions in anxiety outcomes,
which supports the ndings of Reinholt and Krogh (2014). These results
demonstrate that transdiagnostic treatments have benets over and
above the natural recovery processes that often occur across time dur-
ing the course of depression and anxiety problems (e.g., spontaneous
recovery), as well as the non-specic or common therapeutic factors
associated with treatment that may account for symptom improvement
(e.g., therapeutic alliance, regular assessment and monitoring).
It is unclear why there was a smaller difference between transdiag-
nostic treatments compared to usual care, whereas transdiagnostic
treatments outperformed other control conditions to a larger degree.
It is possible that the use of TAU control conditions was confounded
with the type of treatment, the duration of the treatment period being
evaluated, the quality of the study, or the country where the study
was conducted. Three of the four studies were coded as having low
107
risk of bias on all measures. Two of the four studies that used TAU con-
trol conditions evaluated alternative forms of psychotherapy (other
than CBT) such as problem-solving therapy and short-term psychody-
namic psychotherapy in China and Italy, respectively. The remaining
two studies compared computerised CBT to TAU in the United
Kingdom. If participants with depression and anxiety disorders received
relatively effective psychological and/or pharmacological treatments as
part of their usual care in these studies, this may account for the smaller
difference we observed. For example, one of these studies (Bressi,
Porcellana, Marinaccio, Nocito, & Magri, 2010) compared 40 sessions
of short-term psychodynamic psychotherapy for depression and anxi-
ety delivered by Psychiatrists over a 12-month period to 12 months of
TAU, which consisted of drug treatment combined with interviews
with a Psychiatrist, which could be scheduled up to four times per
month. In this study, it is likely that TAU provided a strong comparison
condition. Because there was no independent coding of what the inter-
views entailed, it is even possible that the Psychiatrists in the TAU con-
dition delivered similar psychotherapy to the treatment condition.
These ndings highlight the need for closer examination of what TAU
entails, and highlight the need for future meta-analytic reviews to dis-
tinguish studies that use TAU versus WL/no-treatment controls when
estimating the value of new treatments (Watts, Turnell, Kladnitski,
Newby & Andrews, 2015).
Another aim of this study was to examine the efcacy of transdiag-
nostic treatments relative to disorder-specic treatments, in an attempt
to clarify the mixed ndings in the literature. When compared to the
effect sizes found in recent meta-analyses of disorder-specic interven-
tions, the overall effect size difference compared to control conditions is
slightly lower than the 0.84 large effect size for reduction in anxiety symp-
toms across psychological treatments for GAD (Cuijpers et al., 2014), but
higher than average effect sizes for psychological interventions for de-
pression (0.53) (Cuijpers, Andersson, Donker, & Van Straten, 2011). We
found only four studies that directly compared transdiagnostic versus
disorder-specic treatments for anxiety and depression, which makes it
difcult to draw denitive conclusions about their relative impact. How-
ever, our preliminary results suggested that on average, transdiagnostic
treatments are at least as efcacious as disorder-specic treatments in
reducing anxiety symptoms, and there may be small effects in favour of
the transdiagnostic treatments on depression symptoms. Although
these results are very preliminary and need to be replicated with a larger
number of studies, it suggests that there may be advantages of using a
transdiagnostic approach for treating depression symptoms when de-
pression is experienced in the context of comorbid anxiety symptoms.
While comparing their differential effects on symptoms is important, fu-
ture studies also need to compare transdiagnostic and disorder-specic
treatments on measures of acceptability to patients and clinicians, their
108 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
effects on both primary and secondary comorbidities, as well as their cost-
effectiveness.
