Cochrane (19)

Cochrane
Library
Cochrane Database of Systematic Reviews
Therapist-supported Internet cognitive behavioural therapy for

anxiety disorders in adults (Review)
Olthuis JV, Watt MC, Bailey K, Hayden JA, Stewart SH
Olthuis JV, Watt MC, Bailey K, Hayden JA, Stewart SH.

Therapist-supported Internet cognitive behavioural therapy for anxiety disorders in adults.
Cochrane Database of Systematic Reviews 2015, Issue 3. Art. No.: CD011565.
DOI: 10.1002/14651858.CD011565.
www.cochranelibrary.com
Therapist-supported Internet cognitive behavioural therapy for anxiety disorders in adults (Review)
Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Cochrane Trusted evidence.
Informed decisions.
Library Better health. Cochrane Database of Systematic Reviews
TABLE OF CONTENTS
HEADER......................................................................................................................................................................................................... 1
ABSTRACT..................................................................................................................................................................................................... 1
PLAIN LANGUAGE SUMMARY....................................................................................................................................................................... 2
SUMMARY OF FINDINGS.............................................................................................................................................................................. 4
BACKGROUND.............................................................................................................................................................................................. 11
OBJECTIVES.................................................................................................................................................................................................. 12
METHODS..................................................................................................................................................................................................... 12
RESULTS........................................................................................................................................................................................................ 18
Figure 1.................................................................................................................................................................................................. 19
Figure 2.................................................................................................................................................................................................. 22
Figure 3.................................................................................................................................................................................................. 23
Figure 4.................................................................................................................................................................................................. 27
Figure 5.................................................................................................................................................................................................. 29
DISCUSSION.................................................................................................................................................................................................. 30
AUTHORS' CONCLUSIONS........................................................................................................................................................................... 33
ACKNOWLEDGEMENTS................................................................................................................................................................................ 34
REFERENCES................................................................................................................................................................................................ 35
CHARACTERISTICS OF STUDIES.................................................................................................................................................................. 46
DATA AND ANALYSES.................................................................................................................................................................................... 135
Analysis 1.1. Comparison 1 Therapist-supported ICBT versus waiting list control, Outcome 1 Clinically Important Improvement 135
in Anxiety at Post-Treatment................................................................................................................................................................
Analysis 1.2. Comparison 1 Therapist-supported ICBT versus waiting list control, Outcome 2 Anxiety Symptom Severity at Post- 136
Treatment..............................................................................................................................................................................................
Analysis 1.3. Comparison 1 Therapist-supported ICBT versus waiting list control, Outcome 3 General Anxiety Symptom Severity 136
at Post-Treatment.................................................................................................................................................................................
Analysis 1.4. Comparison 1 Therapist-supported ICBT versus waiting list control, Outcome 4 Quality of Life at Post-Treatment..... 137
Analysis 2.1. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 1 Clinically Important Improvement in 138
Anxiety at Post-Treatment....................................................................................................................................................................
Analysis 2.2. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 2 Anxiety Symptom Severity at Post- 138
Treatment..............................................................................................................................................................................................
Analysis 2.3. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 3 Anxiety Symptom Severity at Follow- 139
up...........................................................................................................................................................................................................
Analysis 2.4. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 4 General Anxiety Symptom Severity at 139
Post-Treatment......................................................................................................................................................................................
Analysis 2.5. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 5 General Anxiety Symptom Severity at 139
Follow-up...............................................................................................................................................................................................
Analysis 2.6. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 6 Quality of Life at Post-Treatment......... 139
Analysis 2.7. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 7 Quality of Life at Follow-up.................. 140
Analysis 3.1. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 1 Clinically Important Improvement in 141
Anxiety at Post-Treatment....................................................................................................................................................................
Analysis 3.2. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 2 Clinically Important Improvement in 141
Anxiety at Follow-up.............................................................................................................................................................................
Analysis 3.3. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 3 Anxiety Symptom Severity at Post- 141
Treatment..............................................................................................................................................................................................
Analysis 3.4. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 4 Anxiety Symptom Severity at Follow- 142
Up...........................................................................................................................................................................................................
Analysis 3.5. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 5 General Anxiety Symptom Severity 142
at Post-Treatment.................................................................................................................................................................................
Analysis 3.6. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 6 General Anxiety Symptom Severity 142
at Follow-up..........................................................................................................................................................................................
Analysis 3.7. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 7 Quality of Life at Post-Treatment....... 143
Analysis 3.8. Comparison 3 Therapist-supported ICBT versus face-to-face CBT, Outcome 8 Quality of Life at Follow-up.............. 143
ADDITIONAL TABLES.................................................................................................................................................................................... 143
Therapist-supported Internet cognitive behavioural therapy for anxiety disorders in adults (Review) i
Informed decisions.
APPENDICES................................................................................................................................................................................................. 170
CONTRIBUTIONS OF AUTHORS................................................................................................................................................................... 170
DECLARATIONS OF INTEREST..................................................................................................................................................................... 170
SOURCES OF SUPPORT............................................................................................................................................................................... 170
DIFFERENCES BETWEEN PROTOCOL AND REVIEW.................................................................................................................................... 171
NOTES........................................................................................................................................................................................................... 171
INDEX TERMS............................................................................................................................................................................................... 172
Therapist-supported Internet cognitive behavioural therapy for anxiety disorders in adults (Review) ii
Informed decisions.
[Intervention Review]
Therapist-supported Internet cognitive behavioural therapy for anxiety

disorders in adults
Janine V Olthuis1, Margo C Watt2, Kristen Bailey3, Jill A Hayden4, Sherry H Stewart5
1Department of Psychology and Neuroscience, Dalhousie University, Halifax, Canada. 2Psychology, Saint Francis Xavier University,
Antigonish, Canada. 3Department of Psychology and Neuroscience, Dalhousie University & IWK Health Centre, Halifax, Canada.
4Department of Community Health & Epidemiology, Dalhousie University, Halifax, Canada. 5Departments of Psychiatry, Psychology and
Neuroscience, and Community Health and Epidemiology, Dalhousie University, Halifax, Canada
Contact address: Janine V Olthuis, Department of Psychology and Neuroscience, Dalhousie University, 1355 Oxford Street, Halifax, NS,
B3H 4J1, Canada. janine.olthuis@dal.ca.
Editorial group: Cochrane Common Mental Disorders Group

Publication status and date: New, published in Issue 3, 2015.
Citation: Olthuis JV, Watt MC, Bailey K, Hayden JA, Stewart SH. Therapist-supported Internet cognitive behavioural therapy for anxiety
disorders in adults. Cochrane Database of Systematic Reviews 2015, Issue 3. Art. No.: CD011565. DOI: 10.1002/14651858.CD011565.
ABSTRACT
Background
Cognitive behavioural therapy (CBT) is an evidence-based treatment for anxiety disorders. Many people have difficulty accessing
treatment, due to a variety of obstacles. Researchers have therefore explored the possibility of using the Internet to deliver CBT; it is
important to ensure the decision to promote such treatment is grounded in high quality evidence.
Objectives
To assess the effects of therapist-supported Internet CBT on remission of anxiety disorder diagnosis and reduction of anxiety symptoms in
adults as compared to waiting list control, unguided CBT, or face-to-face CBT. Effects of treatment on quality of life and patient satisfaction
with the intervention were also assessed.
Search methods
We searched the Cochrane Depression, Anxiety and Neurosis Review Group Specialized Register (CCDANCTR) to 12 April 2013. The
CCDANCTR includes relevant randomised controlled trials from EMBASE (1974 -), MEDLINE (1950 -) and PsycINFO (1967 -). We also searched
online clinical trial registries and reference lists of included studies. We contacted authors to locate further trials. An update of an
initial search (April 2013), conducted in September 2014, identified seven new completed studies, seven previously ongoing studies now
completed, and four new ongoing studies. This is a fast-moving area; we plan to update this review shortly, incorporating these new studies.
Selection criteria
Each identified study was independently assessed for inclusion by two authors. To be included, studies had to be randomised controlled
trials of therapist-supported ICBT compared to a waiting list, attention, information, or online discussion group; unguided CBT (that is, self-
help); or face-to-face CBT. We included studies that treated adults with an anxiety disorder (panic disorder, agoraphobia, social phobia,
post-traumatic stress disorder, acute stress disorder, generalized anxiety disorder, obsessive compulsive disorder, and specific phobia)
defined according to the Diagnostic and Statistical Manual of Mental Disorders III, III-R, IV, IV-TR or the International Classification of
Disesases 9 or 10.
Data collection and analysis

Two authors independently assessed the risk of bias of included studies and judged overall study quality. We used data from intention-
to-treat analyses wherever possible. We assessed treatment effect for the dichotomous outcome of clinically important improvement in
Therapist-supported Internet cognitive behavioural therapy for anxiety disorders in adults (Review) 1
Informed decisions.
anxiety using a risk ratio (RR) with 95% confidence interval (CI). For disorder-specific and general anxiety symptom measures and quality of
life we assessed continuous scores using standardized mean differences (SMD). We examined statistical heterogeneity using the I2 statistic.
Main results
We screened 1000 citations and selected 30 studies (2181 participants) for inclusion. The studies examined social phobia (11 trials),
panic disorder with or without agoraphobia (8 trials), generalized anxiety disorder (4 trials), post-traumatic stress disorder (1 trial), and
specific phobia (1 trial). Five remaining studies included a range of anxiety disorder diagnoses. Studies were conducted in Sweden (15
trials), Australia (12 trials), Switzerland (2 trials), and the Netherlands (1 trial) and investigated a variety of ICBT protocols. Three primary
comparisons were identified, experimental versus waiting list control, experimental versus unguided ICBT, and experimental versus face-
to-face CBT.
Moderate quality evidence from 9 studies (644 participants) contributed to a pooled RR of 4.18 (95% CI 2.42 to 7.22) for clinically important
improvement in anxiety at post-treatment, favouring therapist-supported ICBT over a waiting list, attention, information, or online
discussion group only. Similarly, the SMD for disorder-specific symptoms at post-treatment (22 studies, 1573 participants; SMD -1.12, 95%
CI -1.39 to -0.85) and general anxiety symptoms at post-treatment (14 studies, 1004 participants; SMD -0.79, 95% CI -1.10 to -0.48) favoured
therapist-supported ICBT. The quality of the evidence for both outcomes was low.
One study compared unguided CBT to therapist-supported ICBT for clinically important improvement in anxiety at post-treatment,
showing no difference in outcome between treatments (54 participants; very low quality evidence). At post-treatment there were no clear
differences between unguided CBT and therapist-supported ICBT for disorder-specific anxiety symptoms (4 studies, 253 participants; SMD
-0.24, 95% CI -0.69 to 0.21; low quality evidence) or general anxiety symptoms (two studies, 138 participants; SMD 0.28, 95% CI -2.21 to
2.78; low quality evidence).
Compared to face-to-face CBT, therapist-supported ICBT showed no significant differences in clinically important improvement in anxiety
at post-treatment (4 studies, 365 participants; RR 1.09, 95% CI 0.89 to 1.34; moderate quality evidence). There were also no clear differences
between face-to-face and therapist supported ICBT for disorder-specific anxiety symptoms at post-treatment (6 studies, 424 participants;
SMD 0.09, 95% CI -0.26 to 0.43; low quality evidence) or general anxiety symptoms at post-treatment (5 studies, 317 participants; SMD 0.17,
95% CI -0.35 to 0.69; low quality evidence).
Overall, risk of bias in included studies was low or unclear for most domains. However, due to the nature of psychosocial intervention
trials, blinding of participants and personnel, and outcome assessment tended to have a high risk of bias. Heterogeneity across a number
of the meta-analyses was substantial, some was explained by type of anxiety disorder or may be meta-analytic measurement artefact due
to combining many assessment measures. Adverse events were rarely reported.
Authors' conclusions
Therapist-supported ICBT appears to be an efficacious treatment for anxiety in adults. The evidence comparing therapist-supported ICBT to
waiting list, attention, information, or online discussion group only control was low to moderate quality, the evidence comparing therapist-
supported ICBT to unguided ICBT was low to very low quality, and comparisons of therapist-supported ICBT to face-to-face CBT was
low to moderate quality. Further research is needed to better define and measure any potential harms resulting from treatment. These
findings suggest that therapist-supported ICBT is more efficacious than a waiting list, attention, information, or online discussion group
only control, and that there may not be a significant difference in outcome between unguided CBT and therapist-supported ICBT; however,
this latter finding must be interpreted with caution due to imprecision. The evidence suggests that therapist-supported ICBT may not be
significantly different from face-to-face CBT in reducing anxiety. Future research should involve equivalence trials comparing ICBT and
face-to-face CBT, examine the importance of the role of the therapist in ICBT, and include effectiveness trials of ICBT in real-world settings.
A timely update to this review is needed given the fast pace of this area of research.
PLAIN LANGUAGE SUMMARY
Internet-based cognitive behavioural therapy with therapist support for anxiety in adults: a review of the evidence
Who may be interested in this review?
People who suffer from anxiety and their families.
General Practitioners.
Professionals working in psychological therapy services.
Developers of Internet-based therapies for mental health problems.
Why is this review important?
Many adults suffer from anxiety disorders, which have a significant impact on their everyday lives. Anxiety disorders often result in high
healthcare costs and high costs to society due to absence from work and reduced quality of life. Research has shown that cognitive
Informed decisions.
behavioural therapy (CBT) is an effective treatment which helps to reduce anxiety. However, many people are not able to access face-to-
face CBT due to long waiting lists, lack of available time for appointments, transportation problems, and limited numbers of qualified
therapists.
Internet-based CBT (ICBT) provides a possible solution to overcome many of the barriers to accessing face-to-face therapy. Therapists can
provide support to patients who are accessing Internet-based therapy by telephone or e-mail. It is hoped that this will provide a way of
increasing access to CBT, particularly for people who live in rural areas. It is not yet known whether ICBT with therapist support is effective
in reducing symptoms of anxiety.
What questions does this review aim to answer?
This review aims to summarise current research to find out whether ICBT with therapist support is an effective treatment for anxiety.
The review aims to answer the following questions:
- is ICBT with therapist support more effective than no treatment (waiting list)?
- how effective is ICBT with therapist support compared with face-to-face CBT?
- how effective is ICBT with therapist support compared with unguided CBT (self-help with no therapist input)?
- what is the quality of current research on ICBT with therapist support for anxiety?
Which studies were included in the review?
Databases were searched to find all high quality studies of ICBT with therapist support for anxiety published until May 2013. To be included
in the review, studies had to be randomised controlled trials involving adults over 18 years with a main diagnosis of an anxiety disorder;
30 studies with a total of 2181 participants were included in the review.
What does the evidence from the review tell us?
ICBT with therapist support was significantly more effective than no treatment (waiting list) at improving anxiety and reducing symptoms.
The quality of the evidence was low to moderate.
There was no significant difference in the effectiveness of ICBT with therapist support and unguided CBT, though the quality of the evidence
was low to very low. Patient satisfaction was generally reported to be higher with therapist-supported ICBT, however patient satisfaction
was not formally assessed.
ICBT with therapist support may not differ in effectiveness as compared to face-to-face CBT. The quality of the evidence was low to
moderate.
There was a low risk of bias in the included studies, except for blinding of participants, personnel, and outcome assessment. Adverse events
were rarely reported in the studies.
SUMMARY OF FINDINGS
Summary of findings for the main comparison. Therapist-supported ICBT compared to waiting list, attention, information, or online discussion
Library
Cochrane
group only control for anxiety disorders in adults
Therapist-Supported ICBT compared to waiting list, attention, information, or online discussion group only control for anxiety disorders in adults
Patient or population: patients with anxiety disorders

Settings: outpatient care via Internet with e-mail or telephone support, or both
Better health.
Informed decisions.
Trusted evidence.
Intervention: therapist-supported ICBT
Comparison: waiting list, attention, information, or online discussion group only control
Outcomes Illustrative comparative risks* (95% CI) Relative ef- No of partici- Quality of the Comments
fect pants evidence
Assumed risk Corresponding risk (95% CI) (studies) (GRADE)
Waiting list, Therapist-supported ICBT

attention, in-
formation,
or online
discussion
group only
controll
Clinically im- Study population RR 4.18 644 ⊕⊕⊕⊝

portant im- (2.42 to 7.22) (9 studies) moderate2
provement 13 per 100 54 per 100
in anxiety at (31 to 93)
post-treat-
ment Moderate
Indexed by a
standardized 8 per 100 33 per 100
interview or

(19 to 57)
clinically ac-
cepted measure
cut-off score1
Anxiety symp- The mean anxiety symptom severity at 1573 ⊕⊕⊝⊝ A standard deviation of 0.80 or
tom severity post-treatment in the intervention groups (24 studies) low3,4,5 greater represents a large difference
at post-treat- was between groups6
ment 1.12 standard deviations lower
Indexed by a (1.39 to 0.85 lower)
range of dis-
order-specific
4
self-report mea-
sures
Library
Cochrane
General anx- The mean general anxiety symptom sever- 1004 ⊕⊕⊝⊝ A standard deviation of 0.80 or
iety symp- ity at post-treatment in the intervention (14 studies) low4,5,7 greater represents a large difference
tom severity groups was between groups6
at post-treat- 0.79 standard deviations lower
ment (1.1 to 0.48 lower)
Indexed by a
range of mea-
Better health.
Informed decisions.
Trusted evidence.
sures of anxiety
symptoms in
general
Quality of life The mean quality of life at post-treatment 1395 ⊕⊕⊕⊝ A standard deviation of 0.50 rep-
at post-treat- in the intervention groups was (20 studies) moderate4,7 resents a moderate difference be-
ment 0.51 standard deviations higher tween groups6
Indexed by self- (0.4 to 0.61 higher)
report mea-
sures of quality
of life or func-
tional disability
Adverse events Study population Not estimable 0 See comment Because adverse events were so
at post-treat- (0) rarely reported, they could not be
ment See comment See comment meaningfully reported by compari-
not reported son and are instead described in the
Moderate review text
Participant Study population Not estimable 0 See comment Studies reported high overall treat-
satisfaction (13) ment satisfaction for therapist-sup-
Indexed by a See comment See comment ported ICBT

mix of qualita-
tive and quan- Moderate
titative self-re-
port measures
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is
based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence

High quality: Further research is very unlikely to change our confidence in the estimate of effect.
5
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.
Library
Cochrane
1 For clinically important improvement in anxiety, an event is indicative of a participant achieving clinically important improvement.
2 Downgraded for risk of bias (-1) primarily because three of the included studies did not blind their outcome assessors to participants' group assignment. Not downgraded for
inconsistency (0) because heterogeneity was reduced following subgroup analysis by anxiety disorder.
3 Downgraded for risk of bias (-1) primarily due to concerns with selective outcome reporting in a few studies.
4 Risk of bias (0). While participants in the included studies were not blind to their treatment condition when completing self-report measures and therapists were not blind to
Better health.
Informed decisions.
Trusted evidence.
the treatment they were delivering, these study characteristics cannot be avoided in this type of clinical treatment.
5 Downgraded for inconsistency (-1) because the heterogeneity amongst the included studies was quite high. This may be explained by the variety of anxiety disorders investigated
and differences in the treatment details; however, the number of studies that could be included in subgroup analyses was not sufficient to provide useful reasons for this
heterogeneity.
6 According to Cohen's (1969) interpretation of effect sizes.
7 Downgraded for risk of bias (-1) primarily because two studies included baseline imbalances in participant severity across study groups.
Summary of findings 2. Therapist-supported ICBT compared to unguided CBT for anxiety disorders in adults
Therapist-supported ICBT compared to unguided CBT for anxiety disorders in adults
Patient or population: patients with anxiety disorders

Comparison: unguided ICBT
fect pants evidence
Unguided Therapist-supported ICBT

ICBT

Clinically im- See comment See comment Not estimable 54 ⊕⊝⊝⊝ Not pooled because only one study
portant im- (1 study) very low2,3 in this comparison for this outcome
provement
in anxiety at
post-treat-
ment
Indexed by a
standardized
interview or
clinically ac-
cepted mea-
6
sure cut-off
score1
Library
Cochrane
Disorder-spe- The mean disorder-specific anxiety symp- 253 ⊕⊕⊝⊝ A standard deviation of 0.20 repre-
cific anxiety tom severity at post-treatment in the inter- (4 studies) low4,5,6 sents a small difference between
symptom vention groups was groups7
severity at 0.24 standard deviations lower
post-treat- (0.69 lower to 0.21 higher)
ment
Indexed by a
Better health.
Informed decisions.
Trusted evidence.
range of dis-
order-specif-
ic self-report
measures
General anx- The mean general anxiety symptom sever- 138 ⊕⊕⊝⊝
iety symp- ity at post-treatment in the intervention (2 studies) low3,4
tom severity groups was
at post-treat- 0.28 higher
ment (2.21 lower to 2.78 higher)
Indexed by a
range of mea-
sures of anxi-
ety symptoms
in general
Quality of life The mean quality of life at post-treatment in 199 ⊕⊕⊝⊝ A standard deviation of 0.10 repre-
at post-treat- the intervention groups was (3 studies) low4,5,6 sents a small difference between
ment 0.07 standard deviations higher groups7
Indexed by (0.37 lower to 0.5 higher)
self-report
measures of
quality of life
or functional

disability
Adverse Study population Not estimable 0 See comment Because adverse events were so
events at (0) rarely reported, they could not be
post-treat- See comment See comment meaningfully reported by compari-
ment son and are instead described in the
not reported Moderate review text
7
Participant See comment See comment Not estimable 0 See comment Studies generally reported higher
satisfaction (2 studies) satisfaction with therapist-support-
Indexed by a ed ICBT
Library
Cochrane
mix of qualita-
tive and quan-
titative self-
report mea-
sures
Better health.
Informed decisions.
Trusted evidence.

1 For clinically important improvement in anxiety, an event is indicative of a participant achieving clinically important improvement.
2 Downgraded for risk of bias (-1) primarily because of lack of blinding of outcome assessors.
3 Downgraded for imprecision (-2) as there is only one or two studies within the comparison for this outcome.
5 Downgraded for inconsistency (-1) as the heterogeneity amongst the included studies was quite high.
6 Downgraded for imprecision (-1) as there is a limited number of studies included in the comparison for this outcome.
Summary of findings 3. Therapist-supported ICBT compared to face-to-face CBT for anxiety disorders in adults
Therapist-supported ICBT compared to face-to-face CBT for anxiety disorders in adults

Patient or population: adults with anxiety disorders
Comparison: face-to-face CBT
fect pants evidence
8
Face-to-face Therapist-supported ICBT
CBT
Library
Cochrane
Clinically im- Study population RR 1.09 365 ⊕⊕⊕⊝
portant im- (0.89 to 1.34) (4 studies) moderate2
provement 41 per 100 44 per 100
in anxiety at (36 to 54)
post-treat-
ment Moderate
Better health.
Informed decisions.
Trusted evidence.
Indexed by a
standardized 45 per 100 49 per 100
interview or (40 to 61)
clinically ac-
cepted measure
cut-off score1
Anxiety symp- The mean anxiety symptom severity at post- 424 ⊕⊕⊝⊝ There was no significant differ-
tom severity treatment in the intervention groups was (6 studies) low3,4,5 ence between groups
at post-treat- 0.09 standard deviations higher
ment (0.26 lower to 0.43 higher)
Indexed by a
range of dis-
order-specific
self-report mea-
sures
General anx- The mean general anxiety symptom severity at 317 ⊕⊕⊝⊝ There was no significant differ-
iety symp- post-treatment in the intervention groups was (5 studies) low3,4,5 ence between groups
tom severity 0.17 standard deviations higher
at post-treat- (0.35 lower to 0.69 higher)
ment
Indexed by a
range of mea-
sures of anxiety

symptoms in
general
Quality of life The mean quality of life at post-treatment in 392 ⊕⊕⊕⊝ A standard deviation of 0.20
at post-treat- the intervention groups was (5 studies) moderate2,4 represents a small difference
ment 0.26 standard deviations higher between groups6
Indexed by self- (0.06 to 0.45 higher)
report mea-
sures of quality
of life or func-
tional disability
9
Adverse events See comment See comment Not estimable - See comment Because adverse events were
at post-treat- so rarely reported, they could
ment - not re- not be meaningfully reported
Library
Cochrane
ported by comparison and are instead
described in the review text
Participant Study population Not estimable 0 See comment Studies reported high overall
satisfaction (2) treatment satisfaction across
Indexed by a See comment See comment both conditions
Better health.
Informed decisions.
Trusted evidence.
mix of qualita-
tive and quan- Moderate
titative self-re-
port measures

1 For clinically important improvement in anxiety, an event is indicative of a participant achieving clinically important improvement at post-treatment.
2 Downgraded for imprecision (-1) primarily due to small sample size.
3 Downgraded for risk of bias (-1) primarily because one included study provided incomplete outcome data (though sensitivity analyses suggest no difference in findings when
this study is excluded).
5 Downgraded for inconsistency (-1) primarily due to unexplained heterogeneity.

10
Informed decisions.
BACKGROUND Unfortunately, certain barriers (for example, time constraints,

transportation problems, stigma, long waiting lists, a lack of
Description of the condition sufficiently qualified clinicians) continue to limit access to CBT
(Alvidrez 1999; Young 2001; Mohr 2006). Many of these barriers are
Individuals with anxiety disorders experience excessive anxiety
particularly relevant for those living in rural communities (Yuen
(fear or worry) which is disproportionate to actual threat or
1996; Rost 2002; Hauenstein 2006). National surveys in Canada
danger and significantly interferes with normal daily functioning.
(Statistics Canada 2004) and the US (Kessler 2004) suggested that
Anxiety disorders can include a range of physical (for example,
less than one third (only 32% and 20%, respectively) of those with
trembling, tense muscles, rapid breathing), cognitive (for example,
a current psychiatric disorder received some form of treatment
worries, difficulty concentrating), emotional (for example, distress,
in the past year. In a Canadian sample, only 11% of individuals
negative affect, irritability), and behavioural (for example, difficulty
with an anxiety disorder had received treatment (Ohayon 2000).
sleeping, hyperarousal) symptoms. Often those with anxiety
Increasingly, efforts are being made to improve access to CBT
disorders develop maladaptive strategies to lessen anxiety, such
on a large scale, particularly for those groups who are most at
as avoidance (Health Canada 2002; Wilson 2006) or substance use
risk due to lack of services (for example, the UK-based National
(Stewart 2008). Studies from Canada (Statistics Canada 2004), the
Health Service 'Improving Access to Psychological Therapies' (IAPT)
USA (Kessler 2005a), Australia (Slade 2007), Nigeria (Gureje 2006),
programme launched in 2006) (Department of Health 2008). A
and Europe (ESEMeD/MHEDEA 2000 Investigators 2004) suggest
distance delivery approach wherein CBT is delivered over the
that 6% to 18% of adults experience an anxiety disorder every year.
Internet with a therapist providing support by telephone or e-mail
Moreover, rates of remission within one year are low, that is, from
is one way to minimize treatment barriers and increase access to
33% to 42% across specific anxiety disorders (Robins 1991).
care while still delivering empirically-supported treatment. Such
There are many types of anxiety disorders, including panic disorder an approach could increase access to mental health professionals
(PD), agoraphobia, social phobia, post-traumatic stress disorder for those in rural areas, facilitate treatment for those of limited
(PTSD), acute stress disorder, generalized anxiety disorder (GAD), mobility, and increase patient confidentiality (that is, by engaging
obsessive compulsive disorder (OCD), and specific phobia. These in treatment from home clients do not 'risk' being seen at
are diagnosed according to criteria outlined in the Diagnostic and mental health clinics) and privacy (for example, a degree of visual
Statistical Manual of Mental Disorders (DSM IV-R) (APA 2000) or anonymity). The widespread availability of the Internet makes
the International Classification of Diseases (ICD 10) (WHO 1999). this type of intervention feasible and worth consideration. Recent
Anxiety disorders often co-occur with each other (Kessler 2005a) systematic reviews of computer- and Internet-based treatment
as well as with mood disorders (Fava 2000) and substance abuse for mental health problems suggest largely that these types of
or dependence (Stewart 2008). They tend to have an early onset treatment are more effective than a waiting list control and equally
(Kessler 2005b) and chronic course (Bruce 2005). Anxiety disorders effective as face-to-face psychotherapy in treating anxiety and
also have a major economic impact; for instance, costs of direct depression symptoms (Spek 2007; Bee 2008; Cuijpers 2009; Reger
treatment, unnecessary medical treatment, and work absences or 2009; Cuijpers 2010).
lost productivity amount to more than USD 40 billion per year in the
United States (DuPont 1996; Greenberg 1999). Studies have shown
How the intervention might work
significantly higher annual per capita medical costs for primary Therapist-supported ICBT should work to treat anxiety in the same
care patients with social phobia than for those with no mental manner as conventional face-to-face CBT. The underlying principles
health diagnosis (GBP 11,952 and EUR 2957 respectively) (Acarturk of CBT posit that psychopathology, or emotional disturbances,
2009); primary care patients with PD versus those with a chronic are the result of cognitive distortions and maladaptive behaviour.
somatic condition (EUR 10,269 versus EUR 3019) (Batelaan 2007); Whereas there are hypotheses about the relative importance of
and primary care patients with GAD as compared to those without cognitive and behavioural techniques, as well as suggestions
GAD (USD 2375 versus USD 1448) (Revicki 2012). that the strong collaborative working relationship between the
therapist and client are key to the success of CBT, the exact
Description of the intervention mechanisms of action in CBT are not yet well understood (Olatunji
Accumulating research supports the efficacy of CBT in the 2010). It is thought that disorder-specific symptoms develop
treatment of anxiety disorders (Bisson 2007; Hunot 2007; Norton as a result of a particular pattern of dysfunctional cognitions
2007; Stewart 2009) and anxiety symptoms (Deacon 2004). As its in combination with a specific set of behaviours that serve to
name suggests, CBT includes both cognitive as well as behavioural exacerbate these dysfunctional cognitions further (Beck 2005).
interventions or techniques. It has no one 'founder' and now As such, CBT works to improve symptoms by treating these
exists in many different forms. Its roots, however, lie largely in maladaptive cognitions and behaviours.
the work of Aaron Beck (Beck 1979). While pharmacotherapy
In essence, cognitive techniques and behaviour modification
(most commonly, benzodiazepines or selective serotonin reuptake
strategies are used to identify, evaluate, and challenge underlying
inhibitors) has been shown to be effective in the treatment of
maladaptive thoughts and beliefs. As an example, it is thought that
anxiety disorders, meta-analyses and review articles suggest that
catastrophic thoughts about the outcomes of experiencing arousal-
CBT is as effective in the acute phase of anxiety and may be
related physiological sensations, as well as inaccurate predictions
more effective than pharmacotherapy or a combination of both
about the probability of these dangerous outcomes, and avoidance
treatments in the long term (Westra 1998; Otto 2000; Otto 2005;
of situations that may induce these sensations contribute to
Pull 2007). Moreover, some anxiety medications pose significant
the development and maintenance of PD (Clark 1986; Barlow
risk for addiction (McNaughton 2008) or serious side effects, or both
1988). Accordingly, CBT for panic uses cognitive restructuring
(Buffett-Jerrot 2002; Choy 2007).
techniques to teach individuals to identify and challenge their
Informed decisions.
maladaptive cognitions and beliefs. This is combined with the contact. With a post-protocol change, they revised their review to
use of gradual, repeated exposure to feared sensations to help include studies that involved therapist contact with the qualifier
individuals revise their perceptions of threat and reduce their fear that the interventions must be able to be delivered stand-alone
of these arousal-related physiological sensations (Landon 2004). A without therapist contact. With this in mind, the focus of their
similar description of the CBT model could be provided for the other review remains largely on self-help therapies in which therapist
anxiety disorders (for example, social phobia) (Heimberg 2002). involvement is not necessary and treatment is largely client-
Whereas the underlying cognitive and behavioural principles are driven. Mayo-Wilson did not conduct analyses separating out those
evident in the CBT interventions for each of the anxiety disorders, interventions with and without therapist contact. As such, a meta-
current forms of CBT also target core components of a particular analysis with a particular emphasis on the efficacy of therapist-
disorder and, as such, specific models of CBT now exist for each supported ICBT is needed, particularly as at this point there
disorder, which modify and adapt CBT principles to fit disorder- remains conflicting evidence of the comparable efficacy of self-
specific symptoms (for example, specific phobia (Ost 1997); OCD help and therapist-supported interventions (for example, Spek
(Salkovskis 1985; Foa 2010); PD (Clark 1986; Casey 2004); social 2007; Titov 2008c; Berger 2011). The present review considered the
phobia (Heimberg 2002); GAD (Dugas 2007); PTSD (Ehlers 2000). specific efficacy of therapist-supported ICBT in comparison to each
of a waiting list control (that is, no treatment), traditional face-
ICBT therapists would be expected to draw on these models in to-face CBT, and self-help interventions and as such will fill a gap
the same manner as face-to-face CBT therapists. Typically, ICBT in the literature and answer current calls for research in this area
involves the client following a written treatment program available (Reger 2009). The protocol for the present review can be found in
on the Internet in conjunction with receiving therapist support, the Cochrane Library (Olthuis 2011).
either via telephone calls, texts, or e-mail (Andersson 2006). The
intervention involves content that mimics that of face-to-face CBT, OBJECTIVES
therapist-client contact (albeit through non-traditional means),
and the client engaging in further 'homework' outside of the To assess the effects of therapist-supported ICBT on remission of
session. As such, we anticipated that ICBT will work in the same way anxiety disorder diagnosis and reduction of anxiety symptoms in
and as well as traditional face-to-face CBT. adults as compared to waiting list control, unguided CBT, or face-
to-face CBT. Effects of treatment on quality of life and patient
Why it is important to do this review satisfaction with the intervention were also assessed.
Recently, research into ICBT has elicited considerable interest METHODS
from within the scientific and clinical communities. With advances
in modern communication technologies and their widespread Criteria for considering studies for this review
availability, this type of treatment is quickly becoming a more
realistic option. These advances have come at a time when long Types of studies
waiting lists and a lack of treatment availability stand in stark We included parallel group randomised controlled trials (RCTs),
contrast to the growing emphasis on the importance of mental cross-over, and cluster randomised trials.
health and provision of evidence-based treatments. A desire to
pursue Internet treatment as a viable option to increase access to Types of participants
treatment is growing. The importance of ensuring that the decision
to promote such treatment is grounded firmly in high quality Participant characteristics
evidence is therefore paramount. We included studies of adults (over 18 years of age; no upper limit).
The present review asked whether therapist-supported ICBT is Diagnosis
efficacious in treating anxiety, and if it is as efficacious as face-
to-face CBT. Past meta-analyses have reviewed the efficacy of Participants with a primary diagnosis of an anxiety disorder
ICBT for anxiety symptoms (Spek 2007). A number of reviews that according to the DSM-III (APA 1980), DSM-III-R (APA 1987), DSM-
have included ICBT have looked more broadly, however, at health IV (APA 1994), DSM-IV-TR (APA 2000), ICD-9 (WHO 1979) or ICD-10
problems in general (Barak 2008; Bee 2008) or all computer-based (WHO 1999) diagnostic criteria.
interventions (Cuijpers 2009; Reger 2009; Andrews 2010). Moreover,
We included studies that focused on or adequately reported
many of these reviews have not focused on the role of therapist
subgroup information for any of the following anxiety disorders:
involvement (for example, Cuijpers 2009; Reger 2009; Andrews
panic disorder (PD) with or without agoraphobia, agoraphobia
2010). Ultimately, as the field of ICBT is growing quickly, an updated
without a history of panic, social phobia (social anxiety disorder),
review on therapist-supported ICBT is needed. The findings of this
post-traumatic stress disorder (PTSD), acute stress disorder,
review will be helpful in guiding the path of future research in this
obsessive compulsive disorder (OCD), specific phobia, generalized
field away from continued replication of established findings and
anxiety disorder (GAD), and anxiety disorder not otherwise
toward addressing gaps in the literature and considering the next
specified. Included studies used diagnoses determined using
steps in ICBT implementation.
a validated diagnostic instrument, for example, the Structured
There is a Cochrane Review on media-delivered CBT and Clinical Interview for DSM-IV-TR Axis I Disorders (SCID-I) (First 2002).
behavioural therapy (BT) (self-help) for anxiety disorders (Mayo-
Setting
Wilson 2013). Mayo-Wilson's review answers questions about the
efficacy of delivering CBT to clients in non-traditional formats, We included studies in which treatment entailed participants
including via the Internet. In the protocol of their review, Mayo- engaging in the treatment from their homes and therapists located
Wilson specified that they would not include studies with therapist at primary care settings, university laboratories, community mental
Informed decisions.
health clinics, or private practice clinics. Participants could be Comparator interventions