We included studies of a range of interventions that adopted the
transdiagnostic approach to treatment. Although some studies speci-
cally labelled their intervention as a transdiagnostic treatment, many
did not use this terminology. The high level of heterogeneity suggests
that there were signicant differences in treatment outcomes across
these studies, which may be due to critical differences between proto-
cols. Our preliminary ndings indicated potential sources of the hetero-
geneity: both treatment type (CBT versus mindfulness/acceptance) and
delivery format (face-to-face individual, face-to-face group, or clinician-
guided computerised/internet delivery) inuenced outcomes. For exam-
ple, on average we found that group-based face-to-face transdiagnostic
treatments had the smallest (but still moderate to large) effects on anxi-
ety and depression symptoms. Internet-delivered and computerised
treatments had the largest effect sizes for depression, and had large effects
on anxiety measures, which were similar to the effects found on anxiety
measures in individual treatments.
Computerised and internet therapies typically comprise online/
computerised modules or lessons that are delivered over a dened
treatment period (e.g., 12 weeks). These text-based lessons/modules
teach the patient about depression/anxiety and how to manage their
symptoms using practical skills (e.g., graded exposure and thought chal-
lenging in CBT). Lessons are typically supplemented with homework
exercises to consolidate new learning and encourage skills practice in
the patient's daily life, and clinician guidance is often provided via
phone or email. Because of their standardised nature, computerised
treatments have high treatment delity. Since our search was conduct-
ed, two additional RCTs have found large and positive effects of
transdiagnostic internet CBT for mixed anxiety disorders, providing
more evidence in support for their efcacy (Dear, Zou, Ali, et al., 2014;
Nordgren et al., 2014). Future research is critically needed to identify
why different treatment effects are observed across delivery formats,
and whether it is actually the format of delivery or other aspects of
the treatment protocol or participant samples that inuence outcomes.
Our results also suggested that, on average, CBT signicantly out-
performed mindfulness/acceptance-based treatments in reducing anxi-
ety symptoms, but not depression. These results need to be interpreted
with caution given that there were relatively few studies of third-wave
therapies, and a diverse range of therapies were included in this catego-
ry. It is possible that CBT was more powerful in reducing anxiety overall
because exposure-based techniques are included as a core component
in these protocols. Given that the most commonly used measure of anx-
iety was the BAI, which has been argued to capture symptoms of auto-
nomic arousal rather than other features of anxiety (e.g., worry), it is
also possible that CBT is simply more powerful in reducing anxious
arousal. This hypothesis is in line with the non-signicant but small ef-
fects in favour of CBT over MBSR on measures of arousal found by Arch
et al. (2013). Nevertheless, our study highlights the need for more high
quality, adequately powered research studies to compare the differ-
ential effects of the CBT versus mindfulness- and acceptance-based
transdiagnostic treatments on a range of outcome measures (anxiety
and depression), and to identify active components that contribute to
positive outcomes. Further studies are also needed to explore whether
combining conventional or adapted transdiagnostic CBT protocols with
mindfulness and/or acceptance-based treatment approaches has any
added benet for the treatment of emotional disorders.
While our results go part way to explaining some of the heterogene-
ity amongst treatment effects, more research is needed to understand
additional sources of heterogeneity. Our review included a diverse set
of protocols that included integral (where all patients receive the
same xed protocol), modular and case-formulation driven treatments,
as well as tailored interventions that target a patient's primary disorder
and comorbid symptoms. The protocols also varied in number and type
of anxiety and depressive disorders that were targeted, and the samples
varied in terms of severity and comorbidity. It remains unclear which type
of transdiagnostic treatment approach is more effective, whether thera-
pist experience inuences outcomes, and whether the transdiagnostic
treatments are more suitable to specic symptom proles or diagnostic
combinations. In future, it would be helpful to independently review all
of the transdiagnostic treatment protocols that have been evaluated to-
date, and compare the treatment components/elements in the protocols
that yield the highest effects to those that yield the lowest effects. This
may assist in identifying the treatment components that promote positive
outcomes for patients, and the most efcacious approach to use when
implementing a transdiagnostic protocol.
5.1. Limitations
The results of this review need to be interpreted in the context of
some limitations. First, due to practical reasons, we only included stud-
ies that were published in English; grey literature and unpublished
studies were not included in the meta-analysis. Second, our search
was restricted to a limited set of databases (PSYCInfo and PubMed). Al-
though we attempted to address this by examining the reference lists of
previous meta-analyses and relevant papers, we may have unintention-
ally omitted articles that met our inclusion criteria because of our re-
stricted search. Third, because we restricted our inclusion criteria to
manualised interventions, this is likely to have introduced some bias to-
wards the inclusion of studies of CBT, and exclusion of psychological
treatments from other treatment orientations such as psychodynamic
psychotherapies (e.g., see Knekt et al., 2008). Interestingly, evaluation
of a unied protocol for the transdiagnostic psychodynamic treatment
of anxiety disorders is now underway which will provide a better un-
derstanding of the efcacy of psychodynamic treatments that adopt a
transdiagnostic treatment approach (Leichsenring & Salzer, 2014).
Fourth, we only examined the impact of transdiagnostic treatments
using self-report measures because clinician-rated instruments were not
consistently used across studies, which may have resulted in inated es-
timates of effect sizes. Future studies need to examine the impact of
these treatments on both clinician-rated and self-reported instruments.
Fifth, we did not use a measure of bias risk in the non-randomised studies,
although the major methodological limitation of those studies is the ab-
sence of a control condition and no randomised estimate of effects. Final-
ly, we used a range of subgroup analyses to explore the possible reasons
for the high heterogeneity found in the treatment effects. Because the re-
sults from our subgroup analyses are purely observational (Borenstein &
Higgins, 2013), we cannot conclude that treatment type and format
caused the differences in outcomes we observed in outcomes. It is also
possible that the subgroups differed in other important ways that inu-
enced outcomes (e.g., therapist experience, choice of assessment mea-
sure, or quality of study design).
6. Conclusions
Despite these limitations, the strength of our study is that it is the
rst to review systematically the comprehensive set of transdiagnostic
treatments, across treatment types (e.g., CBT, mindfulness/acceptance
and other treatment approaches) and delivery formats (e.g., face-to-
face individual and group, as well as computerised treatments). Our re-
sults provide evidence in support of the efcacy of transdiagnostic treat-
ments in reducing depression and anxiety, and improving quality of life.
The quality of RCTs was low overall, and heterogeneity was high. Fur-
ther high quality RCTs are now needed to explore the sources of this
heterogeneity to identify the most effective treatment components
and designs, and to understand how transdiagnostic treatments work.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.cpr.2015.06.002.
 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Role of funding sources
Funding for this study was provided by the National Health and Med-
ical Research Council of Australia (NHMRC) in the form of an Early Career
Fellowship awarded to Dr Jill Newby (NHMRC grant number 1037787).
This work was supported by the UK Medical Research Council [MRC-
A060-5PQ60], Professor Tim Dalgleish is supported by a UK Medical Re-
search Council Intramural Programme Grant (MC_US_A060_0019). The
MRC and NHMRC had no role in the study design, collection, data analysis
or interpretation of the data, writing the manuscript or the decision to
submit the paper for publication.
Contributors
Dr Newby and Professors Kuyken and Dalgleish designed the
study and wrote the protocol and search strategy. Dr Newby and
Dr McKinnon conducted the searches, screened the titles, abstracts,
and full-texts for eligibility for inclusion into the meta-analysis. Pro-
fessor Dalgleish provided a third independent review for any arti-
cles for which there was disagreement between the rst two
reviewers. Dr Newby and Dr McKinnon coded the risk of bias of all
RCTs. Dr Newby extracted the data from manuscripts and conducted
the data analysis. All authors contributed to and have approved the
nal version of the manuscript.
Conict of interest
All authors declare that they have no conicts of interest.
Acknowledgements
The authors wish to thank Professor Pim Cuijpers who provided
statistical advice regarding this study.
References6
American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, DC: Author (text rev.).
American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders
(Fifth Edition ). Arlington, VA: American Psychiatric Association.
Andrews, G., Goldberg, D.P., Krueger, R.F., Carpenter, W.T., Hyman, S.E., Sachdev, P., et al.