treatment-seeking community members responding to media
1. Waiting list, attention, information, or online discussion group
advertisements for study participation or they could be referred to
only control condition (no intervention for participants beyond
the study by a health professional.
weekly status monitoring by research personnel or accessing
Co-morbidities online non-treatment related disease information or discussion
groups)
We included studies of participants with co-morbid diagnoses (for 2. Unguided CBT (i.e., self-help CBT with no therapist support)
example, major depressive disorder, substance abuse) only if they
3. Conventional face-to-face CBT interventions (including
had been diagnosed with a primary anxiety disorder. We did not
individual or group CBT delivered in a traditional face-to-face
include studies of participants reporting anxiety symptoms that did
format)
not meet criteria for an anxiety disorder (for example, participants
with a clinical presentation of major depressive disorder who Types of outcome measures
reported subthreshold anxiety symptoms or participants scoring
high on measures of anxiety symptoms but who were not assessed Primary outcomes
for a DSM diagnosis).
1. Efficacy of therapist-supported ICBT in leading to clinically
Types of interventions important improvement in anxiety as determined by a
diagnostic interview, for example, the SCID-I (First 2002) or
Experimental interventions the Anxiety Disorders Interview Schedule (ADIS-IV) (DiNardo
Cognitive behavioural therapies 1994) or a defined cut-off on a validated scale, for example,
the Yale Brown Obsessive Compulsive Scale (YBOCS) (Goodman
We included studies that investigated the efficacy of a 1989). In case the Clinical Global Impression scale change or
therapist-supported Internet cognitive behavioural therapy (CBT), improvement items (CGI) (Guy 1976) were used, we employed
behavioural therapy (BT), or cognitive therapy (CT) intervention for a score of 1 = 'very much' or 2 = 'much improved' to indicate
anxiety, defined as the following. clinically important improvement.
2. Efficacy of therapist-supported ICBT in leading to reduction in
• BT interventions must have been designed to change the
anxiety symptom severity measured by scores on a validated,
behaviours that result from maladaptive anxiety-related
observer-rated instrument, for example, the Hamilton Anxiety
cognitions (we included interventions including, but not limited
Rating Scale (Hamilton 1959), or a validated self-report measure
to, exposure, desensitization, and behavioural experiments).
of: (a) disorder-specific symptoms, for example, the Social
• CT must have been focused on elements of cognitive Phobia Scale (SPS) (Mattick 1998), and (b) anxiety symptoms
restructuring of irrational or maladaptive anxiety-related in general, for example, the Beck Anxiety Inventory (BAI) (Beck
cognitions. 1991).
• CBT interventions consisted of some combination of the
elements of CT and BT. Secondary outcomes
Whereas psychoeducation often is an important part of CBT, 1. Quality of life as assessed by either measures of quality of life,
we did not consider psychoeducation alone to be a sufficient for example, the Quality of Life Inventory (QOLI) (Frisch 1992), or
CBT intervention unless it included some of the other treatment measures of disability, for example the Sheehan Disability Scales
components described here. (SDS) (Leon 1997) as increasing disability entails decreased
quality of life. While research suggests that quality of life and
Internet interventions disability are distinct but somewhat overlapping constructs
(Hambrick 2003), quality of life measures have not often been
To be considered an Internet intervention, CBT must have been
conceptually or operationally distinguished from measures of
delivered over the Internet through the use of web pages or e-
disability, resulting in considerable overlap amongst indices
mail, or both. Crucially, Internet interventions must have included
of quality of life and disability (Mogotsi 2000). With this in
therapist support but this interaction could not be face-to-face.
mind, we anticipated an overlapping conceptualization of these
However, we included interventions that involved an initial face-
two constructs in the included studies and included both
to-face intake or interview session or an initial session to orient
types of measures within the meta-analysis in order to capture
clients to the Internet delivery method or to engage in treatment
all possible information about treatment outcome related to
planning, or a combination of these. Thus, therapist support must
quality of life.
have occurred via e-mail or the telephone, or both. Including only
interventions that could be delivered entirely by distance methods 2. Participant satisfaction with the intervention. Participant
reflected a primary motive for conducting this review, to find satisfaction tends to be measured uniquely across different
ways to increase access to treatment for those who may not be studies using a mix of qualitative and quantitative indices. In
able to visit provider centres. While it was possible that Internet- anticipation of this, we evaluated participants' satisfaction with
based interventions that provided some support in a face-to-face the intervention of interest as compared to the comparator
setting could be just as effectively restructured to be delivered interventions in a qualitative manner.
completely by distance, it was more rigorous to include only studies 3. Adverse events, in whatever manner reported by study authors.
that provided evidence specifically on the efficacy of Internet CBT
Timing of outcome assessment
delivered completely via distance methods. We did not select
interventions based on their length, or the number or duration of We performed separate analyses based on different periods of
sessions. assessment: immediately post-treatment and at one follow-up
Informed decisions.
period at least six months post-treatment but not more than one Details of CCDAN's generic search strategies can be found on the
year. When studies reported more than one follow-up assessment Group‘s website.
point, we used the longest follow-up period so as to provide the
best estimate of the long-term outcomes of the intervention. We searched the CCDAN Specialised Registers to 12 April 2013
and the results from this search were fully incorporated in the
Hierarchy of outcome measures present review. Additionally, prior to publication, CCDAN's Trials
For primary outcomes, separate meta-analyses were conducted Search Co-ordinator performed a precise update search of the
for the two outcomes. The clinically important improvement CCDANCTR Registers in September 2014 (Appendix 1). The results
in anxiety outcome measures were selected according to the were screened at the CCDAN's editorial base and by the first
following hierarchy, based on availability in a particular study: author and relevant studies were placed in awaiting classification
(1) diagnostic interview, (2) cut-off on a validated scale, (3) CGI or ongoing (as appropriate). These studies will be fully incorporated
scores. For reduction in anxiety symptom severity, the outcomes of in a timely, future update of this review.
available observer-rated and self-report measures were statistically CCDANCTR-Studies
combined and a mean score was created across the measures
within a particular study. Measures of variance for this mean score We searched the CCDANCTR-Studies Register using the following
were created by combining standard deviations across studies search strategy:
according to the method described by Borenstein 2009. This
method requires that the correlation between two measures be 1. Condition = (anxiety or *phobi* or PTSD or post-trauma* or “post
known; as such, in the case that this correlation was not known, the trauma*” or posttrauma* or "stress disorder" or panic or OCD or
measures with better psychometric properties were included in the obsess* or compulsi* or GAD)
analysis. 2. Intervention = (CBT or cognitive or behavio* or *therap* or
treatment or intervention or training or counsel*)
For secondary outcomes, quality of life outcome measures were 3. Age Group = (adult or aged or unclear or “not stated”)
treated in the same way as anxiety symptom severity measures. 4. Free-Text = (computer* or distance* or remote or tele* or Internet*
Due to the qualitative nature of the other secondary outcome, or web* or WWW or phone or mobile or e-mail* or email* or online*
participant satisfaction with the intervention, a hierarchy of or on-line or videoconferenc* or video-conferenc* or "chat room*"
outcome measures was not required. or "instant messaging" or iCBT)
5. (1 and 2 and 3 and 4)
Search methods for identification of studies
CCDANCTR-References
We used several methods to identify both published and
unpublished studies for possible inclusion in this review (see We searched the CCDANCTR-References Register to identify
below). We did not restrict studies to those reported in any additional untagged or uncoded references using the following
particular language; however, we conducted searches in English strategy:
and initiated contact with authors in English.
1. (anxiety or *phobi* or PTSD or post-trauma* or “post trauma*” or
Electronic searches posttrauma* or (stress and disorder*) or panic or OCD or obsess* or
compulsi* or GAD):ti,ab,kw
The Cochrane, Depression, Anxiety and Neurosis Review Group's
2. (therap* or train*):ti,ab
Specialised Register (CCDANCTR)
3. (psychotherap* or cognitive or behavio* or CBT):ti,ab,kw
The Cochrane Depression, Anxiety and Neurosis Group (CCDAN) 4. (acceptance* or assertive* or brief* or commitment* or exposure
maintains two clinical trials registers at their editorial base in or group or implosive or “problem solving” or problem-solving or
Bristol, UK, a references register and a studies-based register. "solution focused" or solution-focused or schema):ti,ab,kw
The CCDANCTR-References Register contains over 37,000 reports 5. (CBT or cognitive or behavio* or “contingency management”
of RCTs in depression, anxiety, and neurosis. Approximately 60% or “functional analys*” or mindfulness* or “mind training” or
of these references have been tagged to individual, coded trials. psychoeducat* or relaxation or “role play*”):ti,ab,kw
The coded trials are held in the CCDANCTR-Studies Register and 6. ((2 or 3) and 4) or 5
records are linked between the two registers through the use 7. (computer* or distance* or remote or tele* or Internet* or web* or
of unique Study ID tags. Coding of trials is based on the EU- WWW or phone or mobile or e-mail* or email* or online* or on-line
Psi coding manual. Please contact the CCDAN Trials Search Co- or videoconferenc* or video-conferenc* or "chat room*" or "instant
ordinator for further details. Reports of trials for inclusion in messaging" or iCBT):ti,ab,kw
the Group's registers are collated from routine (weekly), generic 8. 1 and 6 and 7
searches of MEDLINE (1950 to date), EMBASE (1974 to date) and
PsycINFO (1967 to date); quarterly searches of the Cochrane Searching other resources
Central Register of Controlled Trials (CENTRAL), and review- Reference lists
specific searches of additional databases. Reports of trials are
also sourced from international trial registers via the World Health We examined the reference lists of previous related meta-analyses
Organisation (WHO International Clinical Trials Registry Platform (Spek 2007; Bee 2008; Cuijpers 2009; Reger 2009; Andrews 2010;
(ICTRP), ClinicalTrials.gov, drug companies, and the handsearching Cuijpers 2010) and of articles selected for inclusion in the present
of key journals, conference proceedings, and other (non-Cochrane) review.
systematic reviews and meta-analyses.
Informed decisions.
Personal contacts and correspondence 5. outcome measures employed, as listed in Types of outcome
measures, as well as the dropout rates for participants in each
We contacted experts in the field, including principal authors of
treatment condition and whether the data reflected intention-
RCTs in the field of ICBT for anxiety, via e-mail and asked them
to-treat (ITT) analyses with last observation carried forward
if they were aware of any further studies which meet the present
(LOCF) or another method.
review’s inclusion criteria.
We subsequently recorded data in RevMan 5.3 data tables (RevMan
Unpublished studies
2014).
In order to search for unpublished studies, we searched
international trial registries including via the WHO Main planned comparisons
ICTRP (http://apps.who.int/trialsearch/) and ClinicalTrials.gov We planned to compare each of the outcomes of interest, at post-
(www.clinicaltrials.gov) in June 2013. treatment and 6 to 12 month follow-up, for each of the following
comparisons:
Data collection and analysis
Selection of studies 1. therapist-supported ICBT versus waiting list, attention,
information, or online discussion group only control,
In collaboration with the CCDAN Trials Search Co-ordinator, one 2. therapist-supported ICBT versus unguided CBT, and
review author (JVO) conducted searches of electronic databases
3. therapist-supported ICBT versus face-to-face CBT.
and reference lists and contacted authors in order to locate
potential trials to be included in the review. Two review authors Assessment of risk of bias in included studies
(JVO and KMB) independently assessed the titles and abstracts
of the resulting lists of studies for relevance. We then obtained We assessed the risk of bias in each included study using the
full articles for potentially relevant abstracts. Both review authors Cochrane Collaboration’s 'risk of bias' tool (Higgins 2011a). We
independently assessed the identified trials to determine eligibility assessed the following six areas for risk of bias.
as outlined in Criteria for considering studies for this review. We
1. Sequence generation: was the allocation sequence of
collated and compared assessments. In the case of disagreement
participants adequately randomised?
with respect to trial eligibility, we made the final decision by
discussion and consensus, if necessary with the involvement of 2. Allocation concealment: was the allocation sequence
another member of the review group (MCW or SHS, or both). adequately concealed from participants as well as those
involved in the enrolment and assignment of participants?
Data extraction and management 3. Blinding: were participants, study personnel, and those
We independently extracted data from the included studies assessing outcomes kept unaware of participants’ allocation to
regarding methodology and treatment outcomes, and recorded the a study condition throughout the course of the investigation?
data using a data extraction spreadsheet designed by one of the 4. Incomplete outcome data: were there incomplete data for the
review authors (JVO). If the included trials did not provide complete main or secondary outcomes (e.g., due to attrition)? Were
information (for example, details of dropout, group means and incomplete data adequately addressed?
standard deviations), we contacted the primary investigator by e- 5. Selective reporting: was the study free of suggestions of
mail to attempt to obtain unreported data to permit an intention- selective reporting of outcomes (e.g., reporting of a subset of
to-treat (ITT) analysis. We contacted other investigators as needed. outcomes on the basis of the results)?
6. Other potential threats to bias: was the study free of any other
Two review authors (JVO and KMB) independently extracted the problems (e.g., early stopping, baseline imbalance, cross-over
following data from each trial report: trials) that could have introduced bias?
1. description of trial, including primary researcher and year of We did not assess risk of bias related to therapist experience and
publication; qualifications. Evidence in the field as to the impact of therapist
2. characteristics of trial methodology, including the diagnostic experience on treatment outcomes remains mixed (for example,
criteria employed, participant inclusion and exclusion criteria, Hahlweg 2001; Andersson 2012; Norton 2014), as such, it would be
the screening instrument(s) used, the inclusion or exclusion of inappropriate to impose bias on a study based on a characteristic
co-morbidity, the receipt of other interventions simultaneously, we are unsure would actually introduce bias. In addition, we did not
and the number of centres involved; assess risk of bias related to therapist allegiance. This was because:
3. characteristics of participants, including age, gender, primary (a) all studies investigated CBT, and (b) it was impossible to know if
diagnosis, any co-morbid diagnoses, and duration of primary researchers were allied with a particular type of delivery method.
symptoms;
4. characteristics of the intervention (for both the experimental Two review authors (JVO and KMB) independently assessed risk
and comparator interventions), including intervention of bias for each included study. We resolved disagreements by
classification (i.e., CBT, BT, CT), content and components consensus and discussion with a third review author (MCW or SHS)
(e.g., psychoeducation, relaxation training, exposure, cognitive where necessary. If further information about a particular trial
restructuring), method of delivery of therapist support was required to assess its risk of bias, we contacted the primary
(e.g., telephone, e-mail), duration, amount of therapist and investigator of the relevant study. We created 'risk of bias' tables
experimenter contact, and number of participants randomised describing the information outlined above, as reported in each
to each intervention; and study. These tables also include a judgement on the risk of bias,
made by the review authors for each of the six areas, based on the
Informed decisions.
following three categories: (1) low risk of bias, (2) high risk of bias, analysis presented log-transformed data, we back-transformed
and (3) unclear or unknown risk of bias. these data and included untransformed data in the meta-analysis.
We then conducted a sensitivity analysis excluding any studies that
Measures of treatment effect presented transformed data.
Dichotomous outcomes
Unit of analysis issues
We analysed our only dichotomous outcome, clinically important
Parallel group randomised controlled trials (RCTs)
improvement in anxiety (yes or no) (as measured by no longer
meeting diagnostic criteria on a diagnostic interview, no longer In some parallel group RCTs, participants randomly assigned to a
meeting a designated cut-off on a validated scale, or meeting the waiting list, attention, information, or online discussion group only
criteria for very much or much improved on the CGI) using risk ratios control were permitted to pursue the active treatment after their
(RRs) and 95% confidence intervals (CIs) within studies. period on the waiting list was complete. To analyse dichotomous
and continuous data for these trials, we only included data from
Continuous outcomes participants before they crossed over to their second treatment
As most studies that were selected for inclusion used different condition; in other words, only data from the original comparison
measures to assess sufficiently similar constructs, we compared (waiting list, attention, information, or online discussion group only
continuous outcomes (that is, general and disorder-specific anxiety control versus therapist-supported ICBT) was used in the meta-
symptoms, quality of life) by calculating the standardized mean analyses.
difference (SMD) and its 95% CI. However, when all of the studies
Cross-over trials
within a meta-analysis used the same measure to assess an
outcome (for example, if all studies within a meta-analysis used the When included studies were cross-over trials, we planned to
BAI to assess general anxiety symptoms), we compared continuous include only data from the first phase of the trial.
outcomes by calculating the mean difference (MD) to facilitate the
interpretation of the clinical relevance of the findings. Cluster randomised trials
When cluster randomised trials had accounted for clustering
Most included studies used more than one measure to assess each
within their analyses (through the use of multilevel modelling
of the continuous outcomes. Thus, a mean score was created across
or general estimating equations, for example) we planned to
the measures included within each study. Measures of variance for
include data directly in the meta-analyses. For studies that failed
this mean score were created by combining standard deviations
to appropriately account for clustering, we planned to impute the
across studies according to the method described by Borenstein
data based on the number of clusters reported in each intervention
2009. This method requires that the correlation between two
group, the size of each cluster, summary statistics, and an estimate
measures be known; as such, on the rare occasion when this
of intracluster correlation. We also planned to exclude cluster trials
correlation was not known and could not be identified in prior
with a high risk of bias (that is, where clustering was not accounted
literature the measure in question was excluded from analyses.
for in analyses) from sensitivity analyses.
This occurred in four instances (Klein 2006, Richards 2006, and
Kiropoulos 2008 for the Body Vigilance Scale; Andersson 2009 for Multiple intervention arms
the Fear Survey Schedule III).
When multiple intervention arms met our inclusion criteria,
To combine measures of quality of life and disability into one we planned to combine eligible groups to create a pair-wise
outcome, we reversed the scores of the disability measures (that is, comparison following the procedure outlined in the Cochrane
by subtracting mean scores from the measure total scores) to align Handbook for Systematic Reviews of Interventions (Higgins 2011a).
them with the quality of life measures. We planned to conduct sensitivity analyses excluding any studies
with multiple intervention arms that did not report all intervention
Endpoint versus change data comparisons.
We anticipated that we might encounter some studies that
Dealing with missing data
reported analyses based on changes from baseline and other
studies that reported analyses based on final values. We planned We used data from ITT analyses whenever they were reported by
to present the two types of analysis results in separate subgroups study authors. In 21 studies, authors employed a LOCF method to
to avoid confusion for readers and, where appropriate, to combine address missing data with the assumption that participants who
both types of scores in the final results. Despite these plans, none were missing data following randomisation (that is, dropouts) did
of the included studies reported change data so we used endpoint not respond to treatment. Of the remaining studies, one study used
data in all meta-analyses. multiple imputation methods to create ITT data (van Ballegooijen
2013). Seven studies used a mixed effects models approach in an
Skewed data ITT approach to deal with missing data (Bergstrom 2010; Berger
We dealt with skewed data according to the guidelines in 2011; Hedman 2011; Paxling 2011; Andersson 2012a; Andersson
the Cochrane Handbook for Systematic Reviews of Interventions 2012b; Silfvernagel 2012). One study did not include ITT data
(Higgins 2011a) and Higgins 2008. In order to conduct the final (Andersson 2009).
analysis, transformed or untransformed data had to be obtained
Because included studies did not report individual participant data,
for all studies because log-transformed and untransformed data
if authors did not provide ITT analyses in their manuscript we
cannot be combined in meta-analyses (Higgins 2011a). In the
contacted the primary investigator by e-mail to attempt to obtain
case that a limited number of studies included in one meta-
unreported data to permit an ITT analysis. When we did not receive
Informed decisions.
responses from study authors we simply included their reported, mean difference (SMD) and 95% CI. However, as indicated
non-ITT, continuous outcome data in the analysis. This was the case previously, when outcomes were measured similarly across studies
for one study (Andersson 2009). For dichotomous outcomes, we we used a mean difference method. We used the RevMan 5.3
were able to impute ITT data by assuming that participants who software for data synthesis.
had dropped out did not meet the target event (that is, clinically
important improvement in anxiety). We conducted sensitivity Subgroup analysis and investigation of heterogeneity
analyses excluding studies for which ITT data were not available We conducted subgroup analyses but interpreted these with
(either from the published manuscript or from study authors) to caution due to the risk of false positive conclusions. We planned to
determine the extent to which missing data influenced effect sizes. perform the following subgroup analyses:
If included trials did not provide complete information (that is, 1. gender of participants;
group means, standard deviations, and sample size), we contacted
2. type of anxiety disorder (i.e., PD with or without agoraphobia,
the primary investigator by e-mail to attempt to obtain unreported
agoraphobia without a history of panic, social phobia (social
data. We contacted other study investigators as needed. The only
anxiety disorder), PTSD, acute stress disorder, OCD, specific
sources for outcome data were the original published report or
phobia, GAD, and anxiety disorder not otherwise specified);
author correspondence. If standard deviations were not available
from the authors, we planned to calculate these using other data 3. amount of therapist contact, designated as low (90 min or less),
reported in the article, including t-values, CIs, and standard errors. medium (91 to 299 min), or high (300 min or more);
If that was not possible, we planned to impute standard deviations 4. type of CBT (i.e., BT, CT, or CBT); and
from other investigations using similar measures and populations. 5. research group (i.e., the laboratory from which the study was
generated).
Assessment of heterogeneity
We were not able to conduct a subgroup analysis based on gender
We tested the extent of statistical heterogeneity in meta-analyses
of participants as none of the included studies distinguished
using the I2 statistic (Higgins 2002), which calculates the percentage
outcomes based on this participant variable. We also were not able
of variability due to heterogeneity rather than chance. According
to conduct a subgroup analysis based on type of CBT. Only one
to the guidelines outlined in the Cochrane Handbook for Systematic
study (Andersson 2009) had a stronger focus on BT, as compared to
Reviews of Interventions, I2 values may be interpreted as follows: CT or CBT, and no studies examined a CT only intervention. For the
final subgroup analysis by research group, three research groups
• 0% to 40% might not be important;
were identified: a group in Sweden, and two distinct groups in
• 30% to 60% may represent moderate heterogeneity; Australia.
• 50% to 90% may represent substantial heterogeneity; and
• 75% to 100% represents considerable heterogeneity (Higgins Sensitivity analysis
2011a). We conducted sensitivity analyses to determine the extent to which
observed pooled effect sizes depend on the quality of the design
We interpreted the importance of these I2 values in consideration of
characteristics of studies. We planned to conduct the following
the magnitude and direction of effects and the strength of evidence
sensitivity analyses:
for heterogeneity (as indexed by the P value from the Chi2 test). If
there was evidence of heterogeneity, we first re-checked the data 1. exclusion of studies with a designation of high risk of bias for one
for accuracy. We considered sources of heterogeneity according or more of the categories as outlined in Assessment of risk of bias
to the pre-specified subgroup and sensitivity analyses listed in in included studies;
Subgroup analysis and investigation of heterogeneity. 2. exclusion of cluster randomised trials where clustering was not
appropriately accounted for in analysis;
Assessment of reporting biases
3. exclusion of studies with multiple intervention arms with
Where there were sufficient numbers of trials to make such a plot selective reporting of intervention comparisons;
meaningful (that is, at least 10 included studies (Higgins 2011a)) 4. exclusion of studies with a somewhat more active waiting
we constructed funnel plots to determine the possible influence of list control condition (i.e., attention, information, or online
publication bias. We planned to enhance funnel plots with contour discussion group only control)
lines delineating areas of statistical significance (as suggested by
5. exclusion of studies with imputed standard deviations for
Peters 2008) to assist in the differentiation of asymmetry due to
continuous outcomes;
publication bias or other causes.
6. exclusion of studies with back transformed data for continuous
Data synthesis outcomes;
We combined data using an inverse-variance random-effects 7. exclusion of studies not reporting: (a) dichotomous, and (b)
model due to expected variation in the characteristics of continuous outcomes according to the ITT principle;
the interventions investigated and participant populations. We 8. exclusion of studies with continuous outcomes analysed using
combined dichotomous outcome measures by computing a LOCF; and
pooled risk ratio (RR) and 95% CI. We combined continuous 9. assuming treatment dropouts were responders for dichotomous
outcomes when means and standard deviations were available. outcomes.
When sufficiently similar continuous outcomes were measured
differently across studies we calculated an overall standardized
Informed decisions.
Summary of findings Results of the search

Summary of findings tables were created to present the main The electronic search of databases (conducted April 2013), yielded
findings of the review. We imported meta-analytic data from 826 citations for consideration for inclusion in the review, including
RevMan into GRADEprofiler version 3.6 to create summary of manuscripts in peer-reviewed journals, conference abstracts, and
findings tables for each of the three most clinically relevant clinical trial registrations. Employing secondary search methods,
comparisons: ICBT with therapist support versus waiting list including searching clinical trial registries, contacting experts in the
control, ICBT with therapist support versus unguided ICBT, and field, and searching the reference lists of eligible studies, resulted
ICBT with therapist support versus face-to-face CBT. The summary in another 471 citations for consideration. After de-duplication and
of findings tables present meta-analytic outcomes for each of following a brief screening of the titles and abstracts, 212 were
the continuous and dichotomous outcomes at post-treatment and retrieved for a more detailed evaluation of eligibility. One hundred
summarize the number of studies and participants included in each and fifty-one studies were subsequently excluded for failing to
analysis. In addition, GRADEprofiler allowed us to rate the quality meet our inclusion criteria. The PRISMA flow diagram shown in
of the evidence for each outcome for each comparison considering: Figure 1 outlines the study selection process and broad reasons
(a) risk of bias, (b) inconsistency, (c) indirectness, (d) imprecision, for exclusion. Studies were excluded if: (a) participants did not
and (e) publication bias. meet diagnostic criteria for an anxiety disorder, as assessed by
study authors (population), (b) the intervention of interest was not
RESULTS ICBT, did not involve a therapist, or included too much face-to-
face therapist contact (intervention), (c) the comparator was not
Description of studies appropriate given our selection criteria (comparator), (d) the trial
See Characteristics of included studies; Characteristics of excluded was not randomised or did not use adequate diagnostic measures
studies (methods), or (e) the trial was ongoing (ongoing).
Informed decisions.
Figure 1. PRISMA diagram of the search process.
Informed decisions.
Figure 1. (Continued)
After accounting for duplicate reports of the same trial, 30 studies trials), a research group in Switzerland (2 trials), and one in the
were eligible for inclusion in the meta-analyses. Seven studies Netherlands (1 trial).
identified from an updated search in September 2014 have been
added to studies awaiting classification (Andersson 2013; Berger Whereas researchers and treating clinicians were located
2014; Ivarsson 2014; Newby 2013), as have seven studies which at university-affiliated hospitals or mental health centres,
were completed during the process of completing this review participants received the intervention of interest in their home.
(Andrews 2011b; Andrews 2011c; Andrews 2012a; Berger 2012; Treatment took place over the Internet and by telephone. Face-to-
Carlbring 2012; Greist 2012; Nordgren 2012). These studies will be face CBT, when included in a trial, was conducted in a psychiatric
fully incorporated into the review in a timely, future update. setting (for example, hospital, mental health clinic).
E-mail correspondence to collect information to supplement Participants

data provided in the published reports was exchanged with Dr Participants were men and women over 18 years of age. The
Tomas Furmark (Furmark 2009a; Furmark 2009b), Dr Per Carlbring average mean age of study participants was 37.3 years. Women
(Carlbring 2001; Carlbring 2006; Carlbring 2007; Carlbring 2011), Dr represented an average of 67.1% of participants in each study.
Nickolai Titov (Titov 2008a; Titov 2008b; Titov 2008c; Titov 2009; The ethnicity of participants was not reliably reported. For most
Titov 2010; Titov 2011), Dr Britt Klein (Klein 2006; Richards 2006; studies, participants were recruited via media advertisements
Kiropoulos 2008), and Dr Wouter van Ballegooijen (van Ballegooijen or a recruitment website (28 studies); in a minority of studies
2013). participants were recruited via clinic referrals (Bergstrom 2010;
Hedman 2011).
Included studies
See Characteristics of included studies for details of individual All included participants qualified for one of the following anxiety
studies and Table 1 for a summary table of the characteristics of the disorder diagnoses: social phobia (11 trials), PD with or without
included studies. agoraphobia (8 trials), GAD (4 trials), PTSD (1 trial), and specific
phobia (1 trial). The five remaining studies included participants
Design with a range of anxiety disorder diagnoses. Twenty-six trials
included participants with co-morbid diagnoses and four studies
All of the 30 included studies were parallel group RCTs. For studies
did not report on their inclusion and exclusion criteria. Among all
in which participants in the waiting list, attention, information, or
studies, regardless of their inclusion or exclusion of co-morbidities,
online discussion group only control were given the opportunity
25 studies excluded participants who scored above a certain
to complete the treatment after their time on the waiting list, only
threshold on a measure of depressive symptoms, for example,
data from the original comparison were used in the meta-analyses.
above 30 on the Montgomery-Asberg Depression Rating Scale
There were no cross-over or cluster randomised trials.
(MADRS) (Svanborg 1994), and 26 studies excluded participants
Seven studies included multiple intervention arms: two (Titov who endorsed suicidal ideation, for example, on the MADRS
2008c; Furmark 2009a) compared the intervention of interest to suicide item, with the rationale that they were unclear about
two eligible comparators (a waiting list, and unguided CBT) so how to handle this high risk participant via a distance treatment.
were included in multiple meta-analyses (ICBT versus waiting Sixteen studies excluded participants with substance misuse or
list control, and ICBT versus unguided CBT), and five (Richards dependence problems and 16 studies excluded participants with
2006, Furmark 2009b; Robinson 2010; Berger 2011, Johnston 2011) active psychosis with the rationale that these problems would
included a third treatment arm not relevant to the present review. interfere with anxiety treatment.
Sample sizes Twenty-eight trials included participants who were using

psychiatric medication (including selective serotonin reuptake
Sample sizes of included studies ranged from 21 (12 in the inhibitors, serotonin norepinephrine reuptake inhibitors,
intervention arm, 9 in the comparator arm (Richards 2006)) to benzodiazepines, benzodiazepine derivatives, neuroleptics,
204 participants (102 in both the intervention and comparator tricyclic antidepressants, beta-blockers) concurrent with study
arms (Andersson 2012a)). The average study sample size was 74 participation (Berger 2009 excluded those using medication,
participants. In most studies there was an equal distribution of and Andersson 2009 did not report on co-use of medication).
participants between the treatment and control arms. Only 2 Participants were typically included only if they had been at a stable
studies had < 30 participants, 14 studies had 30 to 60 participants, dose for a certain time period (one to three months) preceding the
7 studies had 60 to 90 participants, and 6 studies had 90 to 130 study. Four studies (Carlbring 2001; Carlbring 2005; Carlbring 2006;
participants, with 1 outlier at 204 participants (Andersson 2012a). van Ballegooijen 2013) included participants engaged in another
type of psychological therapy concurrent with study participation,
Setting one of which had no participants that met this characteristic
Included studies came primarily from one research group in (Carlbring 2006).
Sweden (15 trials), two groups in Australia (Klein: 2 trials; Titov: 10
Informed decisions.
Interventions Four studies compared therapist-supported ICBT to unguided CBT

(that is, self-help). Finally, six studies compared the experimental
Experimental interventions
intervention to traditional, face-to-face group or individual CBT.
Twenty-nine studies included in the present review tested ICBT This number of studies adds up to more than the total number of
while one study investigated Internet-based BT with a focus studies because two studies included more than one comparator.
on exposure (Andersson 2009). ICBT interventions involved
participants following 5 (Andersson 2009; Berger 2009; Berger Outcomes
2011) to 15 (Hedman 2011) online treatment modules (mean = Primary outcomes
8; median = 7; mode = 6) with e-mail support from therapists.
Six studies also provided therapist support by telephone (Titov Each of the included studies reported on the efficacy of
2009; Robinson 2010; Titov 2010; Johnston 2011; Spence 2011; therapist-supported ICBT. Fourteen studies assessed participants
Titov 2011) while another seven of these studies also included post-treatment for clinically important improvement in anxiety
participation in an online discussion forum (Tillfors 2008; Furmark (a dichotomous outcome) and three studies reassessed this
2009a; Furmark 2009b; Bergstrom 2010; Spence 2011, Titov 2011; outcome at a follow-up of 6 to 12 months later. Each of the
Andersson 2012a). included studies reported on participants' disorder-specific anxiety
symptom severity using a validated self-report or observer-
Interventions ranged in length from 4 (Andersson 2009) to 15 rated instrument (a continuous outcome) at post-treatment. Eight
weeks (Hedman 2011) (mean = 9; median = 9; mode = 10). The studies assessed anxiety symptom severity at a follow-up of 6 to
degree of therapist involvement in the included interventions was 12 months later. Twenty of the included studies also measured
widely variable; the average total time spent by a therapist with a participants' symptoms of general anxiety using validated self-
participant ranged from a minimum of 25 minutes (Carlbring 2001) report instruments at post-treatment. Six studies assessed general
to a maximum of 376 minutes (Richards 2006) with the overall anxiety at a follow-up of 6 to 12 months later. Please see Table 1
mean = 132 minutes and median = 120 minutes (2 studies (Berger and Characteristics of included studies for more details of outcome
2009; Berger 2011) did not report therapist contact time). Similarly, assessment.
among studies that reported this information, the average number
of e-mails sent by study therapists ranged from a minimum of 5 It was rare for studies to report adverse events. In fact, adverse
(Spence 2011; Titov 2011) to a maximum of 24 (Titov 2009; Titov events could only be assumed from measures of participants'
2010) with the overall mean = 14 e-mails and median = 12 e-mails. symptom deterioration during the study or reasons for participant
dropout related to the treatment.
Of the 30 included studies, 28 specified that treatment was
provided by a licensed clinical psychologist (22 studies) or Secondary outcomes
clinical psychology graduate students in training, or both (15 Twenty-six studies measured quality of life at post-treatment, while
studies). Clinical psychology graduate students providing therapy six studies included quality of life as an outcome at 6 to 12 month
were enrolled in master's or doctoral psychology programs as follow-up.
required for them to practice in their country. Of those studies
in which licensed clinical psychologists delivered the treatment, Participant satisfaction with treatment was indexed by 15 studies
six specified that clinicians were provided with supervision from at post-treatment. A variety of different measures of treatment
an expert in the field. Similarly, of those studies in which satisfaction were used ranging in degrees of comprehensiveness
clinical psychology students provided treatment, 12 specified and complexity. Across different measurement approaches,
that supervision from an expert in the field was provided. The participants were most commonly asked to indicate their overall
two remaining studies specified that therapy was delivered by satisfaction with the treatment program, their satisfaction with
therapists trained by the treatment founder (Andersson 2009) particular portions of the treatment program (for example,
and a psychiatry registrar (Wims 2010). Details on the experience therapist correspondence, Internet modules), and their satisfaction
and training of study therapists, if provided, can be found in the with the pace of the treatment program. Of the 15 studies, only
Characteristics of included studies section. four reported treatment satisfaction for both the experimental
and comparator interventions; the remaining trials compared the
Comparator Interventions experimental intervention to a waiting list control, which did not
Twenty-two studies compared the experimental intervention lend itself to an evaluation of satisfaction.
to a waiting list, attention, information, or online discussion
Excluded studies
group only control. All but five of these studies included strict
waiting list control conditions with no treatment provided to Studies were excluded for a variety of reasons (see Characteristics
participants and assessments occurring after the designated of excluded studies and Figure 1, the former of which lists a
waiting list period. Of the remaining five studies, the control number of studies that were most like the included studies but
condition in one study included a weekly self-report assessment differed in important ways that prevented inclusion). Studies were
but no intervention (Furmark 2009a). The control condition in two frequently excluded because the intervention was: (a) not distance-
studies (Richards 2006; van Ballegooijen 2013) provided basic non- based, (b) distance-based but included more than two sessions
treatment disease-related information to participants and one of of face-to-face contact between therapist and participant, (c) not
these studies (Richards 2006) included weekly status check-ins delivered by a therapist (that is, was a self-help program), or (d)
by phone. Finally, the control condition in two studies permitted not CBT. Similarly, studies were excluded if participants did not
participants to engage in an online discussion group (Carlbring meet our criteria because they had subclinical anxiety symptoms
2011; Andersson 2012a). or an anxiety disorder was not their primary diagnosis. We also
excluded a number of studies because a closer look showed that
Informed decisions.
they were not RCTs or did not compare the intervention of interest Studies awaiting classification
to a comparison group that met the eligibility criteria.
The updated search in September 2014 showed that seven
Ongoing studies previously ongoing studies had been completed; we classified
these studies as awaiting classification, with the plan to fully
We originally identified 14 ongoing studies (16 references; see incorporate them in an immediate update of the review. The
Ongoing studies). The updated search in September 2014 showed updated search also identified seven additional studies which
that seven of these studies (Andrews 2011b; Andrews 2011c; we have classified as awaiting classification; these will also be
Andrews 2012a; Berger 2012; Carlbring 2012; Greist 2012; Nordgren fully incorporated in the updated version of the review. For more
2012) had been completed and so we classified these as studies details on these studies see Characteristics of studies awaiting
awaiting classification (see below). We will fully incorporate these classification.
studies in the upcoming update of the review. The updated search
also identified four additional ongoing studies (Rollman 2012; Titov Risk of bias in included studies
2012; Lindner 2013; Miclea 2014). We found details in the full report
of one study (Carlbring 2010) which revealed it no longer qualified Results of the risk of bias assessments of included studies are
for inclusion, and so we moved this study from ongoing studies to summarized succinctly in Figure 2 and Figure 3. Overall, the risk of
excluded studies. This left a total of nine ongoing studies; for more bias in the included studies was low, with some notable exceptions
details see Characteristics of ongoing studies. related to the nature of clinical trials of psychological treatments.
Figure 2. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages
across all included studies.
Informed decisions.
Figure 3. Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Informed decisions.
Informed decisions.
Allocation As participants were not blind to their treatment condition in

the included studies, self-report outcomes measured in all of
The majority of included studies (n = 27) used an adequate
the included studies were not blinded. Fifteen studies measured
method of randomisation, primarily an online random number
outcomes using observer-rated instruments. In 10 of these studies
generator, to avoid selection bias. The three remaining studies
interviewers who were blind to the treatment condition conducted
reported that participants were randomised but did not describe
the outcome assessments ensuring a low risk of bias. Of the
the randomisation procedure.
remaining five studies, one was compromised by participants who
Most study authors (n = 25) did not adequately report allocation too frequently revealed their treatment condition to interviewers
concealment. The remaining five studies reported allocation (Berger 2011) and four used at least one interviewer who was aware
concealment procedures that would have minimized the risk of of participants' random assignment (Richards 2006; Wims 2010;
selection bias (for example, random assignment was maintained Spence 2011; Titov 2011).
by an independent research team member not involved in other
Incomplete outcome data
study aspects who gave randomisations to participants just prior to
treatment commencement). Attrition bias was not a significant issue in 28 of the included
studies. These 28 studies used an ITT analysis by either carrying
Blinding forward the last observations or using mixed models analyses to
The blinding of participants and study personnel is difficult control for outcomes lost to attrition. Moreover, rates of attrition
when investigating the efficacy of psychological treatments. Unlike were often quite similar between treatment conditions. One study
pharmacological trials in which medication type can be concealed, did not use an ITT approach and as such may have been biased due
it is very difficult to blind participants to the characteristics of to attrition (Andersson 2009). A second study did use ITT analyses
the treatment they are receiving as they are active participants. but had large attrition that may have biased the findings despite
Similarly, it is impossible to blind study therapists to the treatment the use of ITT analyses (van Ballegooijen 2013). We investigated the
they are delivering as they take an active role in its execution. effect of these studies using sensitivity analyses.
As such, each of the included studies could be rated as having
Selective reporting
a high risk of bias because participants and personnel were not
blind to treatment assignment. Though this study characteristic Nineteen of the included studies had been registered as clinical
was a limitation across studies, because it is standard practice with trials allowing for a more accurate analysis of selective reporting.
this type of clinical trial we did not find it necessary to conduct Of these 19 studies, 12 reported on all outcomes outlined in the
sensitivity analyses based on the characteristic. trial registration. For six of the studies, one outcome outlined in
the trial registration was not reported in the final manuscript (Titov
We indexed blinding of outcome assessment separately for self- 2008a; Titov 2008b; Titov 2008c; Berger 2009; Johnston 2011; van
report versus observer or interview-rated outcome measures.
Informed decisions.
Ballegooijen 2013) and they were rated as having an unclear risk 1. Therapist-supported ICBT versus waiting list, attention,
of bias. One study (Titov 2010) had many outcomes indicated in information, or online discussion group only control
the trial registration that were not reported in the final manuscript
Twenty-two studies compared therapist-delivered distance CBT
and was rated as having a high risk of bias. Those studies that
with a waiting list, attention, information, or online discussion
were not registered reported results for each of the outcomes they
group only control: Carlbring 2001; Carlbring 2006; Richards 2006;
measured, as described in their method; however, given the lack of
Carlbring 2007; Titov 2008a; Titov 2008b; Titov 2008c; Berger 2009;
trial registration or protocol publication, these studies were rated
Furmark 2009a; Titov 2009; Robinson 2010; Titov 2010; Wims 2010;
as having an unclear risk of bias.
Carlbring 2011; Johnston 2011; Paxling 2011; Spence 2011; Titov
Other potential sources of bias 2011; Andersson 2012a; Andersson 2012b; Silfvernagel 2012; van
Ballegooijen 2013. See Table 2 for subgroup analysis details.
Two of the included studies had a high risk of bias due to differences
in baseline severity between treatment groups (Richards 2006; Primary outcomes
Titov 2011). Four studies did not report any evaluations of
1.1 Clinically important improvement in anxiety
differences in baseline severity at baseline and so were rated
as having unclear risk of bias in this domain (Tillfors 2008; Nine studies assessed clinically important improvement in anxiety
Andersson 2009; Bergstrom 2010; Paxling 2011). One study reported at post-treatment after therapist-supported ICBT versus a waiting
differences in age and marital status between study groups list, attention, information, or online discussion group only
(Robinson 2010); as it was unclear if this would have an effect on control. A meta-analysis with 325 treatment participants and 319
study results, this study was rated as having unclear risk of bias in controls yielded a RR of 4.18 (95% CI 2.42 to 7.22; Analysis
this domain. 1.1) in favour of the experimental intervention, with substantial
heterogeneity (I2 = 59%). These results did not change significantly
Effects of interventions following sensitivity analyses according to active waiting list
See: Summary of findings for the main comparison Therapist- control conditions, high risk of bias (ROB), or assuming dropouts
supported ICBT compared to waiting list, attention, information, or were treatment responders. There was no difference in treatment
online discussion group only control for anxiety disorders in adults; effect when analyzed by anxiety disorder (PD, social phobia, GAD,
Summary of findings 2 Therapist-supported ICBT compared to and studies of mixed anxiety disorders; only one study examined
unguided CBT for anxiety disorders in adults; Summary of findings PTSD), amount of therapist contact (medium or low; only one study
3 Therapist-supported ICBT compared to face-to-face CBT for examined high), or research group (Sweden, Australia 1; only one
anxiety disorders in adults study was done by Australia 2).
1.2 Reduction in disorder-specific anxiety symptom severity

Primary and secondary outcomes are reported by comparison
below. Because adverse events were so rarely reported, they are All 22 studies that compared therapist-supported ICBT to a
not reported by comparison but are instead reported here. Only waiting list, attention, information, or online discussion group
four studies included a measure that allowed for the assessment only control assessed disorder-specific anxiety symptoms at post-
of participant deterioration over the course of treatment, for treatment. Taken together, these 22 studies included 801 treatment
example, the CGI (Guy 1976). Andersson 2012a and Titov 2011 participants and 772 control participants. Meta-analytic findings
each identified one participant in the treatment condition who showed a significant SMD of -1.12 (95% CI -1.39 to -0.85; Analysis
had deteriorated over the course of the study, but in neither case 1.2; see Figure 4) in favour of the experimental condition, with
could their deterioration be linked to the treatment itself. Carlbring considerable heterogeneity (I2 = 83%). These results did not change
2011 reported that no participants in their treatment condition significantly following sensitivity analyses according to active
had deteriorated. Hedman 2011 found one to two participants had waiting list control conditions or high ROB. One study, Titov 2010,
deteriorated in each of the ICBT and face-to-face CBT conditions, included three separate anxiety disorder subgroups that completed
but there was no difference between conditions. disorder-specific measures so this study was entered as three
studies in this meta-analysis: Titov 2010 GAD; Titov 2010 Panic; and
Titov 2010 Social Phobia. There was no difference in treatment
effect when analyzed by anxiety disorder, amount of therapist
contact, or research group.
Informed decisions.
Figure 4. Forest plot: therapist-supported ICBT versus waiting list control for anxiety symptom severity at post-
treatment.
1.3 Reduction in general anxiety symptom severity participants across these studies reporting being very or mostly
Fourteen studies assessed participants' general anxiety after satisfied with the treatment. Several studies reported that over 90%
therapist-supported ICBT (508 treatment participants) versus a of participants found the quality of the online treatment modules
waiting list, attention, information, or online discussion group and their correspondence with a therapist to be excellent or good.
only control (496 controls). Data analysis resulted in a SMD of Only a few studies mentioned any problems or dissatisfaction
-0.79 (95% CI -1.10 to -0.48; Analysis 1.3) showing a significantly with the intervention. Most notably, three studies reported that
greater decrease in general anxiety following the experimental a majority of participants (70%) found the treatment moved too
quickly (Carlbring 2006; Titov 2008a; Titov 2008b). Several studies
intervention, with considerable heterogeneity (I2 = 80%). Results
reported small numbers of participants who had been dissatisfied
were consistent following sensitivity analyses according to active
with treatment: 3% dissatisfied with treatment (Carlbring 2006);
waiting list control conditions and high ROB. There was no
6% rated quality of therapist correspondence as neutral or
difference in treatment effect when analyzed by anxiety disorder,
somewhat dissatisfied, 1% rated quality of therapist contact as very
amount of therapist contact, or research group.
dissatisfied (Titov 2008b); 11% dissatisfied with treatment (Berger
Secondary outcomes 2009); 13% neutral or somewhat dissatisfied with treatment,
2% rated quality of therapist correspondence as unsatisfactory
1.4 Quality of life (Robinson 2010); 5% rated quality of therapist correspondence as
Twenty studies reported on participants' quality of life following unsatisfactory (Titov 2010); 16% neutral or somewhat dissatisfied
therapist-supported ICBT (707 treatment participants) versus a with treatment, (Johnston 2011). Titov 2008c also reported that 7%
waiting list, attention, information, or online discussion group only of participants found that their confidence in their ability to manage
control (688 controls). Analysis resulted in a SMD of 0.51 (95% CI their symptoms and their motivation to continue practicing their
0.40 to 0.61; Analysis 1.4) in favour of the experimental intervention, skills had not changed. Berger 2009 reported that one participant
with minimal heterogeneity (I2 = 0%) that may not be important. rated the self-help modules as too difficult and one participant
Results did not change significantly following sensitivity analyses indicated that they did not understand the purpose of the self-help
according to active waiting list control conditions or high ROB. modules.
1.5 Participant satisfaction with the intervention 2. Therapist-supported ICBT versus unguided CBT
A comparison of treatment satisfaction was not warranted as Four studies compared therapist-supported ICBT with unguided
authors expectedly did not report on the satisfaction of participants CBT: Titov 2008c; Furmark 2009a; Furmark 2009b; Berger 2011. See
in the waiting list, attention, information, or online discussion Table 3 for subgroup analysis details for this comparison.
group only controls. Thirteen studies reported on participants'
satisfaction with treatment. Overall, participants reported a high
level of satisfaction with the intervention, with roughly 90% of
Informed decisions.
Primary outcomes difficult to estimate heterogeneity (I2 = 0%). No sensitivity analyses