(2009). Exploring the feasibility of a meta-structure for DSM-V and ICD-11: Could it
improve utility and validity? Psychological Medicine, 39(12), 19932000. http://dx.
doi.org/10.1017/s0033291709990250.
*Arch, J.J., Ayers, C.R., Baker, A., Almklov, E., Dean, D.J., & Craske, M.G. (2013). Randomized
clinical trial of adapted mindfulness-based stress reduction versus group cognitive
behavioral therapy for heterogeneous anxiety disorders. Behaviour Research and
Therapy, 51(45), 185196. http://dx.doi.org/10.1016/j.brat.2013.01.003.
*Arch, J.J., Eifert, G.H., Davies, C., Vilardaga, J.C.P., Rose, R.D., & Craske, M.G. (2012). Ran-
domized clinical trial of cognitive behavioral therapy (CBT) versus acceptance and
commitment therapy (ACT) for mixed anxiety disorders. Journal of Consulting and
Clinical Psychology, 80(5), 750765. http://dx.doi.org/10.1037/a0028310.
Balk, E.M., Earley, A., Patel, K., Trikalinos, T.A., & Dahabreh, I.J. (2012). AHRQ methods for
effective health care empirical assessment of within-arm correlation imputation in trials
of continuous outcomes. Rockville (MD): Agency for Healthcare Research and Quality
(US).
Barlow, D.H. (2000). Unraveling the mysteries of anxiety and its disorders from the per-
spective of emotion theory. American Psychologist, 55(11), 12471263. http://dx.doi.
org/10.1037/0003-066X.55.11.1247.
*Barlow, D.H., Allen, L.B., & Choate, M.L. (2004). Toward a unied treatment for emotional
disorders. Behavior Therapy, 35(2), 205230. http://dx.doi.org/10.1016/S0005-
7894(04)80036-4.
Beck, A.T. (1979). Cognitive therapy and the emotional disorders. New York: USA: Penguin
Books.
Beck, A.T., & Steer, R.A. (1996). Beck anxiety inventory. San Antonio, CA: The Psychological
Corporation.
Beck, A.T., Steer, R.A., & Brown, G.K. (1996). Beck depression inventory (2nd
edition)manual. San Antonio, TX: The Psychological Corporation.
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J.E., & Erbaugh, J.K. (1961). An inventory for
measuring depression. Archives of General Psychiatry, 4, 561571.
Birnbaum, H.G., Kessler, R.C., Kelley, D., Ben-Hamadi, R., Joish, V.N., & Greenberg, P.E.
(2010). Employer burden of mild, moderate, and severe major depressive disorder:
6
* = studies included in the meta-analysis.
109
Mental health services utilization and costs, and work performance. Depression and
Anxiety, 27(1), 7889. http://dx.doi.org/10.1002/da.20580.
Borenstein, M., Hedges, L.V., Higgins, J.P.T., & Rothstein, H.R. (2009). Introduction to meta-
analysis. Chichester, UK: Wiley.
Borenstein, M., & Higgins, J.T. (2013). Meta-analysis and subgroups. Prevention Science,
14(2), 134143. http://dx.doi.org/10.1007/s11121-013-0377-7.
*Bressi, C., Porcellana, M., Marinaccio, P.M., Nocito, E.P., & Magri, L. (2010). Short-term
psychodynamic psychotherapy versus treatment as usual for depressive and anxiety
disorders: A randomized clinical trial of efcacy. Journal of Nervous and Mental
Disease, 198(9), 647652. http://dx.doi.org/10.1097/NMD.0b013e3181ef3ebb.
Brown, T.A., Campbell, L.A., Lehman, C.L., Grisham, J.R., & Mancill, R.B. (2001a). Current
and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinical
sample. Journal of Abnormal Psychology, 110(4), 585599. http://dx.doi.org/10.1037/
0021-843x.110.4.585.