2.1 Clinically important improvement in anxiety were required.
Only Berger 2011 assessed clinically important improvement in 2.5 Participant satisfaction with the intervention
anxiety after therapist-supported ICBT versus unguided CBT. They Two studies indexed participant satisfaction with the intervention.
reported that 16/27 participants receiving therapist-supported Berger 2011 found that treatment satisfaction was significantly
ICBT and 15/27 participants completing unguided CBT no longer higher in the therapist-supported ICBT condition as compared
met the diagnostic criteria post-treatment (Analysis 2.1). to the self-help condition according to the Client Satisfacton
2.2 Reduction in disorder-specific anxiety symptom severity Questionnaire (Attkisson 1982). Similarly, Titov 2008c found that
a significantly greater number of participants in the therapist-
The four studies that compared therapist-supported ICBT to supported ICBT condition as compared to the self-help condition
unguided CBT (that is, self-help) assessed disorder-specific anxiety were very or mostly satisfied with their treatment (no participants
symptoms at post-treatment. Combined, these studies included reported being dissatisfied with treatment). However, Titov 2008c
127 treatment and 126 control participants and resulted in a non- reported no differences between conditions in perceptions of
significant SMD of -0.24 (95% CI -0.69 to 0.21; Analysis 2.2), with how logical the treatment was, participants' confidence in
substantial heterogeneity (I2 = 68%). At 6 to 12 month follow- recommending the treatment to a friend, and the extent to which
up, 3 studies reported on this outcome; a meta-analysis of 96 treatment had increased participants' confidence in managing
treatment and 96 comparator participants resulted in a SMD of -0.30 their symptoms. Seven per cent of participants in the ICBT
(95% CI -0.58 to -0.01; Analysis 2.3) in favour of the experimental condition reported that the treatment had not changed their
intervention with minimal but difficult to estimate heterogeneity (I2 confidence in managing their symptoms or their motivation to keep
= 0%). No sensitivity analyses were required. practicing techniques they had learned.
Subgroup analyses based on anxiety disorder were not warranted 3. Therapist-supported ICBT versus face-to-face CBT
as all studies investigated the efficacy of ICBT for social phobia.
Six studies compared therapist-supported ICBT with face-to-face
Subgroup analyses based on therapist contact did result in one
CBT: Carlbring 2005; Kiropoulos 2008; Tillfors 2008; Andersson
change in treatment effect: For the primary outcome disorder-
2009; Bergstrom 2010; Hedman 2011. See Table 4 for subgroup
specific anxiety symptoms at follow-up, a meta-analysis of two
analysis details for this comparison.
studies with medium therapist contact (Furmark 2009a; Furmark
2009b) resulted in a non-significant difference with minimal but Primary outcomes
difficult to estimate heterogeneity (SMD -0.31, 95% CI -0.65 to 0.03;
I2 = 3%). There was no difference in treatment effect when analyzed 3.1 Clinically important improvement in anxiety
by research group. Four studies assessed clinically important improvement in
anxiety at post-treatment after therapist-supported ICBT (185
2.3 Reduction in general anxiety symptom severity
treatment participants) versus face-to-face CBT (180 comparator
Only two studies assessed participants' general anxiety after participants). Meta-analysis yielded a non-significant RR of 1.09
therapist-supported ICBT (69 treatment participants) versus self- (95% CI 0.89 to 1.34; Analysis 3.1), with minimal heterogeneity
help interventions (69 comparator participants). Data analysis that may not be important (I2 = 0%). At 6 to 12 month follow-
resulted in a non-significant mean difference of 0.28 (95% CI up, the results of 3 studies that reported on clinically important
-2.21 to 2.78; Analysis 2.4), with minimal but difficult to estimate improvement in anxiety, with 139 treatment and 140 comparator
heterogeneity (I2 = 0%). A similar result was found at 12 month participants, resulted in a non-significant RR of 1.10 (95% CI 0.94
follow-up with the same studies; the mean difference was 0.72 (95% to 1.27; Analysis 3.2), again with minimal heterogeneity (I2 = 0%).
CI -2.12 to 3.57; Analysis 2.5), with minimal but difficult to estimate Results did not change significantly following a sensitivity analysis
heterogeneity (I2 = 0%). No sensitivity analyses were required. assuming dropouts were treatment responders. Results for this
Subgroup analyses based on anxiety disorder were not warranted outcome remained non-significant following subgroup analyses by
as both studies investigated the efficacy of ICBT for social phobia. anxiety disorder (PD and social phobia; only one study examined
Subgroup analyses based on therapist contact and research group specific phobia), therapist contact (low and high; only one study
did not change the treatment effect. involved medium therapist contact), and research group (Sweden;
only one study was done by Australia 2 and no studies by Australia
Secondary outcomes 1).
2.4 Quality of life
3.2 Reduction in disorder-specific anxiety symptom severity
Three studies indexed quality of life of participants following The six studies that compared therapist-supported ICBT to face-
therapist-supported ICBT (100 treatment participants) versus to-face CBT assessed changes in symptom specific anxiety. Using
unguided CBT (99 control participants). Data analysis resulted in these 6 studies, including 215 treatment participants and 209
a non-significant SMD of 0.07 (95% CI -0.37 to 0.50; Analysis 2.6), control participants, meta-analysis resulted in a non-significant
with moderate to substantial heterogeneity (I2 = 58%). At six to 12 SMD of 0.09 (95% CI -0.26 to 0.43; Analysis 3.3; see Figure 5), with
month follow-up, only two of these studies indexed quality of life of substantial heterogeneity (I2 = 66%). At 6 to 12 month follow-
participants following treatment (69 treatment and 69 comparator up, data from 5 studies, including 171 treatment participants and
participants), with meta-analysis showing a similar non-significant 170 comparator participants, could be used to assess changes
SMD of -0.19 (95% CI -0.53 to 0.14; Analysis 2.7), with minimal but in symptom specific anxiety. Meta-analysis resulted in a non-
significant SMD of -0.21 (-0.42 to 0.0; Analysis 3.4) with minimal
Informed decisions.
heterogeneity that may not be important (I2 = 0%). Results one study that did not use ITT analysis and had high ROB
remained non-significant following a sensitivity analysis excluding (Andersson 2009).
Figure 5. Forest plot: therapist-supported ICBT versus face-to-face CBT for anxiety symptom severity at post-
treatment.
For disorder-specific anxiety symptoms, at post-treatment a meta- 3.6) with minimal heterogeneity that may not be important (I2
analysis of the studies investigating PD (Carlbring 2005; Kiropoulos = 0%). Results remained non-significant following a sensitivity
2008; Bergstrom 2010) found a significant SMD of 0.29 (95% CI analysis excluding one study that did not use ITT analyses and had
0.03 to 0.54) with minimal but difficult to estimate heterogeneity high ROB (Andersson 2009). Results for this outcome remained non-
(I2 = 0%) in favour of face-to-face CBT. In contrast, at post- significant following subgroup analyses by anxiety disorder (PD and
treatment a meta-analysis of two studies investigating social social phobia; only one study examined specific phobia), therapist
phobia (Tillfors 2008; Hedman 2011) remained non-significant (in contact (low and high; only one study involved medium therapist
line with the overall meta-analysis) with substantial to considerable contact), and research group (Sweden; only one study was done by
heterogeneity (SMD -0.18, 95% CI -0.92 to 0.5; I2 = 76%). Australia 2 and no studies by Australia 1).
Unexpectedly, at 6 to 12 month follow-up it was only the meta-
analysis of social phobia studies (Tillfors 2008; Hedman 2011) Secondary outcomes
that showed a significant difference between groups, with an 3.4 Quality of life
SMD of -0.39 (95% CI -0.71 to -0.08) with minimal but difficult
Five studies reported on participants' quality of life following
to estimate heterogeneity (I2 = 0%) in favour of the experimental
therapist-supported ICBT (198 treatment participants) versus face-
intervention, while the meta-analysis of PD studies (Carlbring
to-face CBT (194 comparator participants). Analysis resulted in
2005; Bergstrom 2010) was non-significant, also with minimal but
a SMD of 0.26 (95% CI 0.06 to 0.45; Analysis 3.7) in favour of
difficult to estimate heterogeneity (SMD -0.04, 95% CI 0.36 to 0.28;
the experimental intervention, with minimal heterogeneity that
I2 = 0%).
may not be important (I2 = 0%). This trend continued at 6 to 12
A subgroup analysis of studies with high therapist contact for month follow-up. Four studies comprising 158 treatment and 158
the disorder-specific anxiety symptoms outcome (Kiropoulos 2008; comparator participants resulted in a SMD of 0.33 (95% CI 0.11 to
Tillfors 2008) resulted in a significant SMD of 0.42 (95% CI 0.05 0.55; Analysis 3.8) in favour of the experimental intervention, again
to 0.78), with minimal but difficult to estimate heterogeneity (I2 with minimal heterogeneity (I2 = 0%). No sensitivity analyses were
= 0%), in favour of face-to-face CBT at post-treatment. The meta- required.
analysis of studies with low therapist contact remained non-
3.5 Participant satisfaction with the intervention
significant (SMD -0.08, 95% CI -0.63 to 0.46) with substantial to
considerable heterogeneity (I2 = 76%) (Andersson 2009; Bergstrom Two studies indexed participant satisfaction with the intervention.
2010; Hedman 2011). Overall, treatment satisfaction was high across both therapist-
supported ICBT and face-to-face CBT. In one study (Tillfors
3.3 Reduction in general anxiety symptom severity 2008), only one participant in the ICBT condition and two
Five studies reported participants' levels of general anxiety post- participants in the face-to-face condition reported being "neutral/
treatment. The five studies combined in the meta-analysis included somewhat dissatisfied with treatment" and no participants
163 treatment participants and 154 comparator participants and reported being "very dissatisfied" with treatment. Both studies
resulted in a non-significant SMD of 0.17 (95% CI -0.35 to 0.69; found no significant difference between conditions in participants'
overall satisfaction with the intervention or their perceptions of
Analysis 3.5), with substantial to considerable heterogeneity (I2
improvement as a result of treatment.
= 79%). When the Kiropoulos 2008 study was removed from the
analysis (because it presented transformed data, which we back- A notable significant difference between treatment conditions
transformed to include in the analysis), the resulting SMD remained appeared in one instance: Kiropoulos 2008 found that participants
non-significant at -0.13 (95% CI -0.38 to 0.13) and heterogeneity was receiving therapist-supported ICBT reported significantly less
reduced (I2 = 0%). At 6 to 12 month follow-up 4 studies reported enjoyment in communicating with their therapist as compared to
participants' level of general anxiety. The 4 studies included participants receiving face-to-face CBT.
121 treatment participants and 116 comparator participants and
yielded a non-significant SMD of -0.16 (95% CI -0.42 to 0.09; Analysis
Informed decisions.
Sensitivity analysis our ability to draw firm conclusions based on these analyses. More
research is needed in these areas.
Sensitivity analyses are detailed in the results section above.
Given the available studies for this review, some of the planned Overall completeness and applicability of evidence
sensitivity analyses were not warranted. First, sensitivity analyses
based on the blinding of participants or personnel, or both, in Taken together, the studies included in the present review go a long
the included studies were not conducted because blinding of way toward answering the question, is therapist-supported ICBT an
participants and personnel is not standard practice with this type efficacious treatment for anxiety disorders in adults? In particular,
of clinical trial. Second, as none of the included studies were the included studies are of sufficient number to comprehensively
cluster randomised trials and none of the included studies with compare the efficacy of therapist-supported ICBT to a waiting list,
multiple intervention arms had selective reporting of intervention attention, information, or online discussion group only control.
comparisons, sensitivity analyses based on these characteristics There are fewer, but still sufficient, studies to compare the efficacy
were not conducted. Third, as we were not required to impute any of therapist-supported ICBT to traditional face-to-face CBT. In
standard deviations, we also eliminated that planned sensitivity comparison, the number of studies comparing therapist-supported
analysis. Only one study included transformed data (Kiropoulos ICBT to unguided CBT (that is, self-help) is limited and therefore
2008) and it was discussed in section 3.3 above, and only one study findings with respect to this comparison must be interpreted with
did not use ITT data (Andersson 2009) and is discussed in sections some caution.
3.2 and 3.3 above. Finally, as LOCF was the primary method of
ITT analysis reported by authors, we did not exclude studies using In terms of the applicability of the evidence to ICBT interventions
LOCF. and particular patient populations, several factors warrant
consideration when interpreting the present findings. First,
DISCUSSION the included interventions are quite heterogeneous. While all
studies investigated therapist-supported ICBT, the nuances of each
Summary of main results intervention (for example, length, number of online modules,
nature of therapist support) varied widely. It seems prudent to
Please refer to the Summary of findings for the main comparison, note that while these interventions seem efficacious as a whole,
Summary of findings 2, and Summary of findings 3 for a summary the optimal characteristics of these interventions have yet to be
of the main results. identified.
The present review investigated the efficacy of therapist-supported Second, the included studies investigated a number of different
ICBT in treating anxiety disorders in adults. We identified 30 studies anxiety disorders with a particular focus on PD, social phobia, and
to be included in the review, comparing the intervention of interest GAD, either separately or as part of a transdiagnostic treatment
to a waiting list, attention, information, or online discussion group package. As such, we can be most confident that the present
only control, unguided CBT, and face-to-face group or individual findings apply to the treatment of these disorders. More research
CBT. is needed into ICBT for other anxiety disorders, such as OCD, PTSD,
and specific phobia.
The present findings suggest that therapist-supported ICBT is
more efficacious than a waiting list, attention, information, or Third, researchers have previously raised some concerns about
online discussion group only control in leading to clinically the participants included in investigations of ICBT, as many
important improvement in anxiety, reducing anxiety symptoms of these studies recruit participants from the community via
(both disorder-specific and general), and improving quality of life. media advertisements (for example, Cuijpers 2009). There is some
Results also generally showed no difference in outcomes following question as to whether these participants are similar enough to
therapist-supported ICBT versus unguided CBT at post-treatment, participants recruited for face-to-face CBT RCTs, who tend to be
though results are limited by low quality evidence due to a limited recruited via clinic referrals. Despite this concern, research by
number of studies (that is, imprecision). Moreover, results suggest Titov and colleagues (Titov 2010b) found that ICBT participants
that therapist-supported ICBT may not be significantly different are as severe in terms of symptom severity, distress, and disability
from face-to-face group and individual CBT in treating anxiety as individuals attending a face-to-face clinic and more severe
disorders. Meta-analyses revealed no significant differences in than individuals identified via an epidemiological survey. We also
clinically important improvement in anxiety or reduction in anxiety attempted to account for this possible difference in participant
symptoms (both disorder-specific and general) at post-treatment characteristics by including only individuals with an anxiety
or follow-up for these two interventions. disorder diagnosed using a standardized instrument.
At 6 to 12 month follow-up, results generally mirror the post- Despite the heterogeneity of the interventions and populations
treatment findings but are limited by the small number of across studies, the robustness of findings following sensitivity
studies and the degree of variability in the interventions under analyses lends credence to the efficacy of therapist-supported ICBT
investigation across studies. Thus, these findings should be as an alternative method of delivering CBT to those with anxiety
interpreted with caution. disorders who are in need of intervention. It is important to note,
however, that the search for studies for this review was conducted
All findings largely remained robust following sensitivity analyses
over one year ago. An updated search conducted in September 2014
conducted to explore the impact of potential sources of bias or
identified four new completed studies, seven previously ongoing
heterogeneity. Subgroup analyses suggest that there may be some
studies that have now been completed, and three new ongoing
differences in outcome based on the type of anxiety disorder being
studies that should be included in the present review. This is a fast-
treated or the amount of therapist contact in the intervention;
moving area of research and as such we plan to conduct an update
however, the small number of studies within each subgroup limits
Informed decisions.
of this review after its publication, in which these new studies will participants, etc. Nevertheless, these studies did vary over the
be fully incorporated. years in terms of the anxiety disorder investigated and amount and
nature of therapist contact with participants.
Quality of the evidence
All subgroup analyses are complicated by the fact that only a
We considered the quality of the evidence of the included studies small number of trials tended to be included in each analysis,
using the GRADE tool (Higgins 2011b). With respect to risk of bias, making it difficult to estimate heterogeneity. Thus, heterogeneity
the included evidence is of moderate quality as there were only does remain somewhat of a concern in the present review. While
a few concerns with the internal validity of the included studies. this was unexpected, it may be that there is simply too much
The nature of clinical psychotherapy trials is such that keeping variability in study methods, populations, outcome measures, etc.
the treatment condition concealed from the participant or the across studies and not enough studies to support meaningful
therapist delivering the treatment is difficult, if not impossible, subgroup analyses at this time. Importantly, our speculation is
thus introducing some potential for bias but nothing that could that this heterogeneity might be explained by the expected factors
be improved upon by higher quality study design. Conversely, discussed here as opposed to any bias in the included studies. An
there were some difficulties with (a) blinding of outcome assessors, increase in the number of studies in this area in the future may allow
and (b) incomplete outcome data in several of the included us to explore heterogeneity more robustly and meaningfully.
studies. Sensitivity analyses excluding these studies suggest that
any potential bias introduced by these studies did not affect the In considering the quality of the evidence, we also examined
meta-analytic outcomes. It should be noted that there may be indirectness of the included studies (that is, the degree to which the
some concerns with selective outcome reporting, but these remain included studies address the review objective) and the imprecision
unclear. Selective outcome reporting has been found to be an of each study's findings. Across included studies, we had no
important concern in non-pharmaceutical trials (Milette 2011), concerns with indirectness. As far as imprecision, some of the
such as those included here. Approximately one third of the studies included studies are limited by small sample size. The meta-
included in this review were not prospectively registered on a trial analyses attempt to address such small samples by combining
database. As such, it is impossible to discern if these studies are studies, where appropriate. Precision of findings may also be
biased by selective reporting. It may be that with the advent of affected by rates of dropout across interventions, particularly
trial registration becoming more common, and expected, updates if there is the chance that one of the two interventions being
to this review will be able to provide a more clear estimate of the compared is likely to lead to greater dropout. Given some of the
risk of selective outcome reporting. characteristics of ICBT (for example, engaging from a distance, no
requirement to commit to appointment times or be accountable),
There is a large degree of heterogeneity in a number of the one might expect there to be greater dropout rates with this
meta-analyses in this review, reducing the quality of some of type of treatment. However, the present findings suggest this
the evidence. Subgroup and sensitivity analyses provide some may not be the case. There were generally quite similar rates of
indication of what may account for the heterogeneity, but there is dropout across the interventions investigated (experimental and
by no means a clear answer. Some degree of heterogeneity may comparator). Almost all studies used a rigorous and somewhat
have emerged because we included studies of a range of anxiety conservative method to account for missing data. Sensitivity
disorders, including PD, social phobia, GAD, specific phobia, and analyses on dichotomous outcomes, assuming dropouts were
PTSD, in the meta-analyses. It seems possible there are nuances treatment responders, did not significantly change the meta-
unique to each of these disorders and their treatment that might analytic outcomes. These details suggest that the precision of
facilitate or hamper the efficacy of their treatment via therapist- findings are not significantly threatened by treatment dropout
supported ICBT. Some subgroup analyses by anxiety disorder rates.
resulted in an important decrease in heterogeneity, however
in other cases heterogeneity did not decrease at all. This may Finally, we considered whether publication bias might have
have been in part because even within studies of the same affected the evidence. The number of studies within each
disorder, researchers employed different outcome measures to comparison in the present review only permitted the analysis
assess treatment outcomes. The variability in outcome measures of funnel plots for several outcomes for the comparison of the
within and across studies that were amalgamated in the meta- intervention and a waiting list control. A visual inspection of these
analyses may account for some important heterogeneity. Support funnel plots suggested that there may have been a small study
for this hypothesis may be found in the fact that the quality effect (that is, the potential for some publication bias). Because
of life outcome tends to show the least heterogeneity across there were less than 10 studies in the other meta-analyses in this
comparisons as well as the least variability in measures used to review (in accordance with the guidelines for the use of funnel plots
assess quality of life. Also of importance, the nature of the ICBT in the Cochrane Handbook for Systematic Reviews of Interventions
interventions included in this review is quite diverse in terms (Higgins 2011a)), we did not analyze publication bias using funnel
of length, number of online modules, and nature of therapist plots for the remaining comparisons. To complement the use of
contact. It may be that the nuances of these treatments led to funnel plots, we looked to match trials recorded in clinical trial
nuanced differences in treatment outcome. However, subgroup registries with published manuscripts. Accounting for the fact that
analyses based on amount of therapist contact, for example, did many of the most recent registered trials are still ongoing or may
not sufficiently and consistently reduce heterogeneity. Similarly, be in the process of being published, we only observed a handful
subgroup analyses by research group did not consistently lead to of registries that could not be matched with a published trial.
decreases in heterogeneity. This is surprising given the assumption This would suggest that, at least recently (since trial registration
that studies conducted within the same research laboratory would has been strongly encouraged), publication bias may not be a
have some degree of consistency in methods, outcome measures, significant concern for this review. However, our findings with
Informed decisions.
respect to the consistency between trial registries and published the field of ICBT and is a result of the lack of consensus in
studies do not rule out earlier publication bias or the possibility clinical psychology research in general regarding the most robust
of bias due to smaller-scale, unfunded studies that may not have way to assess clinically significant improvement resulting from
been registered. In a further effort to assess for publication bias, treatment. Consequently, there are a variety of ways to assess
we contacted authors in the field to inquire about any unpublished treatment outcome, including measures of remission, recovery,
findings and were only informed of one study that was unfinished clinically significant improvement, and high end state functioning.
and unpublished due to difficulties with funding. With these factors Because this issue is quite prevalent in ICBT for anxiety studies
in mind, we cannot make a conclusive statement about publication as well, we elected to combine each of these unique ways of
bias. Publication bias may not limit the quality of the included determining notable changes in symptomatology and functioning
evidence but readers should keep the possibility of this bias in mind post-treatment by creating the clinically important improvement
when interpreting the review findings. outcome. While relevant and useful, the nuances of remission and
recovery may be lost by being subsumed within this category.
Overall, the included evidence, across studies and comparisons, As the field expands, and consistency in reporting treatment
is of low to moderate quality. In many cases reductions in outcomes increases, it may be useful to subdivide this outcome to
quality tend to be due to heterogeneity, which may be explained more clearly capture remission and recovery.
by meta-analytic methods rather than the evidence itself. This
finding lends confidence to the present meta-analytic results and It is also worth mentioning that given the conceptual and
conclusions about the efficacy of therapist-supported ICBT for operational overlap between quality of life and disability measures
anxiety disorders. in the anxiety disorder literature (Mogotsi 2000), we included
both outcomes within the same meta-analysis. However, given
Potential biases in the review process research suggesting that these concepts are overlapping but also
distinct (Hambrick 2003), it may be that some variability in the
Given the variability within ICBT interventions, it is possible that
impact of treatment on these measures was missed through their
there are several biases inherent in the present review. First,
amalgamation. Future studies on ICBT should consider assessing
we elected to include only those interventions that did not
both quality of life and disability as separate treatment outcomes.
include face-to-face therapist contact during active treatment. This
may have excluded studies that were simply conducted within Finally, it is necessary to note that our method of statistical
therapists' offices for practical purposes, and only included brief analysis may have introduced some bias into the results. In
therapist interaction but could in fact have been followed online combining multiple measures within one study that assessed the
by a client at home as well. In this way, the included studies may same outcome (for example, combining several measures of panic
not comprehensively include all possible ICBT treatments. Second, symptoms into one mean and standard deviation) we made use of a
we included only interventions with active therapist involvement. method described by Borenstein 2009 that requires the availability
This decision was made because (a) there seems to be an important of bivariate correlations between the study measures in order
distinction between guided and unguided treatment, and (b) some for them to be combined. In four studies in the present review,
prior research has suggested that therapist involvement may be these correlations were not available. In these situations we simply
an important part of distance treatment (Spek 2007; Andersson excluded the measure in question from the overall mean and
2009b). Nevertheless, this decision impacted the types of trials measure of variance. The general concordance between each of the
included in the present review and led to the exclusion of some symptom measures within each study (that is, a series of symptom
Internet-based studies that did not directly involve therapists but measures tended to show similar direction and degree of change
included interactive voice response software (Greist 2002). While from pre- to post-treatment) means that the exclusion of one
preliminary work has been done, further research will need to measure should not significantly impact the results. However, this
investigate the importance of active versus automated versus no process may have introduced some small degree of bias into the
therapist involvement in ICBT. findings.
Another potential bias in the review process may have been
Agreements and disagreements with other studies or
introduced as we elected to include only those studies in which
participants were identified as meeting diagnostic criteria for an
reviews
anxiety disorder, as determined by a validated measure. While A number of prior meta-analyses have investigated the efficacy
this is good practice for the empirical validity of the present of ICBT. These meta-analyses have ranged from a quite broad
review, it may not accurately reflect clinical practice. It is likely scope investigating the efficacy of Internet interventions for any
that as a part of regular clinical practice, clients with subclinical health problems (d = 0.53; Barak 2008) to a more focused scope
diagnoses might be assigned to pursue ICBT. We might assume investigating the efficacy of ICBT for clinical and subclinical anxiety
that these treatments would be as effective for individuals with and mood symptoms (d = 0.96; Spek 2007) or anxiety and mood
subclinical symptom patterns as they are for those with diagnosed disorders in Sweden (d = 0.91; Andersson 2007). Taken together,
disorders (for example, Spek 2007); however, our exclusion of these these reviews support the present findings that ICBT is efficacious
populations prevents any firm statements in this regard. in reducing anxiety symptoms as compared to control interventions
(for example, waiting list control). Within their broad meta-analysis,
The present review is also potentially biased in the way that Barak 2008 found that interventions designed to treat PTSD and
we have measured one of our primary outcomes, clinically those targeting PD showed the largest effect sizes (g = 0.88 and
important improvement in anxiety. This outcome would possibly 0.80, respectively). Spek 2007 found that those interventions that
be more clinically useful had it been narrowed to assess diagnostic included therapist contact, as opposed to those that did not,
remission, in particular, or divided into two outcomes assessing showed a particularly large effect size (d = 1.00). It should be
remission and recovery separately. This issue is larger than noted that some concerns were raised about the methodological
Informed decisions.
quality of the studies included in these types of reviews given their AUTHORS' CONCLUSIONS
small sample sizes, the absence of details about randomisation and
treatment allocation methods, and lack of adequate information Implications for practice
about treatment compliance and credibility (Postel 2008).
The present findings, in consideration of the quality and quantity
Recently, Mayo-Wilson 2013 completed a review of media-delivered of the included studies, suggest that therapist-supported ICBT is
self-help BT and CBT for anxiety disorders. Within their review they more efficacious in treating anxiety disorders among adults than
included ICBT studies delivered both with and without therapist a waiting list, attention, information, or online discussion group
contact. In line with the present findings, their review suggested only control. The evidence also suggests that therapist-supported
that media-delivered self-help BT and CBT were more efficacious ICBT may not differ from unguided ICBT in efficacy; however,
than no treatment (that is, a waiting list control). In contrast to this evidence is significantly limited by a lack of studies in this
the present findings, their review resulted in some suggestion that comparison and must be interpreted with caution. In addition,
media-delivered self-help BT and CBT were somewhat inferior to findings suggest that therapist-supported ICBT may not result in
face-to-face CBT with the conclusion that for those who can access significantly different anxiety outcomes as compared to face-to-
it, face-to-face CBT is probably superior. The differences between face CBT. Face-to-face CBT is currently the intervention of choice
these findings and those in the present review may be due in for the treatment of anxiety disorders (Bisson 2007; Hunot 2007;
part to the differences in therapist involvement between included Norton 2007; Stewart 2009).
studies across the two reviews. Therapist involvement in media-
Our results, in conjunction with the findings of prior meta-analyses
delivered treatments, such as ICBT, may lead the treatments to be
(for example, Cuijpers 2009; Andrews 2010), clearly support further
more similar in efficacy to face-to-face CBT than those interventions
research and development of this technology and type of treatment
without therapist support.
delivery. The benefit of Internet delivery is in its ability to extend
In addition, several meta-analyses have investigated the effects treatment to individuals who may not be able to access treatment
of computer-based psychotherapy for mental health problems through traditional means. It is evidently a promising method to
more broadly, the results of which are also in accordance overcome obstacles to treatment delivery.
with the present findings. In a meta-analysis of computer-
One important next step for this field is to extend research beyond
aided psychotherapy (including treatment delivered via stand-
the two laboratories that are responsible for almost all of the
alone or Internet-linked computers, smartphones, palm pilots,
studies in this area. Replication of results across research sites and
interactive voice response, and CDs or DVDs) for anxiety disorders,
groups will go far toward increasing practitioners' confidence in the
Cuijpers 2009 found that computer-aided psychotherapy was more
intervention as well as interest in exploring greater incorporation
effective than control conditions (d = 1.08) in reducing anxiety
of this type of treatment into general practice. Another important
symptoms, and computer-aided psychotherapy outcomes did not
step will be to uncover the most effective way to make this type
differ significantly from those outcomes achieved through face-
of service delivery available to potential clients. For example, it
to-face treatment. Similarly, Reger 2009 found medium to large
may be possible to administer it as an extended service through
effects sizes when comparing computer-based CBT and ICBT
regular mental health clinics, offering clients the choice to engage
to waiting list, placebo, or treatment as usual comparators in
in Internet-based or face-to-face treatment. Alternatively, ICBT may
treating anxiety. When they investigated the effects of therapist
be more easily administered through clinics or divisions of clinics
involvement on their findings, no significant differences were
devoted entirely to this type of treatment.
identified based on amount of therapist contact. Most recently,
Andrews 2010 investigated the effects of computer therapy for Given the findings of the present review, it seems timely to start to
anxiety and depression (including both computer- and Internet- think about the best ways to incorporate ICBT into clinical practice
aided treatments) as compared to control conditions and face-to- and exploring the effectiveness of these methods. Widespread
face treatment. They found computer-based therapy to be superior rollout of ICBT may not yet be warranted, but continued steps
to control for the treatment of social phobia (g = 0.92), PD (g = 0.83), toward this goal should be pursued. Internet-based programs
and GAD (g = 1.12). They also found a non-significant difference in appear to be efficacious in reducing anxiety symptoms and there
outcome between computer-based and face-to-face CBT. are many individuals in need of treatment who could benefit from
this type of delivery.
It is important to note that these latter meta-analyses looked
more broadly at methods of administering treatment via computer Implications for research
technology, including but not limited to the Internet. Moreover,
they included research into non-therapist supported interventions, The present review suggests some important directions for future
interventions administered using interactive voice response, as research. First, further research is needed into the efficacy of
well as those that included substantial face-to-face contact. ICBT for the anxiety disorders that have not yet been extensively
Nevertheless, despite the differences between these meta-analyses examined, including OCD, PTSD, and specific phobia. The fact
and our own, the overall body of research serves to add further that these disorders tend to be thought of as more complex,
evidence for the efficacy of therapist-supported ICBT in treating and rely heavily on exposure-based elements, may have deterred
anxiety disorders. researchers from translating them into an ICBT intervention.
However, given the similarities in CBT for these disorders and
CBT for the already investigated disorders (PD, social phobia, and
GAD), including other disorders with a heavy emphasis on exposure
(that is, PD), it seems possible that ICBT would also be efficacious
in treating these disorders and thus warrants investigation. The
Informed decisions.
ongoing studies outlined in this review suggest that research therapist contact did not suggest many differences from the overall
into these disorders is indeed increasing. With respect to specific pooled analysis; however, the subgroups were rather small and set
phobia, it is possible that this category of disorder has received somewhat arbitrarily. Future research into the optimal amount of
substantial attention as it is less commonly treated in clinics therapist contact would help maximize the efficacy and efficiency
because it tends to be less functionally impairing relative to other with which ICBT could be delivered.
anxiety disorders and often requires only short interventions.
An extensive collection of self-help manuals to treat specific Fourth, with respect to the assessment of study outcomes, the
phobias exist; suggesting that this type of treatment would be very inclusion of diagnostic assessment post-treatment is encouraged
amenable to an ICBT delivery and would likely lead to efficacious in future trials in this field. While all studies measured anxiety
interventions. Further research into these disorders would be an symptoms via self-report, more objective measures of participants'
important area of investigation. diagnostic profile will help in determining the clinical significance
of treatment outcomes. Fifth, this review highlighted the
Second, while research comparing therapist-supported ICBT to a limited number of studies conducting follow-up assessments of
waiting list, attention, information, or online discussion group only participants' symptoms. While the present results do not suggest a
control is substantial, studies comparing therapist-supported ICBT significant relapse in symptoms after a six month follow-up period,
to face-to-face therapy are somewhat fewer. Subgroup analyses further studies are needed.
in the present review suggest some ambiguity with respect to
the comparable efficacy of treatment between ICBT and face- Sixth, it is important to note that any adverse effects of ICBT
to-face CBT for social phobia and PD. Further studies would have not been well-examined. Evidently, this type of treatment did
help clarify this question. Moreover, the nature of the included not result in significant symptom reduction for each participant.
studies is only sufficient for us to conclude that there may not be There were also a small number of participants across studies
significant differences in treatment outcome between face-to-face who reported being dissatisfied with this type of treatment. More
CBT and ICBT with therapist support. Future equivalence trials are research is needed to better define and measure 'harms' that might
warranted to further clarify the direct comparability of ICBT with result from this type of treatment.
therapist support and face-to-face CBT for anxiety.
Finally, as suggested above, future effectiveness studies examining
Third, the importance of the therapist in ICBT remains somewhat the best way to incorporate ICBT into regular clinical practice seem
unclear. On the one hand prior work has suggested an important to be an important next step in the field.
association between therapist involvement and ICBT treatment
outcome (Spek 2007; Andersson 2009b). On the other hand the ACKNOWLEDGEMENTS
studies included in our comparison of therapist-supported ICBT
The authors would like to thank Karen Neves, Reference and
versus unguided CBT (each of which focused on social phobia)
Research Librarian at Dalhousie University, and Sarah Dawson, Trial
suggest no difference in treatment outcome between the two
Search Co-ordinator for the Cochrane Collaboration Depression,
interventions. More studies comparing therapist-supported ICBT
Anxiety, and Neurosis Group (CCDAN), for their help in developing
versus unguided CBT are needed to clarify the role of the therapist.
and conducting the search strategy for the present review. We
Moreover, if therapist contact is important, the amount of contact
would also like to thank Leah Jones for her administrative and
that would optimize treatment outcome as well as the use of
research assistance with the review process. Finally, we would like
resources has yet to be determined. Each of the included studies
to thank the CCDAN editorial team, particularly Chris Champion and
in this review employed various amounts of therapist contact
Jessica Sharp, for their advice and assistance.
in delivering ICBT. Subgroup analyses based on the amount of
Informed decisions.
REFERENCES
References to studies included in this review Caspar F. A randomized controlled trial of a web-
based treatment for social phobia [ISRCTN62304985].
Andersson 2009 {published data only (unpublished sought but not
ControlledTrials.com [www.controlled-trials.com] 2007.
used)}
Andersson G, Jonsson U, Malmaeus F, Carlbring P, Waara J, Berger 2011 {published data only}
Ost LG. Treatment of specific phobia delivered via the Berger T, Caspar F, Richardson R, Kneubuhler B, Sutter D,
internet vs. one-session exposure treatment. 32nd Congress Andersson G. Internet-based treatment of social phobia: A
of the British Association for Behavioural and Cognitive randomized controlled trial comparing unguided with two
Psychotherapist (jointly with the European Association of types of guided self-help. Behaviour Research and Therapy
Behavioural and Cognitive Therapies). Manchester, UK, 2011;49(3):158-69. [DOI: 10.1016/j.brat.2010.12.007]
September 2004:166.
Bergstrom 2010 {published data only (unpublished sought but not
* Andersson G, Waara J, Jonsson U, Malmaeus F, Carlbring P, used)}
Ost, L. Internet-based self-help versus one-session exposure
* Bergstrom J, Andersson G, Ljotsson B, Ruck C, Andreewitch S,
in the treatment of spider phobia: A randomized controlled
Karlsson A, et al. Internet-versus group-administered
trial. Cognitive Behaviour Therapy 2009;38(2):114-20. [DOI:
cognitive behaviour therapy for panic disorder in a psychiatric
10.1080/16506070902931326]
setting: A randomised trial. BMC Psychiatry 2010;10:54. [DOI:
Andersson 2012a {published data only} 10.1186/1471-244X-10-54]
* Andersson G, Carlbring P, Furmark T, on behalf of the El Alaoui S, Hedman E, Ljotsson B, Bergstrom J, Andersson E,
SOFIE Research Group. Therapist experience and knowledge Ruck C, et al. Predictors and moderators of internet- and group-
acquisition in Internet-delivered CBT for social anxiety disorder: based cognitive beahviour therapy for panic disorder. PLoS One
A randomized controlled trial. PLoS ONE 2012;7:e37411. [DOI: 2013;8:e79024. [DOI: 10.1371/journal.pone.0079024]
10.1371/journal.pone.0037411]
Lindefors N. Internet- versus group-administered cognitive
Andersson G, Paxling B, Wiwe M, Vernmark K, Felix CB, behavior therapy for panic disorder [NCT00845260].
Lundborg L, et al. Therapeutic alliance in guided internet- ClinicalTrials.gov [www.clinicaltrials.gov] 2009.
delivered cognitive behavioural treatment of depression,
generalized anxiety disorder and social anxiety disorder. Lonsdorf TB, Ruck C, Bergstrom J, Andersson G, Ohman A,
Behaviour Research and Therapy 2012;50:544-50. [DOI: 10.1016/ Lindefors N, et al. The COMTval158met polymorphism is
j.brat.2012.05.003] associated with symptoms relief during exposure-based
cognitive-behavioral treatment in panic disorder. BMC
Furmark T. Treating social phobia via the Internet: Genetic Psychiatry 2010;10:99. [DOI: 10.1186/1471-244X-10-99]
influences on the response to Internet-delivered cognitive-
behavior therapy [UMIN000001383]. WHO International Clinical Carlbring 2001 {published and unpublished data}
Trials Registry Platform [www.who.int/ictrp] 2008. Carlbring P, Westling BE, Ljungstrand P, Ekselius L, Andersson G.
Treatment of panic disorder via the internet: A randomized trial
Andersson 2012b {published data only}
of a self-help program. Behavior Therapy 2001;32(4):751-64.
Andersson G. Psychodynamic internet treatment versus [DOI: 10.1016/S0005-7894(01)80019-8]
cognitive behavioral therapy (CBT) for generalized
anxiety disorder [NCT01312116]. ClinicalTrials.gov Carlbring 2005 {published data only}
[www.clinicaltrials.gov] 2011. Andersson G, Carlbring P, Grimlund A. Predicting treatment
outcome in internet versus face to face treatment of panic
* Andersson G, Paxling B, Roch-Norlund P, Ostman G, Norgren A,
disorder. Computers in Human Behavior 2008;24:1790-801. [DOI:
Almlov J, et al. Internet-based psychodynamic versus cognitive
10.1016/j.chb.2008.02.003]
behavioral guided self-help for generalized anxiety disorder: A
randomized controlled trial. Psychotherapy and Psychosomatics * Carlbring P, Nilsson-Ihrfelt E, Waara J, Kollenstam C,
2012;81:344-55. [DOI: 10.1159/000339371] Buhrman M, Kaldo V, et al. Treatment of panic disorder: Live
therapy vs. self-help via the internet. Behaviour Research and
Berger 2009 {published data only}
Therapy 2005;43(10):1321-33. [DOI: 10.1016/j.brat.2004.10.002]
* Berger T, Hohl E, Caspar F. Internet-based treatment for
social phobia: A randomized controlled trial. Journal of Clinical Carlbring P, Nilsson-Ihrfelt E, Waara J, Kollenstam C, Burman M,
Psychology 2009;65(10):1021-35. [DOI: 10.1002/jclp.20603] Kaldo V, et al. Treatment of panic disorder: Live therapy vs.
self-help via the Internet. Congress of the British Association
Berger T, Hohl E, Casper F. Internet-based treatment for social for Behavioural and Cognitive Psychotherapies (jointly with
phobia: A 6-month follow-up [Internetbasierte therapie der the European Association of Behavioural and Cognitive and
sozialen phobie: Ergebnisse einer 6-monate-katamnese]. Therapies). Manchester, UK, September 2004:166.
Zeitschrift fur Klinische Psychologie und Psychotherapie:
Forschung und Praxis 2010;39:217-21. [DOI: 10.1026/1616-3443/
a000050]
Informed decisions.
Carlbring 2006 {published data only (unpublished sought but not Lindefors N. Five year follow-up of internet-based cognitive
used)} behaviour therapy (CBT) for social anxiety disorder (SAD)
Almlov J, Carlbring P, Kallqvist K, Paxling B, Cuijpers P, (SOFIE-5YFU). ClinicalTrials.gov [www.clinicaltrials.gov] 2010.
Andersson G. Therapist effects in guided internet-delivered CBT
Nordgreen T, Havik OE, Ost LG, Furmark T, Carlbring P,
for anxiety disorders. Behavioural and Cognitive Psychotherapy
Andersson G. Outcome predictors in guided and unguided self-
2011;39:311-22. [DOI: 10.1017/S135246581000069X]
help for social anxiety disorder. Behaviour Research and Therapy
* Carlbring P, Bohman S, Brunt S, Buhrman M, Westling BE, 2012;50:13-21. [DOI: 10.1016/j.brat.2011.10.009]
Ekselius L, et al. Remote treatment of panic disorder: A
Furmark 2009b {published data only}
randomized trial of internet-based cognitive behavior
therapy supplemented with telephone calls. American * Furmark T, Carlbring P, Hedman E, Sonnenstein A,
Journal of Psychiatry 2006;163(12):2119-25. [DOI: 10.1176/ Clevberger P, Bohman B, et al. Guided and unguided self-
appi.ajp.163.12.2119] help for social anxiety disorder: Randomised controlled trial.
British Journal of Psychiatry 2009;195(5):440-7. [DOI: 10.1192/
Nordgreen T, Havik OE, Ost LG, Furmark T, Carlbring P, bjp.bp.108.060996]
help for social anxiety disorder. Behaviour Research and Therapy Hedman E, Furmark T, Carlbring P, Ljótsson B, Rück C,
2012;50:13-21. [DOI: 10.1016/j.brat.2011.10.009] Lindefors N, et al. A 5-year follow-up of internet-based cognitive
behavior therapy for social anxiety disorder. Journal of Medical
Carlbring 2007 {published data only (unpublished sought but not Internet Research 2011;13:e39. [DOI: 10.2196/jmir.1776]
used)}
Lindefors N. Five year follow-up of internet-based cognitive
Carlbring P, Bergman Nordgren L, Furmark T, Andersson G.
behaviour therapy (CBT) for social anxiety disorder
Long-term outcome of Internet-delivered cognitive-behavioural
(SAD) (SOFIE-5YFU) [NCT01145690]. ClinicalTrials.gov
therapy for social phobia: A 30-month follow-up. Behaviour
[www.clinicaltrials.gov] 2010.
Research and Therapy 2009;47:848-50. [DOI: 10.1016/
j.brat.2009.06.012] Nordgreen T, Havik OE, Ost LG, Furmark T, Carlbring P,
Carlbring P, Bohman S, Brunt S, Buhrman M, Westling BE,
help for social anxiety disorder. Behaviour Research and Therapy
Ekselius L, et al. Remote treatment of panic disorder: A
2012;50:13-21. [DOI: 10.1016/j.brat.2011.10.009]
randomized trial of internet-based cognitive behavior
therapy supplemented with telephone calls. American Hedman 2011 {published data only}
Journal of Psychiatry 2006;163:2119-25. [DOI: 10.1176/
Hedman E, Andersson E, Ljotsson B, Andersson G, Ruck C,
appi.ajp.163.12.2119]
Lindefors N. Cost-effectiveness of Internet-based cognitive
* Carlbring P, Gunnarsdottir M, Hedensjo L, Andersson G, behavior therapy vs. cognitive behavioral group therapy for
Ekselius L, Furmark T. Treatment of social phobia: Randomised social anxiety disorder: Results from a randomised controlled
trial of internet-delivered cognitive-behavioural therapy trial. Behaviour Research and Therapy 2011;49:729-36. [DOI:
with telephone support. British Journal of Psychiatry 10.1016/j.brat.2011.07.009]
2007;190(2):123-8. [DOI: 10.1192/bjp.bp.105.020107]
* Hedman E, Andersson G, Ljotsson B, Andersson E, Ruck C,
Nordgreen T, Havik OE, Ost LG, Furmark T, Carlbring P, Mortberg E, et al. Internet-based cognitive behavior therapy
Andersson G. Outcome predictors in guided and unguided self- vs. cognitive behavioral group therapy for social anxiety
help for social anxiety disorder. Behaviour Research and Therapy disorder: A randomized controlled non-inferiority trial. PLoS
2012;50:13-21. [DOI: 10.1016/j.brat.2011.10.009] ONE 2011;6(3):e18001. [DOI: 10.1371/journal.pone.0018001]
Carlbring 2011 {published data only} Hedman E, Ljotsson B, Andersson G, Andersson E, Schalling M,
Lindefors N, et al. Clinical and genetic outcome determinants
Carlbring P, Maurin L, Torngren C, Linna E, Eriksson T,
of Internet- and group-based cognitive behavior therapy
Sparthan E, et al. Individually-tailored, internet-based
for social anxiety disorder. Acta Psychiatrica Scandinavia
treatment for anxiety disorders: A randomized controlled
2012;126:126-36. [DOI: 10.1111/j.1600-0447.2012.01834.x]
trial. Behaviour Research and Therapy 2011;49(1):18-24. [DOI:
10.1016/j.brat.2010.10.002] Lindefors N. A comparison between internet therapy and
group therapy for social phobia - A trial using cognitive
Furmark 2009a {published data only}
behavioural therapy [NCT00564967]. ClinicalTrials.gov
* Furmark T, Carlbring P, Hedman E, Sonnenstein A, [www.clinicaltrials.gov] 2007.
Clevberger P, Bohman B, et al. Guided and unguided self-
help for social anxiety disorder: Randomised controlled trial. Johnston 2011 {published data only}
British Journal of Psychiatry 2009;195(5):440-7. [DOI: 10.1192/ * Johnston L, Titov N, Andrews G, Spence J, Dear BF. A RCT
bjp.bp.108.060996] of a transdiagnostic internet-delivered treatment for three
anxiety disorders: Examination of support roles and disorder-
Hedman E, Furmark T, Carlbring P, Ljótsson B, Rück C,
specific outcomes. PLoS ONE 2011;6:e28079. [DOI: 10.1371/
Lindefors N, et al. A 5-year follow-up of internet-based cognitive
journal.pone.0028079]
behavior therapy for social anxiety disorder. Journal of Medical
Internet Research 2011;13:e39. [DOI: 10.2196/jmir.1776]
Informed decisions.
Titov N. Internet-based treatment of generalized anxiety exploring the role of a clinician [ACTRN12609000563268].
disorder, and/or social phobia, and/or panic disorder (the Australian New Zealand Clinical Trials Registry
Anxiety program): A randomized controlled study exploring [www.anzctr.org.au] 2009.
the role of a clinician [ACTRN12610000242022]. Australian New
Zealand Clinical Trials Registry [ www.anzctr.org.au] 2010, Silfvernagel 2012 {published data only}
March. Silfvernagel K, Carlbring P, Kabo J, Edström S, Eriksson J,
Månson L, et al. Individually tailored Internet-based treatment
Kiropoulos 2008 {published data only (unpublished sought but not for young adults and adults with panic attacks: Randomized
used)} controlled trial. Journal of Medical Internet Research
* Kiropoulos LA, Klein B, Austin DW, Gilson K, Pier C, Mitchell J, 2012;14:e65. [DOI: 10.2196/jmir.1853]
et al. Is internet-based CBT for panic disorder and agoraphobia
as effective as face-to-face CBT?. Journal of Anxiety Disorders Spence 2011 {published data only}
2008;22(8):1273-84. [DOI: 10.1016/j.janxdis.2008.01.008] Spence J, Titov N, Dear BF, Johnston L, Solley K, Lorian C, et
al. Randomized controlled trial of internet-delivered cognitive
Kiropoulos LA, Klein B, Austin DW, Pier C, Mitchell J. Is internet- behavioral therapy for posttraumatic stress disorder. Depression
based CBT for panic disorder and agoraphobia as effective as and Anxiety 2011;28:541-50. [DOI: 10.1002/da.20835]
face-to-face CBT?. Australian Association for Cognitive and
Behaviour Therapy Annual Conference. Manly, Australia, 2006. Tillfors 2008 {published data only (unpublished sought but not
used)}
Paxling 2011 {published data only}
Tillfors M, Carlbring P, Furmark T, Lewenhaupt S, Spak M,
Andersson G, Paxling B, Wiwe M, Vernmark K, Felix CB, Eriksson A, et al. Treating university students with social phobia
Lundborg L, et al. Therapeutic alliance in guided internet- and public speaking fears: Internet delivered self-help with or
delivered cognitive behavioural treatment of depression, without live group exposure sessions. Depression and Anxiety
generalized anxiety disorder and social anxiety disorder. 2008;25(8):708-17. [DOI: 10.1002/da.20416]
Behaviour Research and Therapy 2012;50:544-50. [DOI: 10.1016/
j.brat.2012.05.003] Titov 2008a {published data only (unpublished sought but not
used)}
Paxling B. A randomized clinical trial of internet-delivered
Titov N. A randomized controlled study on the effect of Internet
cognitive behaviour therapy for generalized anxiety disorder
based education for treating social phobia using cognitive
[UMIN000001353]. WHO International Clinical Trials Registry
behavioural techniques [ACTRN12607000235404]. Australian
Platform [www.who.int/ictrp] 2008.
New Zealand Clinical Trials Registry [www.anzctr.org.au] May
* Paxling B, Almlov J, Dahlin M, Carlbring P, Breitholtz E, 2007.
Eriksson T, et al. Guided internet-delivered cognitive behavior
Titov N, Andrews G, Johnston L, Schwencke G, Choi I. Shyness
therapy for generalized anxiety disorder: A randomized
programme: Longer term benefits, cost-effectiveness, and
controlled trial. Cognitive Behaviour Therapy 2011;40:159-73.
acceptability. Australian and New Zealand Journal of Psychiatry
[DOI: 10.1080/16506073.2011.576699]
2009;43:36-44. [DOI: 10.1080/00048670802534424]
Paxling B, Lundgren S, Norman A, Almlov J, & Carlbring
* Titov N, Andrews G, Schwencke G, Drobny J, Einstein D.
P. Therapist behaviours in internet-delivered cognitive
Shyness 1: Distance treatment of social phobia over the
behaviour therapy: Analyses of e-mail correspondence in
internet. Australian and New Zealand Journal of Psychiatry
the treatment of generalized anxiety disorder. Behavioural
2008;42(7):585-94. [DOI: 10.1080/00048670802119762]
and Cognitive Psychotherapy 2013;41:280-9. [DOI: 10.1017/
S1352465812000240] Titov N, Gibson M, Andrews G, McEvoy P. Internet treatment
for social phobia reduces comorbidity. Australian and
Richards 2006 {published data only (unpublished sought but not
New Zealand Journal of Psychiatry 2009;43:754-9. [DOI:
used)}
10.1080/00048670903001992]
Richards JC, Klein B, Austin DW. Internet cognitive
behavioural therapy for panic disorder: Does the inclusion Titov 2008b {published data only (unpublished sought but not
of stress management information improve end-state used)}
functioning?. Clinical Psychologist 2006;10(1):2-15. [DOI: Titov N. A randomized controlled study on the effect of internet
10.1080/13284200500378795] based education for treating social phobia using cognitive
behavioural techniques [ACTRN12607000235404]. Australian
Robinson 2010 {published data only (unpublished sought but not
New Zealand Clinical Trials Registry [www.anzctr.org.au] 2007.
used)}
* Robinson E, Titov N, Andrews G, McIntryre K, Schwencke G, Titov N, Andrews G, Johnston L, Schwenke G, Choi I. Shyness
Solley K. Internet treatment for generalized anxiety disorder: A programme: Longer term benefits, cost-effectiveness, and
randomized controlled trial comparing clinician vs. technician acceptability. Australian and New Zealand Journal of Psychiatry
assistance. PLoS ONE 2010;5(6):e10942. [DOI: 10.1371/ 2009;43:36-44. [DOI: 10.1080/00048670802534424]
journal.pone.0010942]
* Titov N, Andrews G, Schwencke G. Shyness 2: Treating social
Titov N. Internet-based treatment of generalized anxiety phobia online: Replication and extension. Australian and
disorder (the Worry program): A randomized controlled study
Informed decisions.
New Zealand Journal of Psychiatry 2008;42(7):595-605. [DOI: Titov 2010 Panic {published data only}
10.1080/00048670802119820] * Titov N, Andrews G, Johnston L, Robinson E, Spence J.
Transdiagnostic internet treatment for anxiety disorders: A
Titov N, Gibson M, Andrews G, McEvoy P. Internet treatment
randomized controlled trial. Behaviour Research and Therapy
for social phobia reduces comorbidity. Australian and
2010;48:890-9. [DOI: 10.1016/j.brat.2010.05.014]
10.1080/00048670903001992] Titov 2010 Social Phobia {published data only}
Titov 2008c {published and unpublished data} Titov N, Andrews G, Johnston L, Robinson E, Spence J.
Transdiagnostic internet treatment for anxiety disorders: A
Titov N. A randomized controlled study on the role of
therapist in an Internet-based treatment program for
2010;48:890-9. [DOI: 10.1016/j.brat.2010.05.014]
social phobia using cognitive behavioural techniques
[ACTRN12608000192381]. Australian New Zealand Clinical Trials Titov 2011 {published data only}
Registry [www.anzctr.org.au] 2008.
Dear BF, Titov N, Schwencke G, Andrews G, Johnston L,
* Titov N, Andrews G, Choi I, Schwencke G, Mahoney A. Shyness Craske MG, et al. An open trial of a brief transdiagnostic internet
3: Randomized controlled trial of guided versus unguided treatment for anxiety and depression [ACTRN12610000555055].
internet-based CBT for social phobia. Australia and New Behaviour Research and Therapy 2011;49(12):830-7.
Zealand Journal of Psychiatry 2008;42(12):1030-40. [DOI:
Titov N. The Wellbeing Program: A randomized controlled
10.1080/00048670802512107]
trial of internet-based treatment of anxiety and depression
Titov N, Andrews G, Johnston L, Schwenke G, Choi I. Shyness [ACTRN12610000247077]. Australian New Zealand Clinical Trials
programme: Longer term benefits, cost-effectiveness, and Registry [www.anzctr.org.au] 2010.
acceptability. Australian and New Zealand Journal of Psychiatry
* Titov N, Dear BF, Schwencke G, Andrews G, Johnston L,
2009;43:36-44. [DOI: 10.1080/00048670802534424]
Craske MG, et al. Transdiagnostic internet treatment for
Titov N, Gibson M, Andrews G, McEvoy P. Internet treatment anxiety and depression: A randomised controlled trial.
for social phobia reduces comorbidity. Australian and Behaviour Research and Therapy 2011;49:441-52. [DOI: 10.1016/
New Zealand Journal of Psychiatry 2009;43:754-9. [DOI: j.brat.2011.03.007]
10.1080/00048670903001992]
van Ballegooijen 2013 {published data only}
Titov 2009 {published data only (unpublished sought but not * van Ballegooijen W, Riper H, Klein B, Ebert DD, Kramer J,
used)} Meulenbeek P, Cuijpers P. An Internet-based guided self-help
Lorian CN, Titov N, Grisham JR. Changes in risk-taking over the intervention for panic symptoms: Randomized controlled trial.
course of an Internet-delivered cognitive behavioral treatment Journal of Medical Internet Research 2013;15:e154. [NTR1639]
for generalized anxiety disorder. Journal of Anxiety Disorders
van Ballegooijen W, Riper H, van Straten A, Kramer J, Conijn B,
2012;26:140-9. [DOI: 10.1016/j.janxdis.2011.10.003]
Cuijpers P. The effects of an Internet based self-help course for
Titov N. Internet-based education for generalized anxiety reducing panic symptoms - Don't Panic Online: Study protocol
disorder (the Worry Program): A randomized controlled trial for a randomised controlled trial. Trials 2011;12:75. [DOI:
[ACTRN12609000136202]. Australian New Zealand Clinical Trials 10.1186/1745-6215-12-75]
Registry [www.anzctr.org.au] 2009.
Wims 2010 {published data only (unpublished sought but not
* Titov N, Andrews G, Robinson E, Schwencke G, used)}
Johnston L, Solley K, et al. Clinician-assisted internet- Wims E. The effectiveness of clinician assisted internet
based treatment is effective for generalized anxiety based cognitive behaviour therapy for panic disorder
disorder: Randomized controlled trial. Australian and New [ACTRN12608000548336]. Australian New Zealand Clinical Trials
Zealand Journal of Psychiatry 2009;43(10):905-12. [DOI: Registry [www.anzctr.org.au] 2008, October.
10.1080/00048670903179269]
* Wims E, Titov N, Andrews G, Choi I. Clinician-assisted internet-
Titov 2010 {published data only} based treatment is effective for panic: A randomized controlled
* Titov N, Andrews G, Johnston L, Robinson E, Spence J. trial. Australian and New Zealand Journal of Psychiatry
Transdiagnostic internet treatment for anxiety disorders: A 2010;44(7):599-607. [DOI: 10.3109/00048671003614171]
2010;48(9):890-9. [DOI: 10.1016/j.brat.2010.05.014]
References to studies excluded from this review
Titov 2010 GAD {published data only} Andersson 2006 {published data only}
Titov N, Andrews G, Johnston L, Robinson E, Spence J. Andersson G, Carlbring P, Holmström A, Sparthan E, Furmark T,
Transdiagnostic internet treatment for anxiety disorders: A Nilsson-Ihrfelt E, et al. Internet-based self-help with therapist
randomized controlled trial. Behaviour Research and Therapy feedback and in vivo group exposure for social phobia: A
2010;48:890-9. [DOI: 10.1016/j.brat.2010.05.014] randomized controlled trial. Journal of Consulting and Clinical
Psychology 2006;74:677-86. [DOI: 10.1037/0022-006X.74.4.677]
Informed decisions.
Andersson 2012c {published data only} Febbraro 2005 {published data only}
* Andersson E, Enander J, Andren P, Hedman E, Ljotsson B, Febbraro GAR. An investigation into the effectiveness of
Hursti T, et al. Internet-based cognitive behaviour therapy bibliotherapy and minimal contact interventions in the
for obsessive-compulsive disorder: A randomized controlled treatment of panic attacks. Journal of Clinical Psychology
trial. Psychological Medicine 2012;42:2193-203. [DOI: 10.1017/ 2005;61:763-79. [DOI: 10.1002/jclp.20097]
S0033291712000244]
Gilson 2006 {published data only}
Ruck C. Internet-based cognitive behavior therapy (CBT) Gilson K, Pier C, Austin DW, Mitchell J, Schattner P, Pierce D, et
for obsessive compulsive disorder (OCD) [NCT01347099]. al. Evaluation of internet-based cognitive behavioural therapy
ClinicalTrials.gov [www.clinicaltrials.gov] 2011. for panic disorder in general medical practice. 29th Australian
Association for Cognitive and Behaviour Therapy Annual
Andrews 2011 {published data only}
Conference. Manly, Australia, October 2006.
Andrews G. In patients seeking treatment for social phobia
internet cognitive behaviour therapy will be compared with Greist 2002 {published data only}
face to face cognitive behaviour therapy on the reduction in Greist JH, Marks IM, Baer L, Kobak KA, Wenzel KW, Hirsch MJ, et
social phobia scores [ACTRN12611000626965]. Australian New al. Behavior therapy for obsessive-compulsive disorder guided
Zealand Clinical Trials Registry [www.anzctr.org.au] June 2011. by a computer or by a clinician compared with relaxation as
a control. Journal of Clinical Psychiatry 2002;63:138-45. [DOI:
* Andrews G, Davies M, Titov N. Effectiveness randomized
10.4088/JCP.v63n0209]
controlled trial of face to face versus Internet cognitive
behaviour therapy for social phobia. Australian and Kenardy 2003 {published data only}
Kenardy JA, Dow MGT, Johnston DW, Newman MG, Thomson A,
10.3109/00048674.2010.538840]
Taylor CB. A comparison of delivery methods of cognitive-
Bell 2012 {published data only} behavioral therapy for panic disorder: An international
multicenter trial. Journal of Consulting and Clinical Psychology
Bell CJ, Colhoun HC, Carter FA, Frampton CM. Effectiveness of
2003;71:1068-75. [DOI: 10.1037/0022-006X.71.6.1068]
computerised cognitive behaviour therapy for anxiety disorders
in secondary care. Australian and New Zealand Journal of Kenwright 2005 {published data only}
Psychiatry 2012;46:630-40. [DOI: 10.1177/0004867412437345]
Kenwright M, Marks I, Graham C, Franses A, Mataix-Cols D.
Carlbring 2003 {published data only} Brief scheduled phone support from a clinician to enhance
computer-aided self-help for obsessive-compulsive disorder:
Carlbring P, Ekselius L, Andersson G. Treatment of panic
Randomized controlled trial. Journal of Clinical Psychology
disorder via the internet: A randomized trial of CBT vs. applied
2005;61:499-508. [DOI: 10.1002/jclp.20204]
relaxation. Journal of Behavior Therapy 2003;34(2):129-40. [DOI:
10.1016/S0005-7916(03)00026-0] Klein 2001 {published data only}
Carlbring 2010 {published data only} Klein B, Richards JC. A brief internet-based treatment for
panic disorder. Behavioural and Cognitive Psychotherapy
Carlbring P. Treatment of social phobia over the
2001;29:113-7. [DOI: 10.1017/S1352465801001138]
internet (SOFIE-8) [NCT01312571]. ClinicalTrials.gov
[www.clinicaltrials.gov] 2010, August. Klein 2006 {published data only}
Carlbring 2011b {published data only} Klein B, Richards JC, Austin DW. Efficacy of internet therapy for
panic disorder. Journal of Behavior Therapy and Experimental
Carlbring P, Maurin T, Sjomark J, Maurin L, Westling BE,
Psychiatry 2006;37:213-38. [DOI: 10.1016/j.jbtep.2005.07.001]
Ekselius L, et al. All at once or one at a time? A randomized
controlled trial comparing two ways to deliver bibliotherapy for Klein 2009 {published data only}
panic disorder. Cognitive Behaviour Therapy 2011;40:228-35.
Klein B, Austin D, Pier C, Kiropoulos L, Shandley K, Mitchell J, et
[DOI: 10.1080/16506073.2011.553629]
al. Internet-based treatment for panic disorder: Does frequency
Cunningham 2006 {published data only} of therapist contact make a difference?. Cognitive Behaviour
Therapy 2009;38:100-13. [DOI: 10.1080/16506070802561132]
Cunningham V, Lefkoe M, Sechrest L. Eliminating fears: An
intervention that permanently eliminates the fear of public Knaevelsrud 2007 {published data only}
speaking. Clinical Psychology & Psychotherapy 2006;13:183-93.
Knaevelsrud C, Maercker A. Internet-based treatment for PTSD
[DOI: 10.1002/cpp.487]
reduces distress and facilitates the development of a strong
Ellis 2011 {published data only} therapeutic alliance: A randomized controlled clinical trial. BMC
Psychiatry 2007;7:13. [DOI: 10.1186/1471-244X-7-13]
Ellis LA, Campbell AJ, Sethi S, O'Dea BM. Comparative
randomized trial of an online cognitive-behavioral therapy Lange 2001 {published data only}
program and an online support group for depression
Lange A, van de Ven JP, Schrieken B, Emmelkamp PM.
and anxiety. Journal of CyberTherapy and Rehabilitation
Interapy. Treatment of posttraumatic stress through the
2011;4:461-7.
Internet: A controlled trial. Journal of Behavior Therapy
Informed decisions.
and Experimental Psychiatry 2001;32:73-90. [DOI: 10.1016/ Shandley 2008 {published data only}
S0005-7916(01)00023-4] Shandley K, Austin DW, Klein B, Pier C, Schattner P, Pierce D,
et al. Therapist-assisted, Internet-based treatment for panic
Lange 2003 {published data only}
disorder: Can general practitioners achieve comparable
Lange A, Rietdijk D, Hudcovicova M, van de Ven JP, Schrieken B, patient outcomes to psychologists?. Journal of Medical Internet
Emmelkamp PM. Interapy: A controlled randomized trial of Research 2008;10:65-79. [DOI: 10.2196/jmir.1033]
the standardized treatment of posttraumatic stress through
the Internet. Journal of Consulting and Clinical Psychology Titov 2009b {published data only}
2003;71:901-9. [DOI: 10.1037/0022-006X.71.5.901] Titov N, Andrews G, Schwencke G, Solley K, Johnston L,
Robinson E. An RCT comparing effect of two types of support
Litz 2007 {published data only}
on severity of symptoms for people completing Internet-based
Bruner V, Gore K, DeDeyn J, Jaffer A, Litz B, Bryant R. A cognitive behaviour therapy for social phobia. Australian
therapist-guided internet-based self-management approach to and New Zealand Journal of Psychiatry 2009;43:920-6. [DOI:
post-traumatic stress after military events. International Society 10.1080/00048670903179228]
for Traumatic Stress Studies. New Orleans, LA, November 2004.
van Straten 2008 {published data only}
Litz BT, Engel CC, Bryant RA, Papa A. A randomized, controlled
van Straten A, Cuijpers P, Smits N. Effectiveness of a web-based
proof-of-concept trial of an internet-based, therapist-assisted
self-help intervention for symptoms of depression, anxiety, and
self-management treatment for posttraumatic stress disorder.
stress: Randomized controlled trial. Journal of Medical Internet
American Journal of Psychiatry 2007;164(11):1676-83. [DOI:
Research 2008;10:e7. [DOI: 10.2196/jmir.954]
10.1176/appi.ajp.2007.06122057]
Wagner 2012 {published data only}
Marks 2004 {published data only}
Wagner B, Brand J, Schulz W, Knaevelsrud C. Online working
Marks IM, Kenwright M, McDonough M, Whittaker M, Mataix-
alliance predicts treatment outcome for posttraumatic stress
Cols D. Saving clinicians' time by delegating routine aspects of
symptoms in Arab war-traumatized patients. Depression and
therapy to a computer: A randomized controlled trial in phobia/
Anxiety 2012;29:646-51. [DOI: 10.1002/da.21962]
panic disorder. Psychological Medicine 2004;34:9-17. [DOI:
10.1017/S003329170300878X]
Newman 1997 {published data only}

References to studies awaiting assessment
Newman MG, Kenardy J, Herman S, Taylor CB. Comparison Andersson 2013 {published data only}
of palmtop-computer-assisted brief cognitive-behavioral * Andersson G, Warra J, Jonsson U, Malmaeus F, Carlbring P,
treatment to cognitive-behavioral treatment for panic disorder. Ost L-G. Internet-based exposure treatment versus one-session
Journal of Consulting and Clinical Psychology 1997;65:178-83. exposure treatment of snake phobia: A randomized controlled
[DOI: 10.1037/0022-006X.65.1.178] trial. Cognitive Behaviour Therapy 2013;42:284-91. [DOI:
10.1080/1656073.2013.844202]
Pittaway 2009 {published data only}
Pittaway S, Cupitt C, Palmer D, Arowobusoye N, Milne R, Andrews 2011b {published data only}
Holttum S, et al. Comparative, clinical feasibility study of three Andrews G. A randomised controlled trial comparing clinician-
tools for delivery of cognitive behavioural therapy for mild to assisted Internet based treatment for panic disorder with
moderate depression and anxiety provided on a self-help basis. or without agoraphobia vs. a waitlist control on severity of
Mental Health in Family Medicine 2009;6:145-54. symptoms of panic [ACTRN12611001120965]. Australian New
Zealand Clinical Trials Registry [www.anzctr.org.au] 2011.
Ruwaard 2010 {published data only}
Ruwaard J, Broeksteeg J, Schrieken B, Emmelkamp P, Lange A. Andrews 2011c {published data only}
Web-based therapist-assisted cognitive behavioral treatment Andrews G. The Wellbeing6 Program for Anxiety and Depression
of panic symptoms: A randomized controlled trial with a three- - A Randomised Controlled Trial [ACTRN12611001055998].
year follow-up. Journal of Anxiety Disorders 2010;24:387-96. Australian New Zealand Clinical Trials Registry
[DOI: 10.1016/j.janxdis.2010.01.010] [www.anzctr.org.au] October 2011.
Saul 2007 {published data only} Andrews 2011d {published data only}
Saul JE. A comparison of Internet-based versus face-to-face Andrews G. A randomized controlled trial of Internet
cognitive behavioral therapy for snake phobia. Dissertation based education for social phobia, panic disorder
Abstracts International 2007;67:6077. (with or without agoraphobia), and generalised anxiety
disorder, comparing immediate education versus delayed
Schneider 2005 {published data only} education groups on severity of symptoms of anxiety
Schneider AJ, Mataix-Cols D, Marks IM, Bachofen M. Internet- [ACTRN12611000625976]. Australian New Zealand Clinical Trials
guided self-help with or without exposure therapy for phobic Registry [www.anzctr.org.au] 2011.
and panic disorders. Psychotherapy and Psychosomatics
2005;74:154-64. [DOI: 10.1159/000084000]
Informed decisions.
Andrews 2012a {published data only} disorders for primary care patients: Study protocol for
Andrews G. A randomised controlled trial comparing clinician- a randomised controlled trial. Trials 2012;13:16. [DOI:
assisted Internet based treatment for Obsessive Compulsive 10.1186/1745-6215-13-16]
Disorder (OCD) vs. a waitlist control condition on severity of
* Nordgren LB, Hedman E, Etienne J, Bodin J, Kadowaki A,
symptoms of OCD [ACTRN12612001073897]. Australian New
Eriksson S, et al. Effectiveness and cost-effectiveness of
Zealand Clinical Trials Registry [www.anzctr.org.au] 2012.
individually tailored Internet-delivered cognitive behavior
Berger 2012 {published data only} therapy for anxiety disorders in a primary care population: A
Berger M. Internet-based treatment for obsessive-compulsive
2014;59:1-11. [DOI: 10.1016/j.brat.2014.05.007]
disorder – acceptance and effectiveness [DRKS00004612].
German Clinical Trials Register [www.germanctr.de] 2012. Richards 2014 {published data only}
Berger 2014 {published data only} Richards D. Internet delivered treatment for generalized anxiety
symptoms in students [ISRCTN16303842]. ISRCTN Registry
* Berger T, Boettcher J, Caspar F. Internet-based guided self-
[www.isrctn.com] 2013.
help for several anxiety disorders: A randomized controlled
trial comparing a tailored with a standardized disorder-specific Richards D, Timulak L, Doherty G, Sharry J, McLoughlin O,
approach. Psychotherapy 2014;51:207-19. [DOI: 10.1037/ Rashleigh C, et al. Low-intensity internet-delivered treatment
a0032527] for generalized anxiety symptoms in routine care: protocol for a
randomized controlled trial. Trials 2014;15:145.
Carlbring 2012 {published data only}
Carlbring P. Internet-based treatment of generalized Schreuders 2008 {published data only}
anxiety disorder [NCT01570374]. ClinicalTrials.gov Schreuders B. Phobias under control. Internet-based
[www.clinicaltrials.gov] 2012. guided self-help for treatment of social, agora-, or specific
phobia(s): a pilot trial [NTR1260]. Nederlands Trial Register
Greist 2012 {published data only}
[www.trialregister.nl/trialreg/] 2008.
Greist R, Kobak K. Computer assisted cognitive behavior
therapy for obsessive-compulsive disorder: A comprehensive
stepped-care approach [NCT01522287]. ClinicalTrials.gov References to ongoing studies
[www.clinicaltrials.gov] 2012.
Andrews 2012b {published data only}
Ivarsson 2014 {published data only} Andrews G. The Obsessive Compulsive Disorder (OCD)
* Ivarsson D, Blom M, Hesser H, Carlbring P, Enderby P, Program: A randomised controlled trial of online
Nordberg R, Andersson G. Guided internet-delivered cognitive versus face-to-face cognitive behavioural therapy (CBT)
behavior therapy for post-traumatic stress disorder: A [ACTRN12612001306808]. Australian New Zealand Clinical Trials
randomized controlled trial. Internet Interventions 2014;1:33-40. Registry [www.anzctr.org.au] 2012.
[DOI: 10.1016/j.invent.2014.03.002]
Bishop 2012 {published data only}
Newby 2013 {published data only} Bishop TM, Possemato K, Acosta M, Lantinga LJ, Maisto S,
Newby JM. The Wellbeing6 program for anxiety and depression: Marsch L, et al. Moving forward: Update on the development
A randomised controlled trial [ACTRN12611001055998]. of a web-based cognitive behavioral treatment for OEF/OIF
Australian New Zealand Clinical Trials Registry veterans with PTSD symptoms and substance misuse [abstract].
[www.anzctr.org.au] 2011. Alcoholism: Experimental and Clinical Research 2012;36:347A.
* Newby JM, Mackenzie A, Williams AD, McIntyre K, Watts S, Clark 2012 {published data only}
Wong N, Andrews G. Internet cognitive behavioural therapy for Clark DM. Internet-based cognitive therapy for social
mixed anxiety and depression: A randomized controlled trial anxiety disorder [ISRCTN95458747]. Controlled Trials.com
and evidence of effectiveness in primary care. Psychological [www.controlled-trials.com] 2012.
Medicine 2013;43:2635-48. [DOI: 10.1017/S0033291713000111]
Kok 2012 {published data only}
Newby JM, Williams AD, Andrews G. Reductions in negative Kok RN, van Straten A, Beekman A, Bosmans J, de Neef M,
repetitive thinking and metacognitive beliefs during Cuijpers P. Effectiveness and cost-effectiveness of web-
transdiagnostic internet cognitive behavioural therapy (iCBT) based treatment for phobic outpatients on a waiting list for
for mixed anxiety and depression. Behaviour Research and psychotherapy: protocol of a randomised controlled trial. BMC
Therapy 2014;59:52-60. [DOI: 10.1016/j.brat.2014.05.009] Psychiatry 2012;12:131. [DOI: 10.1186/1471-244X-12-131]
Nordgren 2012 {published data only} Lindner 2013 {published data only}
Carlbring P. Internet-administrated treatment of anxiety Carlbring P. ACT-smart: Smartphone-supplemented iCBT
disorders (NOVA II) [NCT01390168]. ClinicalTrials.gov for social phobia and/or panic disorder [NCT01963806].
[www.clinicaltrials.gov] 2011. ClinicalTrials.gov 2013.
Nordgren LB, Andersson G, Kadowaki A, Carlbring P.
Tailored internet-administered treatment of anxiety
Informed decisions.
Lindner P, Ivanova E, Ly KH, Andersson G, & Carlbring P. Guided Andersson 2012

and unguided CBT for social aniety disorder and/or panic Andersson G, Carlbring P, Furmark T. Therapist experience and
disorder via the Internet and a smartphone application: Study knowledge acquisition in internet-delivered CBT for social
protocol for a randomised controlled trial. Trials 2013;14:437. anxiety disorder: a randomized controlled trial. PLoS ONE
[10.1186/1745-6215-14-437] 2012;7:e37411. [DOI: 10.1371/journal.pone.0037411]
Miclea 2014 {published data only} Andrews 2010
Miclea M. PAXonline: A randomized controlled trial assessing the Andrews G, Cuijpers P, Craske MG, McEvoy P, Titov N. Computer
efficacy of an Internet-based cognitive behavior intervention therapy for the anxiety and depressive disorders is effective,
for Panic Disorder [ACTRN12614000547640]. Australian New acceptable, and practical health care: A meta-analysis. PLoS
Zealand Clinical Trials Registry 2014. ONE 2010;5:e13196. [DOI: 10.1371/journal.pone.0013196]
Rollman 2012 {published data only} APA 1980
Rollman B. Online treatments for mood and anxiety disorders American Psychiatric Association. Diagnostic and Statistical
in primary care [NCT01482806]. ClinicalTrials.gov [http:// Manual of Mental Disorders. 3rd Edition. Washington, DC:
clinicaltrials.gov] 2011. American Psychiatric Association, 1980.
Titov 2012 {published data only} APA 1987
Titov N. A randomized controlled trial of the effects of disorder- American Psychiatric Association. Diagnostic and Statistical
specific vs. trans-diagnostic and self-guided vs. guided Internet- Manual of Mental Disorders. 3rd Edition. Washington, DC:
administered treatment on symptoms of social phobia in American Psychiatric Association, 1987.
Australian adult [ACTRN12612000430831]. Australia New
Zealand Clinical Trial Registry 2012. APA 1994
American Psychiatric Association. Diagnostic and Statistical
Tulbure 2012 {published data only}
Manual. 4th Edition. Washington, DC: American Psychiatric
Tudor TB. Internet treatment for social phobia (iSOFIE-Ro) Association, 1994.
[NCT01557894]. ClinicalTrials.gov [www.clinicaltrials.gov] 2012.
APA 2000
Tulbure BT, Mansson KNT, Andersson G. Internet treatment
American Psychiatric Association. Diagnostic and Statistical
for social anxiety disorder in Romania: Study protocol for
Manual of Mental Disorders. 4th Edition. Washington, DC:
a randomized controlled trial. Trials 2012;13:202. [DOI:
American Psychiatric Association, 2000.
10.1186/1745-6215-13-202]
Attkisson 1982
von Essen 2008 {published data only}
Attkisson CC, Zwick R. The client satisfaction questionnaire:
von Essen L. Treatment of posttraumatic stress disorder among
Psychometric properties and correlations with service
parents of children with cancer with cognitive behavioural
utilization and psychotherapy outcome. Evaluation and
therapy over the Internet. Pediatric Blood Cancer 2008;50:9.
Program Planning 1982;5(3):233-7.
Barak 2008
Additional references
Barak A, Hen L, Boniel-Nissim M, Shapira N. A comprehensive
Acarturk 2009 review and a meta-analysis of the effectiveness of internet-
Acarturk C, Smit F, de Graaf R, van Straten A, Ten Have M, based psychotherapeutic interventions. Journal of
Cuijpers P. Economic costs of social phobia: A population- Technology in Human Services 2008;26:109-60. [DOI:
based study. Journal of Affective Disorders 2009;115:421-9. [DOI: 10.1080/15228830802094429]
10.1016/j.jad.2008.10.008]
Barlow 1988
Alvidrez 1999 Barlow DH. Anxiety and its disorders: the nature and treatment
Alvidrez J, Azocar F. Distressed women's clinic patients: of anxiety and panic. New York: Guilford Press, 1988.
preferences for mental health treatments and perceived
obstacles. General Hospital Psychiatry 1999;21(5):340-7. Batelaan 2007
Batelaan N, Smit F, de Graaf R, van Balkom A, Vollebergh W,
Andersson 2007 Beekman A. Economic costs of full-blown and subthreshold
Andersson G, Cuijpers P, Carlbring P, Lindefors N. Effects of panic disorder. Journal of Affective Disorders 2007;104:127-36.
internet-delivered cognitive behaviour therapy for anxiety and [DOI: 10.1016/j.jad.2007.03.013]
mood disorders. Review Series: Psychiatry 2007;9:9-14.
Beck 1979
Andersson 2009b Beck AT, Rush AJ, Shaw BF, Emergy G. Cognitive therapy of
Andersson G. Using the internet to provide cognitive behaviour depression. New York: The Guilford Press, 1979.
therapy. Behaviour Research and Therapy 2009;47:175-80. [DOI:
10.1016/j.brat.2009.01.010]
Informed decisions.
Beck 1991 Deacon 2004

Beck A, Steer R. Beck Anxiety Inventory manual. San Antonio, Deacon BJ, Abramowitz JS. Cognitive and behavioral
TX: The Psychological Corporation, 1991. treatments for anxiety disorders: a review of meta-analytic
findings. Journal of Clinical Psychology 2004;60(4):429-41.
Beck 2005
Beck AT, Emery G. Anxiety disorders and phobias: A cognitive Department of Health 2008
perspective. New York: Basic Books, 2005. Department of Health. Improving Access to Psychological
Therapies Implementation Plan: National Guidelines for
Bee 2008 Regional Delivery. London: DoH, 2008.
Bee PE, Bower P, Lovell K, Gilbody S, Richards D, Gask L,
et al. Psychotherapy mediated by remote communication DiNardo 1994
technologies: a meta-analytic review. BMC Psychiatry 2008;8:60. DiNardo P, Brown TA, Barlow DH. Anxiety Disorders Interview
Schedule for DSM-IV. Anxiety Disorders Interview Schedule for
Bisson 2007 DSM-IV. San Antonio, TX: The Psychological Corporation, 1994.
Bisson J, Andrew M. Psychological treatment of post-traumatic
stress disorder (PTSD). Cochrane Database of Systematic Dugas 2007
Reviews 2007, Issue 3. [DOI: 10.1002/14651858.CD003388.pub3] Dugas MJ, Robichaud M. Cognitive-behavioral treatment for
generalized anxiety disorder: From science to practice. New
Borenstein 2009 York: Routledge, 2007.
Borenstein M, Hedges LV, Higgins JPT, Rothstein HR.
Introduction to meta-analysis. Hoboken, NJ: John Wiley & Sons, DuPont 1996
Ltd, 2009. DuPont RL, Rice DP, Miller LS, Shiraki SS, Rowland CR,
Harwood HJ. Economic costs of anxiety disorders. Anxiety
Bruce 2005 1996;2:167-72.
Bruce SE, Yonkers KA, Otto MW, Eisen JL, Weisberg RB,
Pagano M, et al. Influence of psychiatric comorbidity on Ehlers 2000
recovery and recurrence in generalized anxiety disorder, social Ehlers A, Clark DM. A cognitive model of posttraumatic stress
phobia, and panic disorder: a 12-year prospective study. disorder. Behaviour Research and Therapy 2000;38:319-45.
American Journal of Psychiatry 2005;162(6):1179-87.
ESEMeD/MHEDEA 2000 Investigators 2004
Buffett-Jerrot 2002 The ESEMeD/MHEDEA 2000 Investigators. Prevalence of mental
Buffett-Jerrott SE, Stewart SH. Cognitive and sedative effects disorders in Europe: results from the European Study of the
of benzodiazepine use. Current Pharmaceutical Design Epidemiology of Mental Disorders (ESEMeD) project. Acta
2002;8(1):45-58. Psychiatrica Scandinavica 2004;109 Suppl 420:21-7.
Casey 2004 First 2002