Brown, T.A., Di Nardo, P.A., Lehman, C.L., & Campbell, L.A. (2001b). Reliability of DSM-IV
anxiety and mood disorders: Implications for the classication of emotional disor-
ders. Journal of Abnormal Psychology, 110(1), 4958. http://dx.doi.org/10.1037/
0021-843X.110.1.49.
Brown, T.A., Dinardo, P.A., & Barlow, D. (1994). Anxiety disorders interview schedule for
DSM-IV. New York: Oxford University Press.
Chorpita, B.F., & Daleiden, E.L. (2009). Mapping evidence-based treatments for children
and adolescents: Application of the distillation and matching model to 615 treat-
ments from 322 randomized trials. Journal of Consulting and Clinical Psychology,
77(3), 566579. http://dx.doi.org/10.1037/a0014565.
Chorpita, B.F., Taylor, A.A., Francis, S.E., Moftt, C., & Austin, A.A. (2004). Efcacy of mod-
ular cognitive behavior therapy for childhood anxiety disorders. Behavior Therapy,
35(2), 263287. http://dx.doi.org/10.1016/s0005-7894(04)80039-x.
Clark, D.A., & Taylor, S. (2009). The transdiagnostic perspective on cognitive-behavioral
therapy for anxiety and depression: New wine for old wineskins? Journal of
Cognitive Psychotherapy, 23(1), 6066. http://dx.doi.org/10.1891/0889-8391.23.1.60.
Cohen, J. (1988). Statistical power analysis for behavioral sciences (2nd ed.). Hillsdale, NJ:
Erlbaum.
Craske, M.G. (2012). Transdiagnostic treatment for anxiety and depression. Depression
and Anxiety, 29(9), 749753. http://dx.doi.org/10.1002/da.21992.
*Craske, M.G., Farchione, T.J., Allen, L.B., Barrios, V., Stoyanova, M., & Rose, R. (2007). Cog-
nitive behavioral therapy for panic disorder and comorbidity: More of the same or
less of more? Behaviour Research and Therapy, 45(6), 10951109. http://dx.doi.org/
10.1016/j.brat.2006.09.006.
Cuijpers, P., Andersson, G., Donker, T., & Van Straten, A. (2011). Psychological treatment of
depression: Results of a series of meta-analyses. Nordic Journal of Psychiatry, 65(6),
354364. http://dx.doi.org/10.3109/08039488.2011.596570.
Cuijpers, P., Sijbrandij, M., Koole, S., Huibers, M., Berking, M., & Andersson, G. (2014). Psy-
chological treatment of generalized anxiety disorder: A meta-analysis. Clinical
Psychology Review, 34(2), 130140. http://dx.doi.org/10.1016/j.cpr.2014.01.002.
Cuthbert, B.N. (2014). The RDoC framework: Facilitating transition from ICD/DSM to di-
mensional approaches that integrate neuroscience and psychopathology. World
Psychiatry, 13(1), 2835. http://dx.doi.org/10.1002/wps.20087.
*Dear, B.F., Zou, J.B., Ali, et al. (2014). Clinical and cost-effectiveness of therapist-guided
internet-delivered cognitive behavior therapy for older adults with symptoms of anx-
iety: A randomized controlled trial. Behavior Therapy. http://dx.doi.org/10.1016/j.
beth.2014.09.007 (0).
Duval, S., & Tweedie, R. (2000a). A nonparametric trim and ll method of accounting for
publication bias in meta-analysis. Journal of the American Statistical Association,
95(449), 8998. http://dx.doi.org/10.1080/01621459.2000.10473905.
Duval, S., & Tweedie, R. (2000b). Trim and ll: A simple funnel-plot-based method of test-
ing and adjusting for publication bias in meta-analysis. Biometrics, 56(2), 455463.
http://dx.doi.org/10.2307/2676988.
Egger, M., Davey Smith, G., Schneider, M., & Minder, C. (1997). Bias in meta-analysis de-
tected by a simple, graphical test. BMJ, 315(7109), 629634.