Casey LM, Oei TPS, Newcombe PA. An integrated cognitive First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical
model of panic disorder: The role of positive and negative Interview for DSM-IV-TR Axis I Disorders, Research Version,
cognitions. Clinical Psychology Review 2004;24:529-55. [DOI: Patient Edition (SCID-I/P). New York: Biometrics Research, New
10.1016/j.cpr.2004.01.005] York State Psychiatric Institute, 2002.
Choy 2007 Foa 2010

Choy Y. Managing side effects of anxiolytics. Primary Psychiatry Foa EB. Cognitive behavioral therapy of obsessive-compulsive
2007;14(7):68-76. disorder. Dialogues in Clinical Neuroscience 2010;12(2):199-207.
Clark 1986 Frisch 1992

Clark DM. A cognitive approach to panic. Behaviour Research Frisch MB, Cornell J, Villanueva M, Retzlaff PJ. Clinical validation
and Therapy 1986;24(4):461-70. of the quality of life inventory: A measure of life satisfaction
for use in treatment planning and outcome assessment.
Cuijpers 2009 Psychological Assessment 1992;4:92-101.
Cuijpers P, Marks IM, van Straten A, Cavanagh K, Gega L,
Andersson G. Computer-aided psychotherapy for anxiety Goodman 1989
disorders: A meta-analytic review. Cognitive Behaviour Therapy Goodman WK, Price LH, Rasmussen SA, Mazure C,
2009;38:66-82. [DOI: 10.1080/16506070802694776] Fleischmann RL, Hill CL, et al. The Yale-Brown Obsessive
Compulsive Scale, I: Development, use, and reliability. Archives
Cuijpers 2010 of General Psychiatry 1989;46(11):1006-11.
Cuijpers P, Donker T, van Straten A, Li J, Andersson G. Is
guided self-help as effective as face-to-face psychotherapy for Greenberg 1999
depression and anxiety disorders? A systematic review and Greenberg PE, Sisitsky T, Kessler RC, Finkelstein SN, Berndt ER,
meta-analysis of comparative outcome studies. Psychological Davidson JRT, et al. The economic burden of anxiety disorders
Medicine 2010;40:1942-57. [DOI: 10.1017/S0033291710000772]
Informed decisions.
in the 1990s. Journal of Clinical Psychiatry 1999;60:427-35. [DOI: Higgins 2011b

10.4088/JCP.v60n0702] Higgins JPT, Altman DG, Gotzsche PC, Juni P, Moher D,
Oxman AD, et al. The Cochrane Collaboration's tool for
Gureje 2006
assessing risk of bias in randomised trials. BMJ 2011;343:d5928.
Gureje O, Lasebikan VO, Kola L, Makanjuola VA. Lifetime and 12- [DOI: 10.1136/bmj.d5928]
month prevalence of mental disorders in the Nigerian Survey
of Mental Health and Well-Being. British Journal of Psychiatry Hunot 2007
2006;188(5):465-71. Hunot V, Churchill R, Teixeira V, Silva de Lima M. Psychological
therapies for generalised anxiety disorder. Cochrane
Guy 1976
Database of Systematic Reviews 2007, Issue 1. [DOI:
Guy W. ECDEU assessment manual for psychopharmacology. 10.1002/14651858.CD001848.pub4]
Rockville, MD: US Department of Health Education and Welfare,
1976. Kessler 2004
Kessler RC, McGonagle KA, Zhoa S, Nelson CB, Hughes M,
Hahlweg 2001
Eshleman S, et al. Lifetime and 12-month prevalence of DSM-
Hahlweg K, Fiegenbaum W, Frank M, Schroeder B, III-R psychiatric disorders in the United States: Results from the
von Witzleben I. Short- and long-term effectiveness of an National Comorbidity Survey. Archives of General Psychiatry
empirically supported treatment for agoraphobia. Journal 1994;51:8-19. [DOI: 10.1001/archpsyc.1994.03950010008002]
of Consulting and Clinical Psychology 2001;69:375-82. [DOI:
10.1037/0022-006X.69.3.375] Kessler 2005a
Kessler RC, Chiu WT, Demler O, Walters EE. Prevalence, severity,
Hambrick 2003
and comorbidity of 12-month DSM-IV disorders in the national
Hambrick JP, Turk CL, Heimberg RG, Schneier FR, Liebowitz MR. comorbidity survey replication. Archives of General Psychiatry
The experience of disability and quality of life in social anxiety 2005;62(6):617-27.
disorder. Depression and Anxiety 2003;18:46-50. [DOI: 10.1002/
da.10110] Kessler 2005b
Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR,
Hamilton 1959
Walters EE. Lifetime prevalence and age-of-onset distributions
Hamilton M. The assessment of anxiety states by rating. British of DSM-IV disorders in the national comorbidity survey
Journal of Medical Psychology 1959;32:50-5. replication. Archives of General Psychiatry 2005;62(6):593-602.
Hauenstein 2006 Landon 2004
Hauenstein EJ, Petterson S, Merwin E, Rovnyak V, Heise B, Landon TA, Barlow DH. Cognitive-behavioural treatment for
Wagner D. Rurality, gender, and mental health treatment. panic disorder: current status. Journal of Psychiatric Practice
Family and Community Health 2006;29(3):169-85. 2004;10(4):211-26.
Health Canada 2002 Leon 1997
Health Canada. A report on mental illness in Canada. Leon AC, Olfson M, Portera L, Farber L, Sheehan DV. Assessing
www.phac-aspc.gc.ca/publicat/miic-mmac/index.html. Ottawa, psychiatric impairment in primary care with the Sheehan
ON: Health Canada Editorial Board, (accessed 25 March 2008). Disability Scale. International Journal of Psychiatric Medicine
1997;27:93-105.
Heimberg 2002
Heimberg RG. Cognitive-behavioral therapy for social anxiety Mattick 1998
disorder: current status and future directions. Biological Mattick RP, Clarke JC. Development and validation of measures
Psychiatry 2002;51(1):101-8. of social phobia scrutiny fear and social interaction anxiety.
Behaviour Research and Therapy 1998;36(4):455-70.
Higgins 2002
Higgins J, Thompson S. Quantifying heterogeneity in a meta- Mayo-Wilson 2013
analysis. Statistical Medicine 2002;21(11):1539-58. Mayo-Wilson E, Montgomery P. Media-delivered cognitive
behavioural therapy and behavioural therapy (self-help) for
Higgins 2008
anxiety disorders in adults. Cochrane Database of Systematic
Higgins JPT, White IR, Anzures-Cabrera J. Meta-analysis of Reviews 2013, Issue 9. [DOI: 10.1002/14651858.CD005330.pub4]
skewed data: Combining results reported on log-transformed or
raw scales. Statistics in Medicine 2008;27:6072-92. [DOI: 10.1002/ McNaughton 2008
sim.3427] McNaughton N. The neurobiology of anxiety: potential for co-
morbidity of anxiety and substance use disorders. In: Stewart
Higgins 2011a
SH, Conrod PJ editor(s). Anxiety and substance use disorders.
Higgins JPT, Green S (editors). Cochrane Handbook for New York, NY: Springer, 2008:19-33.
Systematic Review of Interventions Version 5.1.0 [updated
March 2011]. The Cochrane Collaboration, 2011. Available from
www.cochranehandbook.org. The Cochrane Collaboration.
Informed decisions.
Milette 2011 Review and analysis. Clinical Psychology: Science and Practice
Milette K, Roseman M, Thombs BD. Transparency of outcome 2005;12(1):72-86.
reporting and trial registration of randomized controlled
Peters 2008
trials in top psychosomatic and behavioral health journals:
A systematic review. Journal of Psychosomatic Research Peters J, Sutton AJ, Jones DR, Abrams KR, Rushton L. Contour-
2011;70:205-17. [DOI: 10.1016/j.jpsychores.2010.09.015] enhanced meta-analysis funnel plots help distinguish
publication bias from other causes of asymmetry. Journal of
Mogotsi 2000 Clinical Epidemiology 2008;61(10):991-6.
Mogotsi M, Kaminer D, Stein DJ. Quality of life in the anxiety
Postel 2008
disorders. Harvard Review of Psychiatry 2000;8:273-82. [DOI:
10.1093/hrp/8.6.273] Postel MG, de Haan HA, De Jong CAJ. E-therapy for mental
health problems: A systematic review. Telemedicine and e-
Mohr 2006 Health 2008;14:707-14. [DOI: 10.1089/tmj.2007.0111]
Mohr DC, Hart SL, Howard I, Julian L, Vella L, Catledge C,
Pull 2007
et al. Barriers to psychotherapy among depressed and
nondepressed primary care patients. Annals of Behavior Pull CB. Combined pharmacotherapy and cognitive-behavioral
Medicine 2006;32(3):254-8. therapy for anxiety disorders. Current Opinion in Psychiatry
2007;20(1):30-5.
Norton 2007
Reger 2009
Norton PJ, Price EC. A meta-analytic review of adult cognitive-
behavioral treatment outcomes across anxiety disorders. Reger MA, Gahm GA. A meta-analysis of the effects of internet-
Journal of Nervous and Mental Diseases 2007;195(6):521-31. and computer-based cognitive-behavioral treatments for
anxiety. Journal of Clinical Psychology 2009;65(1):53-75.
Norton 2014
Revicki 2012
Norton PJ, Little TE, Wetterneck CT. Does experience
matter? Trainee experience and outcomes during Revicki DA, Travers K, Wyrwich KW, Svedsater H, Locklear J,
transdiagnostic cognitive-behavioral group therapy for Stoeckl Mattera M, et al. Humanistic and economic burden of
anxiety. Cognitive Behaviour Therapy 2014;43:230-8. [DOI: generalized anxiety disorder in North America and Europe.
10.1080/16506073.2014.919014] Journal of Affective Disorders 2012;140:103-12. [DOI: 10.1016/
j.jad.2011.11.014]
Ohayon 2000
RevMan 2014 [Computer program]
Ohayon MM, Shapiro CM, Kennedy SH. Differentiating DSM-IV
anxiety and depressive disorders in the general population: The Nordic Cochrane Centre, The Cochrane Collaboration.
Comorbidity and treatment consequences. Canadian Journal of Review Manager (RevMan). Version 5.3. Copenhagen: The
Psychiatry 2000;45:166-72. Nordic Cochrane Centre, The Cochrane Collaboration, 2014.
Olatunji 2010 Robins 1991

Olatunji BO, Hollon SD. Preface: the current status of cognitive Robins LN, Locke BZ, Regier DA. An overview of psychiatric
behavioral therapy for psychiatric disorders. Psychiatric Clinics disorders in America. In: LN Robins, DA Regier editor(s).
of North America 2010;33(3):xiii-ix. Psychiatric disorders in America: The Epidemiologic Catchment
Area Study. New York: Free Press, 1991:328-66.
Olthuis 2011
Rost 2002
Olthuis JV, Watt MC, Stewart SH. Therapist-delivered distance
cognitive behavioural therapy for anxiety disorders in adults. Rost K, Fortney J, Fischer E, Smith J. Use, quality, and outcomes
Cochrane Database of Systematic Reviews 2011, Issue 3. [DOI: of care for mental health: the rural perspective. Medical Care
10.1002/14651858.CD009028] Research and Review 2002;59(3):231-65.
Ost 1997 Salkovskis 1985

Ost LG. Rapid treatments of specific phobias. In: GCL Davey Salkovskis P. Obsessional-compulsive problems: a cognitive
editor(s). Phobias: A handbook of theory, research, and behavioural analysis. Behaviour Research and Therapy
treatment. Chichester, UK: Wiley, 1997:227-46. 1985;23:571-83.
Otto 2000 Slade 2007

Otto MW, Pollack MH, Maki KM. Empirically supported Slade T, Johnston A, Oakley Brown MA, Andrews G, Whiteford H.
treatments for panic disorder: costs, benefits, and stepped care. 2007 National Survey of Mental Health and Wellbeing: Methods
Journal of Consulting and Clinical Psychology 2000;68(4):556-63. and key findings. Australian and New Zealand Journal of
Psychiatry 2009;43(7):594-605.
Otto 2005
Spek 2007
Otto MW, Smits JAJ, Reese HE. Combined psychotherapy and
pharmacotherapy for mood and anxiety disorders in adults: Spek V, Cuijpers P, Nyklicek I, Riper H, Keyzer J, Pop V. Internet-
based cognitive behaviour for symptoms of depression
Informed decisions.
and anxiety: A meta-analysis. Psychological Medicine Westra 1998

2007;37:319-28. [DOI: 10.1017/S0033291706008944] Westra HA, Stewart SH. Cognitive behavioural therapy and
pharmacotherapy: complementary or contradictory approaches
Statistics Canada 2004
to the treatment of anxiety?. Clinical Psychology Review
Statistics Canada. Canadian Community Health Survey: 1998;18(3):307-40.
Mental health and well-being. http://www.statcan.ca/english/
freepub/82-617-XIE/index.htm (accessed 25 March 2008). WHO 1979
International Classification of Diseases (9th revision). World
Stewart 2008
Health Organization 1979.
Stewart SH, Conrod PJ, editors. Anxiety and substance use
disorders: the vicious cycle of comorbidity. New York, NY: WHO 1999
Springer, 2008. International Statistical Classification of Diseases and Related
Health Problems (10th revision). World Health Organization
Stewart 2009
1999.
Stewart RE, Chambless DL. Cognitive-behavioral therapy for
adult anxiety disorders in clinical practice: a meta-analysis Wilson 2006
of effectiveness studies. Journal of Consulting and Clinical Wilson KA, Hayward C. Unique contributions of anxiety
Psychology 2009;77(4):595-606. sensitivity to avoidance: a prospective study in adolescents.
Behaviour Research and Therapy 2006;44(4):601-9.
Svanborg 1994
Svanborg P, Asberg M. A new self-rating scale for depression and Young 2001
anxiety states based on the comprehensive psychopathological Young AS, Klap R, Sherbourne CD, Wells KB. The quality of
rating scale. ACTA Psychiatrica Scandinavica 1994;89:21-8. [DOI: care for depressive and anxiety disorders in the United States.
10.1111/j.1600-0447.1994.tb01480.x] Archives of General Psychiatry 2001;58(1):55-61.
Titov 2010b Yuen 1996
Titov N, Andrews G, Kemp A, Robinson E. Characteristics of Yuen EJ, Gerdes JL, Gonzales JJ. Patterns of rural mental
adults with anxiety or depression treated at an internet clinic: health care: an exploratory study. General Hospital Psychiatry
Comparison with a national survey and an outpatient clinic. 1996;18(1):14-21.
PLoS ONE 2010;5:e10885. [DOI: 10.1371/journal.pone.0010885]
* Indicates the major publication for the study
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Andersson 2009
Methods Randomised controlled trial
Participants Diagnosis: DSM-IV Specific Phobia, Spider Type
Method of diagnosis: SCID-IV
N: 27
Age: M = 25.6 (SD = 4.1); range = 18 to 65 years
Sex: 84.8% women
Country of residence: Sweden
Method of enrollment: responded to media advertisements in community
Baseline depression severity: (BDI-II) ICBT M = 7.9 (SD = 5.9); Live exposure M = 6.9 (SD = 6.2)
Interventions Participants were randomly assigned to either:
(1) Internet-based BT with e-mail support (n = 13)
Duration: 5 online modules completed over 4 weeks
Informed decisions.
Andersson 2009 (Continued)

Treatment protocol*: participants completed online modules on psychoeducation and exposure, with
e-mail support from a therapist for module exercises
Therapists: trained and supervised in this treatment protocol by treatment founder (Ӧst)
Therapist contact: 25 min per participant
Face-to-face contact: none
Dropout: n = 0; 0%
(2) Live exposure (n = 14)
Duration: 2 face-to-face sessions over 1 week
Treatment protocol*: participants attended an orientation session and one graded exposure session
with a therapist
Therapists: trained and supervised in this treatment protocol by treatment founder (Ӧst)
Therapist, face-to-face contact: one orientation session and one 3 hr exposure session
Dropout: n = 0; 0%
Outcomes Timepoints for assessment: pre- and post-treatment and 1 year follow-up
Primary outcomes:
(1) specific phobia symptoms: Behavioural Avoidance Test; Spider Phobia Questionnaire; Fear Survey
Schedule-III
(2) general anxiety: Beck Anxiety Inventory
Notes *treatment based on: Ӧst, L.-G. (1997). Rapid treatments of specific phobias. In G.C.L. Davey (Ed.), Pho-
bias: A handbook of theory, research and treatment (pp. 227–246). Chichester, UK: Wiley.
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence genera- Unclear risk Quote: "Thirty participants...were randomised by an independent person to ei-
tion (selection bias) ther..."
Comment: insufficient detail about method of randomisation provided to de-
termine risk
Allocation concealment Unclear risk No mention of allocation concealment

(selection bias)
Blinding of participants High risk Not possible to blind participants to their treatment condition nor therapists
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus face-to-face CBT)
mance bias)
All outcomes
Blinding of outcome as- High risk All outcome measures were self-report and participants were not blind to their
sessment (detection bias) own treatment condition
Self-Report Outcomes
Incomplete outcome data High risk Quote: "Three participants were dropped because of computer problems (n=1)
(attrition bias) or lack of time (n=2)."
All outcomes Comment: unclear which treatment condition the dropouts were from; ITT
analyses were not used
Informed decisions.
Andersson 2009 (Continued)
Selective reporting (re- Unclear risk No study protocol or trial registration available
porting bias)
Other bias Unclear risk Group comparisons at baseline not reported
Andersson 2012a
Participants Diagnosis: DSM-IV Social Phobia
N: 204
Age: for ICBT, M = 38.1 (SD = 11.3); for discussion group, M = 38.4 (SD = 10.9); range = 19 to 71 years
Sex: 61% women
Psychiatric co-morbidity: included
Co-use of adjunct therapy: excluded
Co-use of medication: 13.7%
Method of enrollment: responded to online study advertisement
Baseline depression severity: (MADRS-S) ICBT M = 13.45 (SD = 7.14); Discussion group M = 14.29 (SD =
6.63)
(1) Internet-based CBT with e-mail support (n = 102)
Treatment protocol*: participants completed online modules on psychoeducation, cognitive restruc-

turing, exposure, social skills, and relapse prevention, with email support from a therapist for module
exercises
Therapists: 7 licensed clinical psychologists (avg. 3 years experience; previous experience with Inter-
net treatment) and 6 clinical psychology students in their last year of the master's program; all had ba-
sic CBT training; students had clinical supervision during study
Therapist contact: 15 min per participant each week
Dropout: n = 8; 7.8%
(2) Online discussion group (n = 102)
Duration: 9 weeks
Treatment protocol: participants made weekly posts in an online topic-relevant discussion group
Therapist, face-to-face contact: none
Dropout: n = 2; 2%
Informed decisions.
Andersson 2012a (Continued)
Outcomes Timepoints for assessment: pre- and post-treatment
Primary outcomes:
(1) social phobia symptoms: Liebowitz Social Anxiety Scale, Social Phobia Scale, Social Interaction Anx-
iety Scale, Social Phobia Screening Questionnaire
(3) clinically important improvement: Clinical Global Impression Improvement Scale
Secondary outcome:
(1) quality of life: Quality of Life Inventory
Notes *treatment based on: Furmark, T., Holmstrom, A., Sparthan, E., Carlbring, P., & Andersson, G. (2006). So-
cial fobi – Effectiv hjalp med kognitiv beteendeterapi [Social phobia – effective help via CBT]. Stock-
holm: Liber.
Risk of bias
Random sequence genera- Low risk Quote: "Randomization was performed by an independent third-party using
tion (selection bias) an online true random-number service (www.random.org)."
Comment: adequate randomisation method

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus online discussion
mance bias) group)
All outcomes
Blinding of outcome as- High risk For self-report outcome measures, participants were not blind to their own
sessment (detection bias) treatment condition
Blinding of outcome as- Low risk Quote: "Outcome assessors were not aware of treatment status before the in-
sessment (detection bias) terview."
Observer/Interview-Rated Comment: interviewers were blind to treatment condition
Outcomes
Incomplete outcome data Low risk Quote: "Eight participants in the treatment group and 2 in the control group
(attrition bias) did not complete posttreatment data yielding a 5% dropout. In accordance
All outcomes with the ITT principle, all participants were asked to complete posttreatment
and follow-up assessments, regardless of how many treatment modules they
had completed and all were included in the analyses."
Comment: a small number of dropouts from both conditions was reported; ITT
analyses were used
Selective reporting (re- Low risk Results were reported for all outcome measures outlined in the trial registra-
porting bias) tion
Other bias Low risk Groups did not differ significantly on any measures at pre-treatment
Informed decisions.
Andersson 2012b
Participants Diagnosis: DSM-IV Generalized Anxiety Disorder
N: 81
Age: for ICBT, M = 44.4 (SD = 12.8); for internet psychodynamic therapy, M = 36.4 (SD = 9.7); for WLC, M =
39.6 (13.7); range = 19 to 66 years
Sex: 76.5% women
Psychiatric co-morbidity: 22.2% Social Phobia, 19.8% Panic Disorder, 3.7% OCD, 23.5% Major Depres-
sion
Method of enrollment: responded to study advertisements in community and online
Baseline depression severity: (MADRS-S) ICBT M = 22.30 (SD = 6.52); WLC M = 21.41 (SD = 5.99)
Interventions Participants were randomly assigned to one of:
Treatment protocol*: participants completed online modules on psychoeducation, applied relax-

ation, worry time, cognitive restructuring, problem solving, exposure, sleep management, and relapse
prevention, with e-mail support from a therapist for module exercises
Therapists: 2 licensed psychologists (previous experience with Internet treatment) and 3 psychology
students in their final year; all had CBT training; supervised by a senior researcher and licensed CBT
therapist
Therapist contact: M total time spent by therapist per participant = 92 min (SD = 61)
Dropout: n = 7; 25.9%
(2) Internet-based psychodynamic therapy with email support (n = 27)
Treatment Protocol^: participants completed online modules on seeing, understanding, and break-
ing unconscious patterns that contribute to emotional difficulties and guarding against future relapses,
with email support and encouragement from a therapist
Therapists: a licensed psychologist and 3 students in their final year of a clinical psychology program;
all trained in psychodynamic therapy
Therapist contact: M total time spent by therapist per participant = 113 min (SD = 41)
Dropout: n = 5; 18.5%
(3) Waiting list control (n = 27)
Informed decisions.
Andersson 2012b (Continued)

Duration: 8 weeks
Outcomes Timepoints for assessment: pre- and post-treatment and 3 month follow-up
Primary outcomes:
(1) generalized anxiety disorder symptoms: Penn State Worry Questionnaire, Generalized Anxiety Disor-
der Questionnaire IV
(2) general anxiety: State Trait Anxiety Inventory, Beck Anxiety Inventory
(3) clinically important improvement: SCID-IV
Secondary outcome:
(1) quality of life: Quality of Life Inventory (at post-treatment): participants reported if they were com-
pletely, moderately, or not satisfied with treatment
Notes *treatment based on: Paxling, B., Almlov, J., Dahlin, M., Carlbring, P., Breitholtz, E., Eriksson, T., & Ander-
sson, G. (2011). Internet-delivered cognitive behaviour therapy for generalized anxiety disorder: A ran-
domized controlled trial. Cognitive Behaviour Therapy, 40, 159-173.
^treatment based on: Silverberg, F. (2005). Make the leap: A practical guide to breaking the patterns that
hold you back. New York: Marlow and Company.
Risk of bias
Random sequence genera- Low risk Quote: "The randomization procedure was managed by an external adminis-
tion (selection bias) trator who was not otherwise involved in the study. A true random number ser-
vice (www.random.org) was used to ensure complete randomness. Random-
ization was done after inclusion wherein participants were randomized to the
three groups with no stratification."

(selection bias)
mance bias) group)
All outcomes
Blinding of outcome as- Low risk Quote: "After the treatment period, the interviewers were blinded concerning
sessment (detection bias) participant status and allocation (given that the posttreatment interviewers
Observer/Interview-Rated did not have access to information about the participants). In addition, partici-
Outcomes pants were asked not to reveal whether they had received treatment."
Comment: interviewers were blinded
Incomplete outcome data Low risk A small number of dropouts from each condition (depending on outcome
(attrition bias) measure, 4 to 7 for Internet CBT, 1 to 5 dropouts for Internet psychodynam-
All outcomes
Informed decisions.
Andersson 2012b (Continued)

ic therapy and 1 to 2 dropouts for waiting list control) and intention-to-treat
analyses were used
Selective reporting (re- Low risk All results were reported for all outcome measures outlined in the trial regis-
porting bias) tration
Berger 2009
N: 52
Sex: 44.2% women
Country of residence: 88% Switzerland, 10% France, 2% Belgium
Psychiatric co-morbidity: 26.9% had a co-morbid Axis I diagnosis
Co-use of adjunct therapy or medication: excluded
Baseline depression severity: (BDI-II) ICBT M = 16.6 (SD = 6.2); WLC M = 17.9 (SD = 10.4)
(1) Internet-based CBT with email support (n = 31)
Treatment protocol*: participants completed online modules on psychoeducation, self-focused at-

tention, safety behaviours, in vivo exposure, and cognitive restructuring, with email support from a
therapist for module exercises
Therapists: 6 master’s level clinical psychologists; 4 in their first year of a CBT training program, 2 in a
postgraduate clinical psychology and psychotherapy course
Therapist contact: M e-mails from participant = 5.5 (range = 0 to 16); in addition to responding to
these, therapists sent weekly motivating e-mails
Duration: 10 weeks
Informed decisions.
Berger 2009 (Continued)

Primary outcome:
(1) social phobia symptoms: Liebowitz Social Anxiety Scale – Self-Report; Social Phobia Scale; Social In-
teraction Anxiety Scale
Secondary outcome:
(1) treatment satisfaction (at post-treatment): participants reported if they were completely, moderate-
ly, or not satisfied with treatment
Notes *treatment based on: Stangier, U., Heidenreich, T., & Peitz, M. (2003). Soziale Phobien. Ein kognitiv-ver-
haltenstherapeutisches Behandlungsmanual [Social phobia. A cognitive-behavioral treatment manual].
Weinheim: Beltz.
Risk of bias
Random sequence genera- Low risk Quote: "We used a weighted randomizations procedure (Altman, 1991), such
tion (selection bias) that 60% were assigned to the treatment condition and 40% to the waiting-list
control group. According to a computer-generated randomizations scheme..."

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus waiting list)
mance bias)
All outcomes
Incomplete outcome data Low risk Quote: "After randomizations, 5 participants (3 in the treatment group and 2
(attrition bias) in the control group) dropped out during the course of the study and did not
All outcomes complete post assessment (9.6%). According to an ITT paradigm..."
Comment: a small and similar number of dropouts from both treatment condi-
tions was reported; ITT analyses were used
Selective reporting (re- Unclear risk Trial registration suggests that the State Trait Anxiety Inventory was complet-
porting bias) ed by participants, however, results are not reported for this outcome; all oth-
er outcomes utlined in the protocol are reported in the manuscript
Berger 2011
N: 81
Informed decisions.

Sex: 53.1% women
Country of residence: Switzerland
Psychiatric co-morbidity: 38% had at least one other Axis I diagnosis; 12% PD, 10% Specific Phobia,
2% GAD, 22% MDD or Dysthymia, 2% Eating Disorder
Baseline depression severity: (BDI-II) Guided ICBT M = 18.2 (SD = 11.5); Unguided ICBT M = 17.7 (SD =
9.8)
(1) Guided internet-based CBT (with e-mail support) (n = 27)
Duration: 5 online modules completed over 10 weeks (M hrs spent online = 10)
Treatment protocol*: participants completed online modules on motivational interviewing, psychoe-

ducation, cognitive restructuring, self-focused attention, and exposure, with weekly e-mail support
from a therapist
Therapists: 2 clinical psychology master's level graduate students, 2 master’s level clinical psycholo-
gists in post-graduate CBT training, 2 licensed psychologists with more than 5 years research and clini-
cal experience
Therapist contact: M e-mails from participant = 6.16 (SD = 4.56; range = 1 to 17); M e-mails from thera-
pist = 12.44 (SD = 2.85; range = 6 to 17)
Dropout: n = 3; 11.1%
(2) Unguided internet-based CBT (n = 27)
Duration: 5 online modules completed over 10 weeks (M hrs spent online = 9.5)

ducation, cognitive restructuring, self-focused attention, and exposure independently
(3) Step-up on demand Internet-based CBT (with e-mail or phone support) (n = 27)
Duration: 5 online modules completed over 10 weeks (M hrs spent online = 10.5)

ducation, cognitive restructuring, self-focused attention, and exposure, with e-mail or phone support,
or both, from a therapist as requested
Therapists: 2 clinical psychology master's level graduate students, 2 master’s level clinical psycholo-
gists in post-graduate CBT training, 2 licensed psychologists with more than 5 years research and clini-
cal experience
Therapist contact: 52% of participants did not request contact, 33% requested weekly e-mail contact,
7% requested weekly e-mail and phone contact
Informed decisions.