*Ellard, K.K., Fairholme, C.P., Boisseau, C.L., Farchione, T.J., & Barlow, D.H. (2010). Unied
protocol for the transdiagnostic treatment of emotional disorders: Protocol develop-
ment and initial outcome data. Cognitive and Behavioral Practice, 17(1), 88101.
http://dx.doi.org/10.1016/j.cbpra.2009.06.002.
*Erickson, D.H., Janeck, A.S., & Tallman, K. (2007). A cognitive-behavioral group for pa-
tients with various anxiety disorders. Psychiatric Services, 58(9), 12051211.
Euroqol Group (1990). EuroQol A new facility for the measurement of health-related
quality of life. Health Policy, 16(3), 199208.
Fairburn, C.G., Cooper, Z., & Shafran, R. (2003). Cognitive behaviour therapy for eating dis-
orders: A transdiagnostic theory and treatment. Behaviour Research and Therapy,
41(5), 509528. http://dx.doi.org/10.1016/S0005-7967(02)00088-8.
*Farchione, T.J., Fairholme, C.P., Ellard, K.K., Boisseau, C.L., Thompson-Hollands, J., Carl, J.R.,
et al. (2012). Unied protocol for transdiagnostic treatment of emotional disorders: A
randomized controlled trial. Behavior Therapy, 43(3), 666678. http://dx.doi.org/10.
1016/j.beth.2012.01.001.
First, M.B., Spitzer, R.L., Gibbons, M., & Williams, J.B.W. (1996). Structured clinical interview
for DSM-IV axis I disorders (SCID-I). Washington, DC: American Psychiatric Press.
Harvey, A.G., Watkins, E., Mansell, W., & Shafran, R. (Eds.). (2004). Cognitive behavioural
processes across psychological disorders: A transdiagnostic approach to research and
treatment. Oxford: Oxford University Press.
Hayes, S.C., Strosahl, K.D., & Wilson, K.G. (1999). Acceptance and commitment therapy: An
experiential approach to behavior change. New York, NY, US: Guilford Press.
Higgins, J.P.T., & Green, S. (2011). Cochrane handbook for systematic reviews of interven-
tions. Retrieved from www.cochrane-handbook.org
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), 169183. http://dx.doi.org/10.1037/a0018555.
110 J.M. Newby et al. / Clinical Psychology Review 40 (2015) 91110
Hofmann, S.G., & Smits, J.A. (2008). Cognitive-behavioral therapy for adult anxiety disor-
ders: A meta-analysis of randomized placebo-controlled trials. Journal of Clinical
Psychiatry, 69(4), 621632. http://dx.doi.org/10.4088/JCP.v69n0415.
Hofmann, S.G., Wu, J.Q., & Boettcher, H. (2014). Effect of cognitive-behavioral therapy for
anxiety disorders on quality of life: A meta-analysis. Journal of Consulting and Clinical
Psychology, 82(3), 375391.
*Johnston, L., Titov, N., Andrews, G., Spence, J., & Dear, B.F. (2011). A RCT of a
transdiagnostic internet-delivered treatment for three anxiety disorders: Examina-
tion of support roles and disorder-specic outcomes. PLoS One, 6(11), e28079.
http://dx.doi.org/10.1371/journal.pone.0028079.
Judd, L.L. (1997). The clinical course of unipolar major depressive disorders. Archives of
General Psychiatry, 54(11), 989991.
*Kabat-Zinn, J., Massion, A.O., Kristeller, J., Peterson, L.G., Fletcher, K.E., Pbert, L., et al.
(1992). Effectiveness of a meditation-based stress reduction program in the treat-
ment of anxiety disorders. American Journal of Psychiatry, 149(7), 936943.
Katon, W.J. (2011). Epidemiology and treatment of depression in patients with chronic
medical illness. Dialogues in Clinical Neuroscience, 13(1), 723.
Kendler, K.S. (1996). Major depression and generalised anxiety disorder. Same genes,
(partly) different environmentsRevisited. British Journal of Psychiatry.
Supplement(30), 6875.
Kessler, R.C., Berglund, P., Demler, O., Jin, R., Merikangas, K.R., & Walters, E.E. (2005). Lifetime
prevalence and age-of-onset distributions of DSM-IV disorders in the national comor-
bidity survey replication. Archives of General Psychiatry, 62(6), 593602. http://dx.doi.
org/10.1001/archpsyc.62.6.593.
Knekt, P., Lindfors, O., Harkanen, T., Valikoski, M., Virtala, E., Laaksonen, M.A., et al. (2008).
Randomized trial on the effectiveness of long-and short-term psychodynamic psy-
chotherapy and solution-focused therapy on psychiatric symptoms during a 3-year
follow-up. Psychological Medicine, 38(5), 689703. http://dx.doi.org/10.1017/
s003329170700164x.
Kroenke, K., Spitzer, R., & Williams, J. (2001). The PHQ-9: Validity of a brief depression se-
verity measure. Journal of General Internal Medicine, 16(9), 606613. http://dx.doi.
org/10.1046/j.1525-1497.2001.016009606.x.
Leichsenring, F., & Salzer, S. (2014). A unied protocol for the transdiagnostic psychody-
namic treatment of anxiety disorders: An evidence-based approach. Psychotherapy
(Chicago, Ill.), 51(2), 224245. http://dx.doi.org/10.1037/a0033815.
Lovibond, P.F., & Lovibond, S.H. (1995). The structure of negative emotional states: Com-
parison of the depression anxiety stress scales (DASS) with the Beck depression and
anxiety inventories. Behaviour Research and Therapy, 33(3), 335343. http://dx.doi.
org/10.1016/0005-7967(94)00075-U.
*McEvoy, P.M., & Nathan, P. (2007). Effectiveness of cognitive behavior therapy for diag-
nostically heterogeneous groups: A benchmarking study. Journal of Consulting and
Clinical Psychology, 75(2), 344350. http://dx.doi.org/10.1037/0022-006X.75.2.344.
McEvoy, P.M., Nathan, P., & Norton, P.J. (2009). Efcacy of transdiagnostic treatments: A
review of published outcome studies and future research directions. Journal of
Cognitive Psychotherapy, 23(1), 2033. http://dx.doi.org/10.1891/0889-8391.23.1.20.
*Miller, J.J., Fletcher, K., & Kabat-Zinn, J. (1995). Three-year follow-up and clinical implica-
tions of a mindfulness meditation-based stress reduction intervention in the treat-
ment of anxiety disorders. General Hospital Psychiatry, 17(3), 192200. http://dx.
doi.org/10.1016/0163-8343(95)00025-m.
Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., & The, P.G. (2009). Preferred reporting
items for systematic reviews and meta-analyses: The PRISMA statement. PLoS
Medicine, 6(7), e1000097. http://dx.doi.org/10.1371/journal.pmed.1000097.
Nordgren, L.B., Hedman, E., Etienne, J., Bodin, J., Kadowaki, ., Eriksson, S., et al. (2014). Ef-
fectiveness and cost-effectiveness of individually tailored Internet-delivered cogni-
tive behavior therapy for anxiety disorders in a primary care population: A
randomized controlled trial. Behaviour Research and Therapy, 59(0), 111. http://dx.
doi.org/10.1016/j.brat.2014.05.007.
Norton, P.J. (2006). Toward a clinically-oriented model of anxiety disorders. Cognitive
Behaviour Therapy, 35(2), 88105. http://dx.doi.org/10.1080/16506070500441561.
*Norton, P.J. (2008). An open trial of a transdiagnostic cognitive-behavioral group therapy
for anxiety disorder. Behavior Therapy, 39(3), 242250. http://dx.doi.org/10.1016/j.
beth.2007.08.002.
*Norton, P.J. (2012). A randomized clinical trial of transdiagnostic cognitive-behavioral
treatments for anxiety disorder by comparison to relaxation training. Behavior
Therapy, 43(3), 506517. http://dx.doi.org/10.1016/j.beth.2010.08.011.