Primary outcomes:
teraction Anxiety Scale
Secondary outcome:
(1) treatment satisfaction (at post-treatment): Client Satisfaction Questionnaire
Notes *treatment based on: Stangier, U., Heidenreich, T., & Peitz, M. (2003). Soziale Phobien. Ein kognitiv-ver-
haltenstherapeutisches Behandlungsmanual [Social phobia. A cognitive-behavioral treatment manu-
al]. Weinheim: Beltz.
Risk of bias
Random sequence genera- Low risk Quote: "Participants were randomised into one of the three conditions using a
tion (selection bias) computerized random number generator (www.random.org)."
Allocation concealment Low risk Quote: "The allocation schedule was generated by an independent researcher
(selection bias) and was unknown to the investigators."
Comment: allocation likely concealed sufficiently to prevent deviations from
protocol
and personnel (perfor- to the treatment they delivered (guided versus unguided versus step-up on de-
mance bias) mand Internet-based CBT)
All outcomes
Blinding of outcome as- High risk Quote: "The interviewers could not be kept blind regarding group assignment
sessment (detection bias) at post-assessment because some participants disclosed aspects of the group
Observer/Interview-Rated assignment during the interview."
Outcomes
Comment: attempts were made to ensure interviewers were blind to treat-
ment condition, however, participants revealed their treatment condition
Incomplete outcome data Low risk Quote: "Six participants (7.4%) dropped out before post-treatment assessment
(attrition bias) (one in the self-help group, three in the guided self-help group, and two in the
All outcomes step-up of support on demand condition)."; "There was no significant differ-
ence in terms of demographics, pre-treatment, or post-treatment scores be-
tween those who provided post-treatment and follow-up data and those who
did not..."; "All analyses were based on the ITT sample."
Comment: a small and similar number of dropouts from the three treatment
conditions was reported; reasons were provided for dropouts (self-help: disap-
pointed with group assignment; guided self-help: wanted face-to-face contact
or had internet trouble; step-up on demand: vacation or no reason); ITT analy-
ses were used
porting bias)
Informed decisions.
Bergstrom 2010
Participants Diagnosis: DSM-IV Panic Disorder (n=16) or Panic Disorder with Agoraphobia (n = 88)
Method of diagnosis: MINI
N: 104
Age: for ICBT, M = 33.8 (SD = 9.7); for face-to-face CBT, M = 34.6 (SD = 9.2)
Sex: 61.5% women
Co-use of medication: 45% (34% SSRI or SNRI, 13% benzodiazepines, 24% benzodiazepine derivatives
or neuroleptics, 5% tricyclic antidepressants)
Method of enrollment: referred to study by health professionals or self-referred to study clinic
Baseline depression severity: (MADRS-S) ICBT M = 8.9 (SD = 5.2); face-to-face CBT M = 9.5 (SD = 4.9)

turing, exposure, and relapse prevention, with e-mail support from a therapist for module exercises,
and posted on an online discussion forum
Therapists: psychologists
Therapist contact: M e-mails from therapist = 11.3 (SD = 4.3); M total time spent by therapist per partic-
ipant = 35.4 min (SD = 19)
Dropout: n = 9; 17%
(2) Face-to-face group CBT (n = 54)
Duration: 10 face-to-face group therapy sessions over 10 weeks
Treatment protocol*: group sessions focused on psychoeducation, cognitive restructuring, exposure,

and relapse prevention
Therapists: 2 regular clinical psychologists, not specially trained for this study
Therapist, face-to-face contact: 10 x 2 hr group sessions
Dropout: n = 11; 18.3%
Primary outcomes:
(1) panic symptoms: Panic Disorder Severity Scale; Anxiety Sensitivity Index
(2) clinically important improvement: MINI
Informed decisions.
Bergstrom 2010 (Continued)

Secondary outcome:
(1) quality of life: Sheehan Disability Scale
Notes *treatment based on: Barlow D.H., & Craske M.G. (2000). Mastery of your anxiety and panic (MAP-3). San
Antonio: The Psychological Corporation.
Risk of bias
Random sequence genera- Low risk Quote: "The participants were divided into two groups...by an independent
tion (selection bias) random number procedure..."
Allocation concealment Low risk Quote: "...where each patient was assigned to either treatment by the opening
(selection bias) of sealed numbered envelopes."
Comment: adequate allocation concealment
and personnel (perfor- to the treatment they delivered (Internet-based applied relaxation versus In-
mance bias) ternet-based CBT)
All outcomes
Blinding of outcome as- Low risk Quote: "All outcome measures...were administered during the clinical inter-
sessment (detection bias) view..."
Self-Report Outcomes Comment: self-report outcomes were not completed
Blinding of outcome as- Low risk Quote: "The psychiatrists performing the clinical interviews at post-treat-
sessment (detection bias) ment and follow-up were blind to treatment condition."; "All outcome mea-
Observer/Interview-Rated sures...were administered during the clinical interview..."
Outcomes Comment: interviewers were blind to treatment condition
Incomplete outcome data Low risk Quote: "Nine participants dropped out after randomisation but before com-
(attrition bias) mencing treatment. Various reasons were given for not starting treatment, but
All outcomes all pertained to different life circumstances of the individual participants and
not to randomisation status. These initial dropouts were excluded from the
statistical analyses."; "A number of patients did not return for the clinical inter-
view at post-treatment... a mixed effects models approach was used in the sta-
tistical analysis to adjust for these missing values."
Comment: a similar number of dropouts from both treatment conditions was
reported (during treatment: six from treatment, five from comparator); mixed
effects models were used to account for missing data
Selective reporting (re- Low risk Results for all outcome measures outlined in the trial registration were report-
porting bias) ed
Carlbring 2001
Participants Diagnosis: DSM-IV Panic Disorder
Method of diagnosis: CIDI and ADIS-IV
Informed decisions.
Carlbring 2001 (Continued)

N: 41
Age: M = 34 (SD = 7.5); range = 21 to 51 years
Sex: 71% women
Co-use of adjunct therapy: ongoing for > 6 months and not CBT (n = 1)
Co-use of medication: 64% (44% SSRIs, 10% benzodiazepines, 5% beta-blockers, 5% tricyclic antide-
pressants)
Duration: 6 online modules completed over 7 to 12 weeks
Treatment protocol*: participants completed online modules on psychoeducation, breathing retrain-

ing, cognitive restructuring, exposure, and relapse prevention, with email support from a therapist for
module exercises
Therapists: a clinical psychology graduate student
Therapist contact: M reciprocal e-mail contacts = 7.5 (SD = 1.2; range = 6 to 15); M total time spent by
therapist per participant = 90 min
Dropout: n = 4; 19%
Duration: 7 to 12 weeks
Dropout: n = 1; 5%
Primary outcomes:
(1) panic and agoraphobia symptoms: Body Sensations Questionnaire; Agoraphobic Cognitions Ques-
tionnaire; Mobility Inventory
Secondary outcomes:
(2) treatment satisfaction (at post-treatment): Evaluation of Self-Help Program and Advisory Service
Notes *treatment based on: Barlow, D.H., & Craske, M.G. (1994). Mastery of your anxiety and panic. San Anto-
nio, TX: The Psychological Corporation. AND Zuercher-White, E. (1998). An end to panic: Breakthrough
techniques for overcoming panic disorder (2nd ed.). Oakland, CA: New Harbinger Publications.
Informed decisions.
Risk of bias
Random sequence genera- Low risk Quote: "Participants were divided into two groups by the drawing of lots.
tion (selection bias) These were drawn for the two treatment groupings pairwise for participants
who had completed their baseline measurements. In other words, as soon as
two participants had completed their baseline measurements, one was allo-
cated to the treatment group and the other to the waiting-list group."
Comment: adequate randomiation method
Allocation concealment Low risk Appears that lots were drawn immediately before assignment so allocation
(selection bias) was likely concealed adequately
and personnel (perfor- to the treatment they delivered (waiting list versus Internet-based CBT)
mance bias)
All outcomes
Blinding of outcome as- High risk All outcome measures were self-reported and participants were not blind to
sessment (detection bias) their own treatment condition
Incomplete outcome data Low risk Quote: "After randomizations, five people dropped out during the course of
(attrition bias) the study. There were four dropouts from the treatment group and one from
All outcomes the waiting-list group, χ2(1) = 2.9, P < 0.05. In the treatment group, lack of time
was given as the main reason for discontinuing (n = 3). One patient dropped
out because of a newly discovered cancer. The person who left the waiting-list
group gave no reason."; "...intention-to-treat evaluation of the results."
Comment: though there was a difference in the number of dropouts between
the two treatment conditions, the number of dropouts was small and reasons
did not relate directly to treatment components; ITT analyses were used
porting bias)
Carlbring 2005
Participants Diagnosis: DSM-IV Panic Disorder (49%) or Panic Disorder with Agoraphobia (51%)
N: 49
Age: M = 35 (SD = 7.7); range = 18 to 60 years
Sex: 71% women
Psychiatric co-morbidity: 49% another Anxiety Disorder, 6% Major Depression
Co-use of adjunct therapy: ongoing for > 6 months and not CBT (4%)
Informed decisions.

Co-use of medication: 30.6% SSRIs, 8.2% benzodiazepines, 6.1% beta-blockers, 6.1% tricyclic antide-
pressants
Baseline depression severity: (BDI-II) ICBT M = 11.8 (SD = 7.8); face-to-face CBT M = 15.9 (SD = 9.0);
(MADRS-S) ICBT M = 13.4 (SD = 5.3); face-to-face CBT M = 16.0 (SD = 4.3)
Treatment protocol*: participants completed online modules on psychoeducation, breathing retrain-

ing, cognitive restructuring, exposure, and relapse prevention, with e-mail support from a therapist for
module exercises
Therapists: 4 licensed clinical psychologists (research or clinical experience, or both, with anxiety dis-
orders), 3 advanced graduate students with a master's degree in clinical psychology, 1 student in final
semester of master's degree program; all supervised by a licensed CBT psychologist and supervisor
Therapist contact: M reciprocal e-mail contacts = 15.4 (SD = 5.5; range = 4-31); M total time spent by
therapist per participant = 150 min
Dropout: n = 3; 12%
(2) Face-to-face individual CBT (n = 24)
Duration: 10 individual face-to-face sessions over 10 weeks
Treatment protocol*: sessions focused on psychoeducation, breathing retraining, cognitive restruc-

turing, exposure, and relapse prevention
Therapists: 4 licensed clinical psychologists (research and/or clinical experience with anxiety disor-
ders), 3 advanced graduate students with a master's degree in clinical psychology, 1 student in the final
semester of their master's degree program; all supervised by a licensed CBT psychologist and supervi-
sor
Therapist, face-to-face contact: 10 x 45 to 60 min sessions
Dropout: n = 3; 12.5%
Secondary outcome:
Notes *treatment based on: Barlow, D.H., & Craske, M.G. (1994). Mastery of your anxiety and panic. San Anto-
nio, TX: The Psychological Corporation. AND Zuercher-White, E. (1998). An end to panic: Breakthrough
techniques for overcoming panic disorder (2nd ed.). Oakland, CA: New Harbinger Publications.
Risk of bias
Informed decisions.
Random sequence genera- Low risk Quote: "Participants were divided into two groups...by a true random-num-
tion (selection bias) ber-service (http://www.random.org)."

(selection bias)
and personnel (perfor- to the treatment they delivered (face-to-face CBT or Internet-based CBT)
mance bias)
All outcomes
Blinding of outcome as- Low risk Quote: "...a clinical re-interview (SCID) was administered by an independent
sessment (detection bias) psychologist blind for treatment condition."
Outcomes
Incomplete outcome data Low risk Quote: "After randomizations, six people dropped out during the course of
(attrition bias) the study. There were three dropouts from the LIVE therapy group and three
All outcomes from the IT group. Lack of time was given as the main reason for discontinuing.
However, in accordance with the intention to treat paradigm...post-treatment
data were collected from all dropouts."
Comment: a small and similar number of dropouts reported in the two treat-
ment conditions; used ITT analysis
porting bias)
Carlbring 2006
Method of Diagnosis: SCID-IV
N: 60
Age: M = 36.7 (SD = 10); range = 18 to 60 years
Sex: 60% women
Co-use of adjunct therapy: none
Co-use of medication: 54%
Informed decisions.

Baseline depression severity: (BDI-II) ICBT M = 17.7 (SD = 8.8); WLC CBT M = 15.4 (SD = 7.4); (MADR-S)
ICBT M = 16.4 (SD = 7.2); WLC CBT M = 15.1 (SD = 6.0)
(1) Internet-based CBT with e-mail and phone support (n = 30)
Treatment protocol: participants completed online modules on psychoeducation, breathing retrain-

ing, cognitive restructuring, exposure, and relapse prevention, with email support from a therapist for
module exercises
Therapists: 1 licensed psychologist, 2 students in their final year of a clinical psychology master's pro-
gram; all had regular supervision from an experienced CBT psychologist
Therapist contact: M reciprocal contacts = 13.5 (SD = 4.4; range = 7-29); M time spent by therapist per
participant per week = 12 min; M length of weekly phone conversations = 11.8 min (range = 9.6 to 15.6)
Duration: 10 weeks
Primary outcomes:
Secondary outcomes:
(2) treatment satisfaction (at post-treatment): participants reported if they were satisfied, very satis-
fied, or dissatisfied with treatment and gave their opinion on the pace of the program
Notes
Risk of bias
Random sequence genera- Low risk Quote: "The participants were divided into two groups, treatment or a waiting
tion (selection bias) list, by a true random-number service."

(selection bias)
Informed decisions.
mance bias)
All outcomes
Blinding of outcome as- Low risk Quote: "...a reinterview administered by an independent psychologist who was
sessment (detection bias) blind to treatment condition."
Outcomes
Incomplete outcome data Low risk Quote: "One participant dropped out during the study; shortage of time was
(attrition bias) said to be the main reason. However, in accordance with the intention-to-
All outcomes treat paradigm... posttreatment data were also collected from the participant
who dropped out. Two participants in the treatment condition and one on the
waiting list did not return their posttreatment questionnaires. Therefore, their
pretreatment scores were carried forward to the posttreatment assessment
point."
porting bias)
Carlbring 2007
N: 60
Age: for ICBT, M = 32.4 (SD = 9.1); for WLC, M = 32.9 (SD = 9.2); range = 18 to 60 years
Sex: 64.9% women
Co-use of medication: included
Baseline depression severity: (MADRS-S) ICBT M = 13.4 (SD = 8.4); WLC CBT M = 13.5 (SD = 6.0)
(1) Internet-based CBT with e-mail and phone support (n=30)
Informed decisions.

Treatment protocol*: participants completed online modules, with e-mail support from a therapist for
module exercises
Therapists: 2 students completing their last semester of a clinical psychology master’s degree
Therapist contact: M time spent by therapist per participant per week = 22 min; M length of weekly
phone conversations = 10.5 min (SD = 3.6)
Duration: 9 weeks
Primary outcomes:
teraction Anxiety Scale; Social Phobia Screening Questionnaire
Secondary outcome:
Notes *treatment based on: Furmark, T., Holmstrom, A., Sparthan, E., et al. (2006). Social Phobia – effective
treatment with
cognitive-behavioural therapy (in Swedish). Liber.
Risk of bias
Random sequence genera- Low risk Quote: "...were divided into two groups (treatment or waiting-list control) by
tion (selection bias) an online true random-number service independent of the investigators and
therapists. This service is run by the Department of Computer Science at the
University of Dublin and the numbers are generated using a purely random
process (atmospheric disturbances in space)."
Comment: adequate randomisation process

(selection bias)
mance bias)
All outcomes
Informed decisions.
Incomplete outcome data Low risk Quote: "Two participants, one in each condition, were excluded from the
(attrition bias) analysis since they started other treatment during the period. A total of 27 of
All outcomes the 29 people in the treatment group completed all nine modules within the
intended 9-week time frame. Lack of time was provided as the explanation for
terminating treatment prematurely. One of them did not send in post-treat-
ment measures, which explains why intention-to-treat analysis was used. Fi-
nally, after randomisation but before answering the pre-treatment question-
naires, one person in the waiting-list chose to refrain from participating be-
cause of lack of computer access. Thus, data for 29 participants in the treat-
ment group and 28 in the control group were eligible for analysis."
porting bias)
Carlbring 2011
Participants Diagnosis: DSM-IV Panic Disorder (9%), Panic Disorder with Agoraphobia (22%), Social Phobia (39%),
Generalized Anxiety Disorder (20%), Anxiety Disorder not otherwise specified (13%)
N: 54
Sex: 76% women
Psychiatric co-morbidity: 2% OCD, 2% PTSD, 20% MDD, 7% mild Depression, 15% Dysthymia
Co-use of medication: 26% using an antidepressant or anxiolytic
Baseline depression severity: (MADRS-S) ICBT M = 20.41 (SD = 7.31); attention control M = 19.59 (SD =
7.43)
Duration: 6 to 10 online modules completed over 10 weeks
Treatment protocol*: participants completed online modules as prescribed by a therapist on topics

related to their diagnosis, with e-mail support from a therapist for module exercises
Therapists: 8 clinical psychology master's students in last semester of training
Therapist contact: M time spent by therapist per participant per week = 15 min
Informed decisions.

(2) Attention control (n = 27)
Duration: 10 weeks
Treatment protocol: participants made weekly posts in a confidential online support group based on
a theme posted by a therapist
Therapists: 8 clinical psychology master's students in last semester of training

Therapist contact: therapist spent 1 hr per week monitoring forum
Dropout: n = 0; 0%
Primary outcomes:
(1) anxiety symptoms: Clinical Outcomes in Routine Evaluation – Outcome Measure
(3) clinically important improvement: Clinical Global Impression Scale
Secondary outcome:
Notes *treatment based on Internet-based programs described in: Andersson, G., Carlbring, P., Holmström, A.,
Sparthan, E., Furmark, T., Nilsson-Ihrfelt, E., et al. (2006). Internet-based self-help with therapist feed-
back and in vivo group exposure for social phobia: a randomised controlled trial. Journal of Consult-
ing and Clinical Psychology, 74, 677-686.; Carlbring, P., Westling, B. E., Ljungstrand, P., Ekselius, L, & An-
dersson, G. (2001). Treatment of panic disorder via the Internet: A randomised trial of a self-help pro-
gram. Behavior Therapy, 32, 751-764.; AND Vernmark, K., Lenndin, J., Bjärehed, J., Carlsson, M., Karls-
son, J., Öberg, J., et al. (2010). Internet administered guided self-help versus individualized e-mail ther-
apy: a randomised trial of two versions of CBT for major depression. Behaviour Research and Therapy,
48, 368-376.
Risk of bias
Random sequence genera- Low risk Quote: "The participants were divided into two groups... by an online true ran-
tion (selection bias) dom-number service independent of the investigators and therapists."

(selection bias)
All outcomes
Blinding of outcome as- Low risk Quote: "...a clinical global impression of improvement (CGI-I) was mapped on
sessment (detection bias) a 7-point scale (CGI; Guy, 1976) after a telephone interview by a blind asses-
Informed decisions.

Observer/Interview-Rated sor who had no earlier contact with the participants and no knowledge of to
Outcomes which group they had been randomly allocated."
Comment: interviewers were blind to treatment condition
Incomplete outcome data Low risk Quote: "The response rate was... 96.3% (52/54) at post-treatment."; "Since
(attrition bias) the missing data at post-treatment was only in the treatment group, repeated
All outcomes ANOVAs with conservative imputation according to the last observation-car-
ried-forward method in case of missing data was used in the analysis of the im-
mediate results."
Comment: a small and similar number of dropouts from both treatment con-
ditions was reported (two from treatment, zero from comparator); used ITT
analysis
porting bias)
Furmark 2009a
N: 120
Age: for Internet CBT, M = 35 (SD = 10.2); for bibliotherapy, M = 37.7 (SD = 10.3); for waiting list, M = 35.7
(SD = 10.9)
Sex: 67.5% women

turing, exposure, social skills, and relapse prevention, with e-mail support from a therapist for module
exercises, and posted on an online discussion forum
Therapists: 6 licensed clinical psychologists, 7 clinical psychology students in final year of master's
program; students had clinical supervision during the study
Therapist contact: 15 min per week
(2) Bibliotherapy (n = 40)
Informed decisions.
Furmark 2009a (Continued)

Duration: 9 sections of the manual completed over 9 weeks
Treatment protocol*: participants received a self-help manual in the mail and completed it indepen-
dently
Duration: 9 weeks, completed weekly assessment measure

Primary outcomes:
Secondary outcome:
Notes *treatment based on: Carlbring, P., Furmark, T., Steczkó, J., Ekselius, L., & Andersson, G. (2006). An open
study of internet-based bibliotherapy with minimal therapist contact via email for social phobia. Clin-
ical Psychology, 10, 30-38.; Andersson, G., Carlbring, P., Holmström, A., Sparthan, E., Furmark, T., Nils-
son-Ihrfelt, E., et al. (2006). Internet-based self-help with therapist feedback and in-vivo group expo-
sure for social phobia: A randomised controlled trial. Journal of Consulting and Clinical Psychology, 74,
677-686.; Carlbring, P., Gunnarsdóttir, M., Hedensjö, L., Andersson, G., Ekselius, L., & Furmark, T. (2007).
Treatment of social phobia: randomised trial of internet-delivered cognitive–behavioural therapy with
telephone support. British Journal of Psychiatry, 190, 123-128.; AND Tillfors, M., Carlbring, P., Furmark,
T., Lewenhaupt, S., Spak, M., Eriksson, A., et al. (2008). Treating university students with social phobia
and public speaking fears: internet delivered self-help with or without live group exposure sessions. De-
pression and Anxiety, 25, 708-717.
Risk of bias
Random sequence genera- Low risk Quote: "Randomisation was performed by an independent third party using an
tion (selection bias) online true random-number service."

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus pure bibliotherapy
mance bias) versus waiting list)
All outcomes
Blinding of outcome as- High risk Quote: "...assessors were not masked with regard to the treatment assign-
sessment (detection bias) ment. However, all assessments were conducted online with standardised
Self-Report Outcomes written instructions and automatic scoring, reducing the risk of reactivity or
experimenter effects."
Informed decisions.
Furmark 2009a (Continued)

Comment: all outcome measures were self-report and participants were not
blind to their own treatment condition
Incomplete outcome data Low risk Quote: "Two participants, one each from the pure bibliotherapy and wait-
(attrition bias) ing-list groups, withdrew immediately after randomisation because of per-
All outcomes sonal reasons and one additional participant (ICBT group) did not provide
post-treatment data."; "For all randomised participants, missing data were re-
placed by the last obtained score (pre- or post-treatment), i.e., last observa-
tion carried forward."; "Ten participants (4.3%) withdrew from the study after
the first (n=6) or second (n=4) treatment week, the main reasons being lack of
time or motivation and personal problems unrelated to the treatment. In ac-
cordance with the intention-to-treat principle, all participants were asked to
complete post-treatment and follow-up assessments, regardless of how many
treatment modules they had completed."
Comment: one participant from each of ICBT, waiting list, and bibliotherapy
did not complete post-treatment measures; reasons for dropout from treat-
ment seem unrelated to treatment condition although are not provided based
on treatment condition; ITT analyses were used
porting bias) ed
Furmark 2009b
N: 115
Age: for internet CBT, M = 34.9 (SD = 8.4); for bibliotherapy, M = 32.5 (SD = 8.5); for bibliotherapy and dis-
cussion group, M = 35 (SD = 10.4); for internet applied relaxation, M = 36.4 (SD = 9.8)
Sex: 67.8% women
(1) Internet-based CBT with e-mail support (n 2 9)

exercises, and posted on an online discussion forum
Therapists: 6 licensed clinical psychologists, 7 clinical psychology students in final year of master's
program; students had clinical supervision during the study
Informed decisions.
Furmark 2009b (Continued)

Dropout: n = 0; 0%
(2) Bibliotherapy (n = 29)
dently
Dropout: n = 0; 0%
(3) Bibliotherapy and discussion group (n=28)
dently as well as posting weekly on an online discussion forum
Dropout: n = 0; 0%
(4) Internet-based applied relaxation (n = 29)

Treatment protocol^: participants completed online modules on psychoeducation, relaxation, and re-
lapse prevention, with e-mail support from a therapist for module exercises, and posted weekly on an
online discussion forum
Therapists: a licensed clinical psychologist, clinical psychology graduate students
Dropout: n = 0; 0%
Primary outcomes:
Secondary outcome:
Notes *treatment based on: Carlbring, P., Furmark, T., Steczkó, J., Ekselius, L., & Andersson, G. (2006). An open
study of internet-based bibliotherapy with minimal therapist contact via email for social phobia. Clin-
ical Psychology, 10, 30-38.; Andersson, G., Carlbring, P., Holmström, A., Sparthan, E., Furmark, T., Nils-
Informed decisions.
Furmark 2009b (Continued)

^treatment based on: Ӧst, L.G. (1997). Tillӓmpad avslappning [applied relaxation]. Stockholm, Sweden:
Repro HSC.
Risk of bias
Random sequence genera- Low risk Quote: "Randomisation was performed by an independent third party using an
tion (selection bias) online true random-number service."

(selection bias)
and personnel (perfor- to the treatment they delivered (ICBT versus pure bibliotherapy versus waiting
mance bias) list versus applied relaxation)
All outcomes
Blinding of outcome as- High risk Quote: "...assessors were not masked with regard to the treatment assign-
sessment (detection bias) ment. However, all assessments were conducted online with standardised
Self-Report Outcomes written instructions and automatic scoring, reducing the risk of reactivity or
experimenter effects."
Comment: all outcome measures were self-report and participants were not
blind to their own treatment condition
Incomplete outcome data Low risk Quote: "Two participants, one each from the pure bibliotherapy and wait-
(attrition bias) ing-list groups, withdrew immediately after randomisation because of per-
All outcomes sonal reasons and one additional participant (ICBT group) did not provide
post-treatment data."; "For all randomised participants, missing data were re-
placed by the last obtained score (pre- or post-treatment), i.e., last observation
carried forward."; "In accordance with the intention-to-treat principle, all par-
ticipants were asked to complete post-treatment and follow-up assessments,
regardless of how many treatment modules they had completed."
Comment: there was a very small and similar number of participants from
each treatment condition who did not complete post-treatment measures; ITT
analyses were used
porting bias) ed
Hedman 2011
Method of d: SCID-IV and MINI
N: 126
Age: for ICBT, M = 35.2 (SD = 11.1); for face-to-face CBT, M = 35.5 (SD = 11.6); range = 18 to 64 years
Informed decisions.
Hedman 2011 (Continued)

Sex: 38% women
Psychiatric co-morbidity: 18% another Anxiety Disorder, 15% MDD
Co-use of medication: 19.8% SSRIs, 4.8% SNRIs
Method of enrollment: referred to study by health professionals or self-referred to study clinic
Baseline depression severity: (MADRS-S) ICBT M = 12.7 (SD = 6.5); face-to-face CBT M = 14.0 (SD = 8.0)
Treatment protocol*: participants completed online modules on social phobia treatment themes
such as exposure and cognitive restructuring, with email support from a therapist for module activities
Therapists: 8 clinical psychologists (1 to 4 years experience with Internet CBT)
Therapist contact: M emails by therapists = 17.4; M time spent by therapist per participant per week =
5.5 min (SD = 3.6)
(2) Face-to-face group CBT (n = 62)
Duration: 15 face-to-face group sessions over 15 weeks
Treatment protocol^: participants attended face-to-face group therapy sessions on social phobia
treatment themes including cognitive restructuring and exposure
Therapists: 6 clinical psychologists (2-15 years experience with CBT for social phobia); supervised by a
licensed psychotherapist experienced in CBT for social phobia
Therapist, face-to-face contact: 15 x 2.5 hr group therapy sessions
Dropout: n = 0; 0%
Primary outcomes:
(1) social phobia symptoms: Liebowitz Social Anxiety Scale – Clinician; Social Phobia Scale; Social In-
teraction Anxiety Scale; Anxiety Sensitivity Index
Secondary outcome:
Notes *treatment based on: Andersson, G., Carlbring, P., Holmström, A., Sparthan, E., Furmark, T., Nils-
Informed decisions.
Hedman 2011 (Continued)

^treatment based on: Heimberg, R.G., & Becker, R.E. (2002). Cognitive-behavioral group therapy for so-
cial phobia. Basic mechanisms and clinical strategies. New York: Guilford Press.
Risk of bias
Random sequence genera- Low risk Quote: "A true random number service (http://www.random.org) was used to
tion (selection bias) ensure randomizations... Participants were allocated to CBGT or ICBT in a 1:1
ratio using simple randomizations with no restrictions or matching."
Allocation concealment Low risk Quote: "The randomizations procedure involved two external persons not in-
(selection bias) volved in the study; one provided randomizations data and the other moni-
tored that no manipulation of treatment allocation was performed by the re-
search group."; "The random sequence was generated after inclusion of partic-
ipants to ensure that assignment of intervention was concealed from assess-
ing psychiatrists and researchers of the study."
Comment: adequate allocation concealment
and personnel (perfor- to the treatment they delivered (internet-based CBT versus face-to-face CBT)
mance bias)
All outcomes
Blinding of outcome as- Low risk Quote: "To ensure the integrity of the blinding procedure, participants were in-
sessment (detection bias) structed not to mention which treatment they had received during the post-
Observer/Interview-Rated treatment and follow-up interviews. After completing the interviews, the as-
Outcomes sessing psychiatrists guessed allocation status for each participant."; "In four
instances blinding was broken. On two occasions participants accidentally
mentioned their treatment allocation status to the assessor, and in another
two occasions it was deemed necessary to break the blinding because of the
need to assess increased depressive symptoms during treatment... There was
no significant association between assessors' guess and actual treatment allo-
cation (χ2 = 0.27, df = 1, p = .61), indicating successful blinding."
Incomplete outcome data Low risk A similar number of dropouts from both treatment conditions was reported
(attrition bias) (13 for ICBT; 12 for CBGT); ITT analyses were used
All outcomes
Selective reporting (re- Unclear risk It is unclear if several outcomes in the trial registration (described as WQ, TIC-
porting bias) P, SSP) were reported in the manuscript; all other outcome measures outlined
in the trial registration were reported in the mansucript
Informed decisions.
Johnston 2011
Participants Diagnosis: DSM-IV Panic Disorder with or without agoraphobia (20.6%), Social Phobia (34.4%), general-
ized anxiety disorder (45%)
N: 139
Sex: 58.8% women
Country of residence: Australia
Psychiatric co-morbidity: 29% another Anxiety Disorder only, 9.2% another Affective Disorder only,
32.1% another Anxiety and Affective disorder
Method of enrollment: responded to online study advertisements
Baseline depression severity: (PHQ-9) ICBT with clinician M = 11.63 (SD = 5.96); WLC M = 11.71 (SD =
6.31)
(1) Internet-based CBT with e-mail and phone support from a clinician (n = 47)
Treatment protocol*: participants completed online modules on disorder-specific psychoeducation,

cognitive restructuring, core beliefs, exposure, assertiveness communication and interpersonal bound-
aries, and relapse prevention, with email and phone support from a therapist for module activities
Therapists: 1 clinical psychologist with specialist post-graduate training in clinical psychology and 2.5
years postgraduate experience
Therapist contact: M emails by therapist = 8.83 (SD = 3.19); M phone calls by therapist = 7.54 (SD =
2.43); M time spent by therapist per participant overall = 69.09 min (SD = 32.29)
Dropout: n = 5; 10.6%
(2) Internet-based CBT with e-mail and phone support from a coach (n = 46)
Treatment protocol*: participants completed online modules on disorder-specific psychoeducation,

cognitive restructuring, core beliefs, exposure, assertiveness communication and interpersonal bound-
aries, and relapse prevention, with email and phone support from a coach for module activities (no
clinical support)
Therapists: 1 clinical psychologist with specialist post-graduate training in clinical psychology and 2.5
years postgraduate experience
Therapist contact: M e-mails by coach = 8.88 (SD = 4.38); M phone calls by coach = 7.56 (SD = 1.19); M
time spent by coach per participant overall = 69.09 min (SD = 30.75)
Informed decisions.
Johnston 2011 (Continued)

Duration: 10 weeks
Dropout: n = 5; 10.9%
Primary outcomes:
(1) disorder-specific symptoms: Penn State Worry Questionnaire; Social Phobia Scale/Social Interaction
Anxiety Scale – Short Form; Panic Disorder Severity Scale – Self-Rating
(2) general anxiety symptoms: GAD-7, Depression Anxiety Stress Scales – 21
Secondary outcomes:
(2) treatment satisfaction: A 7-item questionnaire based on the Credibility/Expectancy Questionnaire
Notes *treatment based on: Titov , N., Andrews, G., Johnston, L., Robinson, E., Spence, J. (2010). Transdiag-
nostic Internet treatment for anxiety disorders: A randomised controlled trial. Behaviour Research and
Therapy, 48, 890-9.
Risk of bias
Random sequence genera- Low risk Quote: "...were randomised via a true randomizations process (www.ran-
tion (selection bias) dom.org), generated by an independent person, to either..."
Allocation concealment Low risk Quote: "The allocation sequence preceded pre-treatment diagnostic inter-
(selection bias) views and was concealed from LJ and JS [pre-treatment interviewers]."
Comment: unclear how allocation concealment occurred but it seems to have
taken place
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus waiting list con-
mance bias) trol)
All outcomes
Blinding of outcome as- High risk All outcomes were self-report outcome measures, participants were not blind
sessment (detection bias) to their own treatment condition
Incomplete outcome data Low risk Quote: "All post-treatment and 3-month follow-up analyses involved an inten-
(attrition bias) tion-to-treat (ITT) design and missing data was addressed by carrying forward
All outcomes the first available data (baseline-observation-carried-forward; BOCF)."
Comment: ITT analyses were used
Selective reporting (re- Unclear risk One measure that appears in the trial protocol (Agoraphobic Cognitions Ques-
porting bias) tionnaire) is not reported; all other outcomes in the trial registration are re-
ported
Informed decisions.
Kiropoulos 2008
Participants Diagnosis: DSM-IV Panic Disorder (41.9%) or Panic Disorder with Agoraphobia (58.1%)
Method of diagnosis: ADIS-IV
N: 86
Sex: 72.1% women
Psychiatric co-morbidity: 16% Social Phobia, 17% GAD, 10% Specific Phobia, 3% PTSD, 10% MDD, 5%
Dysthymia, 1% Alcohol Abuse, 8% Hypochondriasis
Method of enrollment: responded to media advertisements in community and online
Baseline depression severity: (DASS depression, log transformed) ICBT M = 2.72 (SD = 1.80); face-to-
face CBT M = 2.98 (SD = 1.61)
Duration: 6 online modules (+ 2 optional modules) completed over 6 weeks
Treatment protocol*: participants completed online modules on deep breathing, cognitive restruc-
turing, and exposure, with e-mail support from a therapist for module activities (optional modules on
stress and benzodiazepines)
Therapists: 9 registered and 1 probationary psychologist; all trained in CBT
Therapist contact: M e-mails by therapist = 18.24 (SD = 9.82); M e-mails by participant = 10.64 (SD =
8.21); M time spent by therapist per participant = 352 min (SD = 240)
Dropout: n = 5; 10.9%
(2) Face-to-face individual CBT (n = 40)
Duration: 12 face-to-face group sessions over 12 weeks
Treatment protocol^: participants attended face-to-face group therapy sessions on social phobia
treatment themes including cognitive restructuring and exposure
Therapists: registered psychologists
Therapist, face-to-face contact: 12 x 60 to 90 min sessions; M = 568 min (SD = 255.12)
Dropout: n = 2; 5%
Primary outcomes:
Informed decisions.
Kiropoulos 2008 (Continued)

(1) panic and agoraphobia symptoms: Panic Disorder Severity Scale; Body Vigilance Scale; Agoraphobic
Cognitions Questionnaire; Anxiety Sensitivity Profile
(2) general anxiety: DASS Stress and Anxiety subscales
(3) clinically important improvement: ADIS-IV
Secondary outcomes:
(1) quality of life: WHO Quality of Life – BREF subscales
(2) treatment satisfaction (at post-treatment): Treatment Satisfaction Questionnaire - Modified
Notes *on-line treatment program: Panic Online
^treatment based on: Barlow, D.H., & Craske, M.G. (2000). Mastery of your anxiety and panic: MAP-3. New
York: Graywind Publications.
Risk of bias
Random sequence genera- Low risk Quote: "...they were randomly allocated using a random numbers table..."
tion (selection bias) Comment: adequate randomisation method

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus face-to-face CBT)
mance bias)
All outcomes
Blinding of outcome as- Low risk Quote: "All assessors were blind to treatment allocation of eligible participants
sessment (detection bias) into the study."
Outcomes
Incomplete outcome data Low risk Quote: "The attrition rates were 10.9% (5/46) and 5% (2/40) for the PO and
(attrition bias) face-to-face treatment conditions, respectively. A Fisher's exact test revealed
All outcomes no difference in attrition rates between the two treatment conditions, χ2 (1, N
= 86) = .44, P > .05. Reasons for non-completion of either treatment included
participants not being contactable, changing their mind about taking part in
the study, because they could no longer commit to the 12-week treatment pro-
gram or because they no longer had access to the Internet."; "Data analysis in-
volved intention-to-treat analyses."
porting bias)
Informed decisions.
Paxling 2011
N: 89
Age: M = 39.3 (SD = 10.8); range = 18 to 66
Sex: 79.8% women
Psychiatric co-morbidity: included; 22.5% MDD
Method of enrollment: responded to media advertisements in the community and online
Treatment protocol*: participants completed online modules on psychoeducation, applied relax-

ation, worry time, cognitive restructuring, problem solving, sleep management, exposure, and relapse
prevention with email support from a therapist for module activities
Therapists: psychologists in their final year of training; all trained for 1 week in CBT protocol; supervi-
sion provided by experienced clinician
Therapist contact: M time spent by therapist per participant = 97 min (SD = 52)
Dropout: n = 6; 13.6%
Duration: 8 weeks
Primary outcomes:
(1) generalized anxiety symptoms: Penn State Worry Questionnaire, Generalized Anxiety Disorder Ques-
tionnaire – IV
(2) general anxiety symptoms: State Trait Anxiety Inventory, Beck Anxiety Inventory
Secondary outcomes:
Informed decisions.
Paxling 2011 (Continued)
Notes *treatment based on parts of: Ost, L.G. (1987) Applied relaxation: Description of a coping technique and
review of controlled studies. Behaviour Research and Therapy, 25, 379-409.; Borkovec, T.D., & Costello,
E. (1993). Efficacy of applied relaxation and cognitive-behavioral therapy in the treatment of general-
ized anxiety disorder. Journal of Consulting and Clinical Psychology, 61, 611-9.; Borkovec, T.D., Wilkin-
son, L., Folensbee, R., & Lerman, C. (1983). Stimulus control applications to treatment of worry. Behav-
iour Research and Therapy, 21, 247-51.; Borkovec, T.D., & Sharpless, B. (2004). Generalized anxiety disor-
der: Bringing cognitive-behavioral therapy into the valued present. In S. C. Hayes, V.M. Follette, & M.M.
Linehan (Eds.), Mindfulness and acceptance (pp. 209-42). New York, NY: Guilford Press.; Zetterqvist, K.,
Maanmies, J., Strom, L., & Andersson, G. (2003). Randomized controlled trial of Internet-based stress
management. Cognitive Behaviour Therapy, 3, 151-60.; Sanderson, W. C., & Rygh, J.L. (2004). Treating
generalized anxiety disorder: Evidence-based strategies, tools, and techniques. New York, NY: Guilford
Press.; Strom, L., Pettersson, R., & Andersson, G. (2004). Internet-based treatment for insomnia: A con-
trolled evaluation. Journal of Consulting and Clinical Psychology, 72, 113-20..
Risk of bias
Random sequence genera- Low risk Quote: "The 89 participants were randomised...by an independent person
tion (selection bias) not involved in the study. A computer-generated random list was obtained
via www.random.org, which utilizes atmospheric noise to create random se-
quences of numbers."