*Norton, P.J., & Barrera, T.L. (2012). Transdiagnostic versus diagnosis-specic CBT for anx-
iety disorders: A preliminary randomized controlled noninferiority trial. Depression
and Anxiety, 29(10), 874882. http://dx.doi.org/10.1002/da.21974.
Norton, P.J., & Philipp, L.M. (2008). Transdiagnostic approaches to the treatment of anxi-
ety disorders: A quantitative review. Psychotherapy: Theory, Research, Practice,
Training, 45(2), 214226. http://dx.doi.org/10.1037/0033-3204.45.2.214.
*Proudfoot, J., Ryden, C., Everitt, B., Shapiro, D.A., Goldberg, D., Mann, A., et al. (2004). Clin-
ical efcacy of computerised cognitive-behavioural therapy for anxiety and depres-
sion in primary care: Randomised controlled trial. The British Journal of Psychiatry,
185(1), 4654. http://dx.doi.org/10.1192/bjp.185.1.46.
Reinholt, N., & Krogh, J. (2014). Efcacy of transdiagnostic cognitive behaviour therapy for
anxiety disorders: A systematic review and meta-analysis of published outcome
studies. Cognitive Behaviour Therapy. http://dx.doi.org/10.1080/16506073.2014.
897367.
*Schmidt, N.B., Buckner, J.D., Pusser, A., Woolaway-Bickel, K., Preston, J.L., & Norr, A.
(2012). Randomized controlled trial of false safety behavior elimination therapy: A
unied cognitive behavioral treatment for anxiety psychopathology. Behavior
Therapy, 43(3), 518532. http://dx.doi.org/10.1016/j.beth.2012.02.004.
Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., et al. (1998).
The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and
validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10.
Journal of Clinical Psychiatry, 59(Suppl. 20), 2233.
Spitzer, R.L., Kroenke, K., Williams, J.B., & Lowe, B. (2006). A brief measure for assessing
generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10),
10921097.
*Titov, N., Andrews, G., Johnston, L., Robinson, E., & Spence, J. (2010). Transdiagnostic In-
ternet treatment for anxiety disorders: A randomized controlled trial. Behaviour
Research and Therapy, 48(9), 890899. http://dx.doi.org/10.1016/j.brat.2010.05.014.
*Titov, N., Dear, B.F., Schwencke, G., Andrews, G., Johnston, L., Craske, M.G., et al. (2011).
Transdiagnostic internet treatment for anxiety and depression: A randomised con-
trolled trial. Behaviour Research and Therapy, 49(8), 441452. http://dx.doi.org/10.
1016/j.brat.2011.03.007.
Ware, J., Jr., Kosinski, M., & Keller, S.D. (1996). A 12-item short-form health survey: con-
struction of scales and preliminary tests of reliability and validity. Medical Care, 34(3),
220233.
Watson, D. (2005). Rethinking the mood and anxiety disorders: A quantitative hierarchi-
cal model for DSM-V. Journal of Abnormal Psychology, 114(4), 522536.
Watts, S. E., Turnell, A., Kladnitski, N., Newby, J. M., & Andrews, G. (2015). Treatment-as-usual
(TAU) is anything but usual: A meta-analysis of CBT versus TAU for anxiety and depres-
sion. Journal of Affective Disorders, 175, 152167. http://dx.doi.org/10.1016/j.jad.2014.12.
025.
Wilamowska, Z.A., Thompson-Hollands, J., Fairholme, C.P., Ellard, K.K., Farchione, T.J., &
Barlow, D.H. (2010). Conceptual background, development, and preliminary data
from the unied protocol for transdiagnostic treatment of emotional disorders.
Depression and Anxiety, 27(10), 882890. http://dx.doi.org/10.1002/da.20735.
World Health Organization (1992). ICD-10 classication of mental and behavioural disor-
der: Clinical descriptions and diagnostic guidelines (Geneva).
World Health Organization (2008). The Global Burden of Disease: 2004 Update. Geneva,
Switzerland: WHO Press.