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus waiting list con-
mance bias) trol)
All outcomes
Blinding of outcome as- Low risk Quote: "The interviewers were blinded concerning participant status (e.g.
sessment (detection bias) treatment or control) since the posttreatment interviewers did not have access
Observer/Interview-Rated to information about the participants and started each interview by asking the
Outcomes participants not to say whether they were in the treatment or control condi-
tion."
Incomplete outcome data Low risk Quote: "Posttreatment measures were obtained from 38 or 44 randomised
(attrition bias) participants in the treatment group (86%) and 44 of 45 in the control group
All outcomes (98%)."; "In order to account for dropouts without assuming that the first mea-
surement was stable (i.e., the last observation carried forward assumption),
we used a mixed-effects models approach...Mixed-effect models are able to
accommodate missing data and integrate time-varying factors."
Comment: very little data was incomplete; an ITT approach was used
Selective reporting (re- Low risk Results were reported for all outcome measures outlined in the trial registra-
porting bias) tion
Informed decisions.
Richards 2006
Participants Diagnosis: DSM-IV Panic Disorder (21.9%) or Panic Disorder with Agoraphobia (78.1%)
Method of diagnosis: ADIS-IV
N: 23
Age: M = 36.59 (SD = 9.9); range = 18 to 70
Sex: 68.8% women
Psychiatric co-morbidity: 22% Social Phobia, 13% GAD, 9% Specific Phobia, 6% PTSD, 9% MDD, 6%
Hypochondriasis, 3% Somatization Disorder
Co-use of medication: 15.6% antidepressants, 12.5% benzodiazepines, 9.4% both antidepressants and
benzodiazepines
Baseline depression severity: (DASS depression) ICBT M = 21.25 (SD = 12.3); control M = 6.79 (SD = 6.4)
Treatment protocol*: participants completed online modules on relaxation strategies, cognitive re-
structuring, and exposure, with e-mail support from a therapist for module activities
Therapists: 1 clinical psychologist, 3 doctoral clinical psychology students; all experienced in CBT
Therapist contact: M emails by therapist = 18 (SD = 6.5); M e-mails by participant = 15.3 (SD = 12.8); M
time spent by therapist per participant = 376.30 min (SD = 156.8)
Dropout: n = 2; 16.7%
(2) Internet-based CBT and stress management with email support (n = 11)
Treatment protocol*: participants completed online modules on relaxation strategies, cognitive re-
structuring, and exposure, as well as several stress management modules, with email support from a
therapist for module activities
Therapists: 1 clinical psychologist, 3 doctoral clinical psychology students; all experienced in CBT
Therapist contact: M e-mails by therapist = 12.9 (SD = 3.8); M e-mails by participant = 11.6 (SD = 13.3);
M time spent by therapist per participant = 309.30 min (SD = 111.3)
Dropout: n = 1; 9%
(3) Internet-based information control (n = 9)
Duration: 8 weeks
Informed decisions.
Richards 2006 (Continued)

Treatment protocol: participants read online non-CBT panic resources and provided weekly status re-
ports to a therapist via e-mail
Therapists: 1 doctoral clinical psychology student
Therapist contact: limited to weekly status update e-mails
Dropout: n = 2; 22.2%
Primary outcomes:
(1) panic and agoraphobia symptoms: Panic Disorder Severity Scale; Body Vigilance Scale; Agoraphobic
Cognitions Questionnaire; Anxiety Sensitivity Profile
(2) general anxiety: DASS Stress and Anxiety subscales
(3) clinically important improvement: ADIS-IV
Secondary outcome:
(1) quality of life: WHO Quality of Life subscales
Notes *on-line treatment program: Panic Online
Risk of bias
Random sequence genera- Unclear risk ABC block randomisation was used (information provided by authors via per-
tion (selection bias) sonal correspondence); unclear if sequential design or a more rigorous ran-
domisation method was used

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based without stress management
mance bias) versus Internet-based with stress management versus Internet-based informa-
All outcomes tion control)
Blinding of outcome as- High risk Quote: "The two assessors were the second author of the present study and
sessment (detection bias) a probationary registered psychologist/PhD candidate. The second author
Observer/Interview-Rated was not blind to treatment allocation, although the other assessor was... To
Outcomes evaluate reliability of assessment, a third assessor (the 3rd author), who was
blind to the treatment allocation, reviewed 15% of the clinical interviews...";
"The two clinicians who conducted the assessments did not provide any treat-
ment."
Comment: not all interviewers were blind to treatment condition
Incomplete outcome data Low risk Quote: "The attrition rate for PO1 was 16.7% (2/12), 9% (1/11) in PO2 and 22%
(attrition bias) (2/9) in IC. Reasons given for discontinuing treatment in the PO1 condition
All outcomes were a lack of motivation or an episode of major depression. The PO2 person
discontinued because of a wish to commence selective serotonin reuptake in-
hibitor medication halfway through the study. Of the two IC participants, no
Informed decisions.

reason for discontinuing was given."; "Data analysis involved intention-to-
treat analyses."
porting bias)
Other bias High risk At baseline, treatment groups scored significantly higher on the DASS depres-
sion subscale than control participants
Robinson 2010
N: 101
Age: M = 46.96 (SD = 12.70); range = 18 to 80
Sex: 68.3% women
Baseline depression severity: (PHQ-9) ICBT with clinician M = 11.40 (SD = 4.63); WLC M = 12.5 (SD =
4.73)
Treatment protocol*: participants completed online modules on cognitive restructuring, challenging

core beliefs, and exposure, with e-mail and phone support from a therapist for module activities
Therapists: 1 registered clinical psychologist
Therapist contact: M e-mails or calls by therapist = 33.2 (SD = 4); M time spent by therapist per partici-
pant = 80.8 min (SD = 22.6)
(2) Internet-based CBT with e-mail and phone support from a technician (n = 50)
Treatment protocol*: participants completed online modules on cognitive restructuring, challenging

core beliefs, and exposure, with e-mail and phone encouragement and instructions from a technician
Informed decisions.
Robinson 2010 (Continued)

Therapists: no therapist; clinic manager acted as technician
Therapist contact: none; M e-mails or calls by technician = 31.1 (SD =3.1); M time spent by clinician per
participant = 74.5 min (SD = 7.8)
Dropout: n = 5; 10%
Face-to-Face Contact: none
Duration: 11 weeks
Primary outcomes:
(1) generalized anxiety symptoms: Penn State Worry Questionnaire
(2) clinically important improvement: GAD-7
Secondary outcomes:
(2) treatment satisfaction (at post-treatment): A 7-item questionnaire based on the Credibility/Ex-
pectancy Questionnaire
Notes *on-line treatment program: Worry Program - Titov N, Andrews G, Robinson E, Schwencke G, Johnston
L, et al. (2009). Clinician-assisted Internet-based treatment is effective for generalized anxiety disorder:
a randomised controlled trial. Australian and New Zealand Journal of Psychiatry, 43, 905–912.
Risk of bias
Random sequence genera- Low risk Quote: "The 150 people accepted into the program were randomised by NT
tion (selection bias) [2nd author] via a true randomisation process (www.random.org)..."
Allocation concealment Unclear risk Quote: "Allocation preceded the diagnostic telephone call."
(selection bias) Comment: insufficient detail about method of allocation concealment provid-
ed to determine risk
and personnel (perfor- to the treatment they delivered (Internet-based CBT clinician versus Inter-
mance bias) net-based CBT technician versus waiting list)
All outcomes
Incomplete outcome data Low risk Quote: "Post-treatment data was collected from 45 (90%) TA, 46 (98%) CA
(attrition bias) group members, and from 47/48 (98%) of control group participants."; "In ac-
All outcomes cordance with the ITT and LOCF paradigm..."
Comment: a small and similar number of participants from both treatment
conditions did not complete post-treatment measures; ITT analyses were used
Informed decisions.
Robinson 2010 (Continued)
porting bias) ed
Other bias Unclear risk There were significant differences in marital status and age between the con-
trol and treatment groups at baseline
Silfvernagel 2012
Participants Diagnosis: DSM-IV Anxiety Disorder
N: 57
Age: M = 32.4 (SD = 6.9); range = 20 to 45
Sex: 65% women
Psychiatric co-morbidity: 32%
Co-use of medication: included if stable dose for past 3 months
Baseline depression severity: (MADRS-S) ICBT M = 15.81 (SD = 7.35); WLC M = 17.93 (SD = 8.38)
(1) Internet-based CBT with e-mail support from a clinician (n = 29)

turing, and exposure, with e-mail support from a therapist for module activities
Therapists: 3 clinical psychology master's students; completed clinical training; supervised by experi-
enced clinical psychologists
Therapist contact: 15 min/week; approximately 19 e-mail exchanges between therapist and partici-
pant during treatment
Dropout: n = 10; 34.5%
Duration: 10 weeks
Primary outcomes:
Informed decisions.
Silfvernagel 2012 (Continued)

(1) panic symptoms: Panic Disorder Severity Scale
(2) general anxiety: Beck Anxiety Inventory; Clinical Outcomes in Routine Evaluation - Outcome Mea-
sure
Secondary outcome:
Notes *treatment based on Internet-based programs described in: Andersson, G., Carlbring, P., Holmström, A.,
Sparthan, E., Furmark, T., Nilsson-Ihrfelt, E., et al. (2006). Internet-based self-help with therapist feed-
back and invivo group exposure for social phobia: a randomised controlled trial. Journal of Consulting
and Clinical Psychology, 74, 677-686.; Carlbring, P., Westling, B.E., Ljungstrand, P., Ekselius, L, & Anders-
son, G. (2001). Treatment of panic disorder via the Internet: A randomised trial of a self-help program.
Behavior Therapy, 32, 751-764.; AND Vernmark, K., Lenndin, J., Bjärehed, J., Carlsson, M., Karlsson, J.,
Öberg, J., et al. (2010). Internet administered guided self-help versus individualized e-mail therapy:
a randomised trial of two versions of CBT for major depression. Behaviour Research and Therapy, 48,
368-376.; Carlbring, P., Maurin, L., Törngren, C., Linna, E., Eriksson, T., Sparthan, E., et al. (2011). Individ-
ually-tailored, Internet-based treatment for anxiety disorders: A randomised controlled trial. Behaviour
Research and Therapy, 49, 18-24.; Andersson, G., Estling, F., Jakobsson, E., Cuijpers, P., & Carlbring, P.
(2011). Can the patient decide which modules to endorse? An open trial of tailored internet treatment
of anxiety disorders. Cognitive Behavior Therapy, 40, 57-64.
Risk of bias
Random sequence genera- Low risk Quote: "The participants were divided into two groups so that the two prede-
tion (selection bias) termined age groups 18–30 years (young adults) and 31–45 years(adults) were
equally represented in each condition. The blocked randomizations process
was conducted through an online true random number-generation service
(random.org) independent of the investigators and therapists."
Allocation concealment Unclear risk Quote: "The blocked randomizations process was conducted...independent of
(selection bias) the investigators and therapists."
Comment: no more specific mention of allocation concealment present
and personnel (perfor- to the treatment they delivered (Internet-based CBT clinician versus Inter-
mance bias) net-based CBT technician versus waiting list)
All outcomes
Blinding of outcome as- Low risk Quote: "At posttreatment participants were instructed via email to complete
sessment (detection bias) the follow-up questionnaires and to participate in a semistructured telephone
Observer/Interview-Rated interview carried out by a blinded assessor who had no earlier contact with
Outcomes the participants."
Comment: assessors were blind to treatment condition
Incomplete outcome data Low risk Quote: "A mixed-models approach with an unstructured covariance structure
(attrition bias) was endorsed as a way to handle missing data at posttreatment."
All outcomes Comment: ITT analysis was used
porting bias) ed
Informed decisions.
Silfvernagel 2012 (Continued)
Spence 2011
Participants Diagnosis: DSM-IV Post-traumatic Stress Disorder
Method of Diagnosis: MINI
N: 44
Age: M = 42.6 (SD = 13.1); range = 21 to 68
Sex: 81% women
Psychiatric co-morbidity: 62% MDD, 33% Social Phobia, 31% PD with or without Agoraphobia, 26%
GAD, 17% OCD
Baseline depression severity: (PHQ-9) ICBT M = 15.61 (SD = 7.35); WLC CBT M = 15.05 (SD = 4.9)
Treatment protocol*: participants completed online modules on psychoeducation, de-arousal strate-

gies, cognitive restructuring, graded exposure, and relapse prevention with e-mail and phone support
from a therapist for module activities
Therapists: 1 clinical psychologist
Therapist contact: M e-mails by therapist = 5.39 (SD = 3.54); M phone calls by therapist = 7.87 (SD =
2.56); M time spent by therapist per participant = 103.91 min (SD = 96.53)
Duration: 8 weeks
Dropout: n = 3; 14.3%
Primary outcomes:
(1) post-traumatic stress symptoms: Post-traumatic Stress Disorder Checklist – Civilian
Informed decisions.
Spence 2011 (Continued)

(2) general anxiety symptoms: GAD-7
(3) clinically important improvement: Post-traumatic Stress Disorder Checklist – Civilian
Secondary outcomes:
(2) treatment satisfaction (at post-treatment): a 7-item questionnaire based on the Credibility/Ex-
Notes *treatment is based on: Andrews, G. (2003). The treatment of anxiety disorders: Clinician guides and
patient manuals. Cambridge: Cambridge University Press.; Perini, S., Titov, N., & Andrews, G. (2008).
The climate sadness program of Internet-based treatment for depression: A pilot study. Journal of Ap-
plied Psychology, 4, 18-24.; Robinson, E., Titov, N., Andrews, G., McIntyre, K., Schwencke, G., & Solley,
K. (2010). Internet treatment for generalized anxiety disorder: A randomised controlled trial compar-
ing clinician vs. technician assistance. PLoS ONE, 5, e10942.; Wims, E., Titov, N., Andrews, G., & Choi,
I. (2010). Clinician-assisted Internet-based treatment is effective for panic: A randomised controlled
trial. Australian and New Zealand Journal of Psychiatry, 44, 599-607.; Titov, N., Andrews, G., Johnston,
L., Robinson, E., & Spence, J. (2010). Transdiagnostic Internet treatment for anxiety disorders: A ran-
domised controlled trial. Behaviour Research and Therapy, 48, 890-9.
Risk of bias
Random sequence genera- Low risk Quote: "...randomised via a true randomizations process (www.random.org),
tion (selection bias) generated by an independent person..."

(selection bias)
mance bias) group)
All outcomes
Blinding of outcome as- High risk Quote: "The assessments were conducted by JS and KS, who were not blind to
sessment (detection bias) the participants' condition."
Observer/Interview-Rated Comment: interviewers were not blind to treatment condition
Outcomes
Incomplete outcome data Low risk Quote: "All post-treatment analyses involved a conservative intention-to-treat
(attrition bias) (ITT) design where missing data was addressed by carrying forward the first
All outcomes available data (i.e. Baseline-observation-carried-forward model)."; "Five par-
ticipants did not complete the program: one for unknown reasons; three be-
cause of competing time commitments; and one because of a relapse of de-
pressive symptoms. There were no formal withdrawals during the treatment
program."; "Post-treatment data were collected from 21/23 (91%) Treatment
and 18/21 (86%) Control group participants."
Comment: very little data were missing; ITT analyses were used
porting bias)
Informed decisions.
Spence 2011 (Continued)
Tillfors 2008
N: 38
Age: for ICBT, M = 32.3 (SD = 9.7); for ICBT + exposure, M = 30.4 (SD = 6.3); range = 19 to 53
Sex: 78.9% women
Baseline depression severity: (MADRS-S) ICBT M = 11.3 (SD = 7.3); ICBT + exposure M = 12.4 (SD = 6.4)
Duration: 9 online modules completed in 9 weeks

activities, and participated in an online discussion forum
Therapists: 2 licensed clinical psychologists (research or clinical experience, or both, in social pho-
bia), 2 clinical psychology students in final year of master's program; supervised by licensed CBT psy-
chotherapist
(2) Internet-based CBT with e-mail support and face-to-face exposure (n = 19)
Duration: 9 online modules and 5 face-to-face group exposure sessions over 9 weeks
Treatment protocol*^: participants completed online modules on psychoeducation, cognitive re-

structuring, exposure, social skills, and relapse prevention, with e-mail support from a therapist for
module activities, and participated in an online discussion forum as well as attending 5 face-to-face
group therapy sessions
Therapists: 2 licensed clinical psychologists, 2 psychologist candidates
Therapist contact: 35 min per week by e-mail and 5 x 2.25 hr exposure sessions
Face-to-face contact: 5 x 2.25 hr exposure sessions
Informed decisions.
Tillfors 2008 (Continued)

Primary outcomes:
Secondary outcomes:
(2) treatment satisfaction (at post-treatment): participants reported on the quality of the overall treat-
ment, its components, and its tempo as well as perceptions of their own improvement
Notes *treatment based on: Rodebaugh, T.L., Holaway, R.M., & Heimberg, R.G. (2004). The treatment of social
anxiety disorder. Clinical Psychology Review, 24, 883–908. AND Clark, D.M., & Wells, A. (1995). A cognitive
model of social phobia. In R.G. Heimberg, M.R. Liebowitz, D.A. Hope, & F.R. Schneier (Eds.), Social pho-
bia: Diagnosis, assessment and treatment (pp. 69-93). New York, NY: Guilford Press.
^exposure sessions based on Heimberg, R.G., & Becker, R.E. (2002). Cognitive-behavioral group therapy
for social phobia: basic mechanisms and clinical strategies. New York, NY: Guilford Press.
Risk of bias
Random sequence genera- Unclear risk Quote: "...38 were eventually randomised into either..."
tion (selection bias) Comment: no information on method of randomisation provided

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based CBT versus Internet-based
mance bias) CBT plus live exposure)
All outcomes
Incomplete outcome data Low risk Quote: "At post-test, all participants except one answered their computerized
(attrition bias) questionnaires. The pre-test score of that participant was carried forward to
All outcomes the post-test assessment point (e.g., last observation carried forward)."
Comment: though there were a number of participants who did not complete
all treatment modules (n = 10 ICBT + Exp; n = 9 ICBT) the numbers were rela-
tively equal across conditions and participants still provided post-treatment
data; ITT analyses used
porting bias)
Informed decisions.
Titov 2008a
Method of diagnosis: CIDI
N: 105
Age: M = 38.13 (SD = 12.24); range = 18 to 72
Sex: 59% women
Baseline depression severity: (PHQ-9) ICBT M = 8.0 (SD = 4.95); WLC M = 8.02 (SD = 5.32)

turing, exposure, and relapse prevention, with email support from a therapist for module activities
Therapist contact: M time spent by therapist per participant = 125 min (SD = 25)
Dropout: n = 6; 12%
Duration: 10 weeks
Dropout: n = 6; 10.9%
Primary outcome:
(1) social phobia symptoms: Social Interaction Anxiety Scale; Social Phobia Scale
Secondary outcomes:
(1) quality of life: WHO Disability Assessment Schedule
Notes *online treatment program: Shyness Programme (based on CLIMATEGP program written by Drobny
and Einstein)
Informed decisions.
Titov 2008a (Continued)
Risk of bias
Random sequence genera- Low risk Quote: "...were randomised via a true randomizations process (www.ran-
tion (selection bias) dom.org) to either..."

(selection bias)
mance bias)
All outcomes
Incomplete outcome data Low risk Quote: "Eleven members of the treatment group (22%) failed to complete all
(attrition bias) six lessons within the required time frame. Of these non-completers, two for-
All outcomes mally withdrew citing lack of time and motivation after experiencing a death
or illness in the family; one reported that the exposure exercises were too anx-
iety provoking; one reported he did not find the programme helpful; one re-
ported taking an overseas holiday; three cited a change in work or study com-
mitments affecting their ability to complete the programme requirements; one
reported complications due to her pregnancy and two did not give a reason.";
"Post-treatment data were collected from 93 participants (44/50 treatment
group participants and 49/49 waitlist control group participants). In accor-
dance with the intention-to-treat paradigm, the pre-treatment scores of these
six participants who did not complete the post-treatment questionnaires were
replicated as their post-treatment scores."
Comment: there were a number of dropouts from the treatment group, how-
ever some of these dropouts still provided post-treatment data; ITT analyses
were used
Selective reporting (re- Unclear risk Results for one outcome measure outlined in the trial registry (GAD-7) are not
porting bias) reported
Titov 2008b
N: 88
Age: M = 36.79 (SD = 10.93); range = 20 to 61
Sex: 62.96% women
Informed decisions.
Titov 2008b (Continued)


Therapist contact: M time spent by therapist per participant = 126.76 min (SD = 30.89)
Dropout: n = 5; 11.6%
Duration: 10 weeks
Dropout: n = 5; 11.1%
Primary outcome:
Secondary outcomes:
(1) quality of life: WHO Disability Assessment Schedule
and Einstein)
Risk of bias
Random sequence genera- Low risk Quote: "The 88 people accepted into the programme were randomised via a
tion (selection bias) true randomizations process (www.random.org) to either..."

(selection bias)
Informed decisions.
Titov 2008b (Continued)
mance bias)
All outcomes
Incomplete outcome data Low risk Quote: "Eight members of the treatment group (20%) failed to complete all
(attrition bias) six lessons within the require time frame. Of these eight non-completers, one
All outcomes said the programme was not helpful, and one reported they had improved
sufficiently."; "...post-treatment data were collected from 78 participants
(38/41 treatment group participants and 40/40 waitlist control group partici-
pants). In accordance with the intention-to-treat paradigm, the pre-treatment
scores...were replicated as their post-treatment scores."
Selective reporting (re- Unclear risk Results for one outcome measure outlined in the trial registry (GAD-7) are not
porting bias) reported
Titov 2008c
N: 98
Age: M = 37.97 (SD = 11.29); range = 18 to 64
Sex: 61.05% women
Baseline depression severity: (PHQ-9) ICBT M = 7.65 (SD = 4.72); Unguided ICBT M = 7.0 (SD = 5.27);
WLC M = 7.03 (SD = 5.28)

Informed decisions.
Titov 2008c (Continued)

Therapists: 2 clinical psychologists
Therapist contact: M time spent by therapist per participant = 168 minutes (SD = 40)
(2) Internet-based CBT (n = 31)

turing, exposure, and relapse prevention independently
Dropout: n = 4; 12.9%
Duration: 10 weeks
Primary outcome:
Secondary outcomes:
(2) treatment satisfaction: a 7-item questionnaire based on the Credibility/Expectancy Questionnaire
and Einstein)
Risk of bias
Random sequence genera- Low risk Quote: "The 98 people accepted into the programme were randomised via a
tion (selection bias) true randomizations process (www.random.org) to either..."

(selection bias)
and personnel (perfor- to the treatment they delivered (clinician-assisted computerized CBT versus
mance bias) non-clinician-assisted computerized CBT versus waiting list)
All outcomes
Informed decisions.
Titov 2008c (Continued)
Incomplete outcome data Low risk Quote: "Post-treatment data were collected from 91 participants (30/31 CaC-
(attrition bias) CBT group participants, 27/30 CCBT group participants, and from 34/34 con-
All outcomes trol group participants). In accordance with the ITT paradigm, the pre-treat-
ment scores of the four participants who did not complete their post-treat-
ment questionnaires were replicated as their post-treatment scores."
Selective reporting (re- Unclear risk Results for one outcome measures (GAD-7) outlined in the trial registration
porting bias) were not reported; all other outcome measures outlined in the trial registra-
tion were reported
Titov 2009
N: 48
Age: M = 44 (SD = 12.98)
Sex: 76% women

turing, exposure, core beliefs, and relapse prevention, with e-mail and phone support from a therapist
for module activities
Therapist contact: M e-mails by therapist = 23.7; M telephone calls by therapist = 4.1; M instant mes-
sages by therapist = 5.5; M time spent by therapist per participant = 130 min
Dropout: n = 5; 20%
Informed decisions.
Titov 2009 (Continued)

Duration: 9 weeks
Dropout: n = 4; 17.4%
Primary outcomes:
(1) general anxiety symptoms: Penn State Worry Questionnaire
Secondary outcomes:
(2) treatment satisfaction: A 7-item questionnaire based on the Credibility/ Expectancy Questionnaire
Notes *online treatment program: Worry Programme, developed for this study
Risk of bias
Random sequence genera- Low risk Quote: "The 48 people accepted into the programme were randomised by NT
tion (selection bias) [Nickolai Titov] via a true randomizations process (www.random.org) to ei-
ther..."
Allocation concealment Unclear risk Quote: "Allocation preceded the screening phone call."
(selection bias) Comment: unclear if allocation was kept concealed from screener
mance bias)
All outcomes
Incomplete outcome data Low risk Quote: "Post-treatment data were collected from 21/24 (88%) treatment group
(attrition bias) participants and 19/21 (90%) of control group participants. In accordance with
All outcomes the ITT paradigm, the pre-treatment scores of the five participants who did
not complete the post-treatment questionnaires were replicated as their post-
treatment scores."
porting bias) ed
Informed decisions.
Titov 2010
Participants Diagnosis: DSM-IV Panic Disorder with Agoraphobia (26.9%), Social Phobia (29.5%), Generalized Anxi-
ety Disorder (43.6%)
N: 86
Age: M = 39.5 (SD = 13)
Sex: 67.9% women
Psychiatric co-morbidity: 28.2% another Anxiety Disorder only, 20.5% another Affective Disorder only,
26.9% another Anxiety and Affective Disorder
(1) Internet-based CBT with email and phone support (n = 42)
Treatment protocol*: participants completed online modules on disorder-specific psychoeduca-

tion, cognitive restructuring, exposure, assertiveness training, and relapse prevention, with email and
phone support from a therapist for module activities
Therapist contact: M e-mails by therapist = 23.6; M time spent by therapist per participant = 46 min (SD
= 16)
Dropout: n = 6; 14.3%
(2) Waiting list control (n=44)
Duration: 8 weeks
Dropout: n = 8; 18.2%
Primary outcomes:
(1) disorder-specific symptoms: Penn State Worry Questionnaire; Social Phobia Screening Question-
naire; Panic Disorder Severity Scale – Self-Rating
Secondary outcomes:
Informed decisions.

Notes *online treatment program: Anxiety Programme, developed for this study
Risk of bias
Random sequence genera- Low risk Quote: "Eighty six applicants met all inclusion criteria and were randomised by
tion (selection bias) NT [Nickolai Titov] via a true randomizations process (www.random.org)..."
(selection bias) Comment: unclear if screener was aware of group allocation
and personnel (perfor- to the treatment they delivered (Internet-based CBT vs. waiting list)
mance bias)
All outcomes
Incomplete outcome data Low risk Quote: "Post-treatment data (Time 2) was collected from 38/40 (95%) treat-
(attrition bias) ment group participants and 40/40 (100%) control group participants... In ac-
All outcomes cordance with the ITT and BOCF principles, the pre-treatment scores of partic-
ipants who did not complete the post-treatment... questionnaires were repli-
cated as their post-treatment... scores."
Selective reporting (re- High risk Results for several outcome measures (Beck Anxiety Inventory, Social Phobia
porting bias) Scale, Social Interaction Anxiety Scale, Agoraphobic Cognitions Questionnaire,
Body Vigilance Scale, WHO Disability Assessment Schedule II) outlined in the
trial registration were not reported and other scales not in the trial registration
(Penn State Worry Questionnaire, Social Phobia Screening Questionnaire, Pan-
ic Disorder Severity Scale) were reported
Titov 2010 GAD

N: 34
Age^: M = 39.5 (SD = 13)
Sex^: 67.9% women
Informed decisions.
Titov 2010 GAD (Continued)

Psychiatric co-morbidity^: 28.2% another Anxiety Disorder only, 20.5% another Affective Disorder on-
ly, 26.9% another Anxiety and Affective Disorder
Co-use of medication^: 47.4%
Baseline depression severity^: (PHQ-9) ICBT M = 10.77 (SD = 5.20); WLC M = 10.84 (SD = 6.26)
(1) Internet-based CBT with email and phone support (n = 18)

turing, exposure, assertiveness training, and relapse prevention, with email and phone support from a
Therapist contact^: M emails by therapist = 23.6; M time spent by therapist per participant = 46 min
(SD = 16)
Dropout^: n = 6; 14.3%
Duration: 8 weeks
Dropout^: n = 8; 18.2%
Primary outcomes:
(1) generalized anxiety disorder symptoms: Penn State Worry Questionnaire
Secondary outcomes:
^statistics for entire Titov 2010 sample
Risk of bias
Informed decisions.
Titov 2010 GAD (Continued)
mance bias)
All outcomes
Selective reporting (re- High risk Results for several outcome measures outlined in the trial registration were
porting bias) not reported
Titov 2010 Panic

Participants Diagnosis: DSM-IV Panic Disorder with Agoraphobia
N: 21
Age^: M = 39.5 (SD = 13)
Sex^: 67.9% women
Informed decisions.
Titov 2010 Panic (Continued)

turing, exposure, assertiveness training, and relapse prevention, with e-mail and phone support from a
Therapist contact^: M e-mails by therapist = 23.6; M time spent by therapist per participant = 46 min
(SD = 16)
Dropout^: n = 6; 14.3%
Duration: 8 weeks
Dropout^: n = 8; 18.2%
Primary outcomes:
(1) panic symptoms: Panic Disorder Severity Scale – Self-Rating
Secondary outcomes:
Risk of bias
mance bias)
All outcomes
Informed decisions.
Titov 2010 Panic (Continued)

Titov 2010 Social Phobia

N: 23
Age^: M = 39.5 (SD = 13)
Sex^: 67.9% women

turing, exposure, assertiveness training, and relapse prevention, with e-mail and phone support from a
Therapist contact^: M e-mails by therapist = 23.6; M time spent by therapist per participant = 46 min
(SD = 16)
Dropout^: n = 6; 14.3%
Duration: 8 weeks
Informed decisions.
Titov 2010 Social Phobia (Continued)

Dropout^: n = 8; 18.2%
Primary outcomes:
(1) social phobia symptoms: Social Phobia Screening Questionnaire
Secondary outcomes:
(2) treatment satisfaction: a 7-item questionnaire based on the Credibility/ Expectancy Questionnaire
Risk of bias
mance bias)
All outcomes
Titov 2011
Informed decisions.
Participants Diagnosis: DSM-IV Panic Disorder with or without Agoraphobia (10%), Social Phobia (11%), General-
ized Anxiety Disorder (28%), MDD (51%; not included in review)
N: 74
Age: M = 43.9 (SD = 14.6); range = 18 to 79
Sex: 73% women
Psychiatric co-morbidity: 81% had another co-morbid Anxiety or Depressive Disorder

turing, de-arousal strategies, behavioural activation, exposure, challenging core beliefs, and relapse
prevention, with e-mail and phone support from a therapist for module activities
Therapist contact: M e-mails by therapist = 5.45 (SD = 3.57); M phone calls by therapist = 9.35 (SD =
2.96); M time spent by therapist per participant = 84.76 min (SD = 50.37)
Duration: 10 weeks
Primary outcomes:
(1) disorder-specific symptoms: Penn State Worry Questionnaire; Social Phobia – 12; Panic Disorder
Severity Scale
(2) general anxiety: GAD-7, Depression Anxiety Stress Scales - 21
Secondary outcomes:
Informed decisions.
Notes *treatment based on: Andrews, G., Creamer, M., Crino, R., Hunt, C., Lampe, L., & Page, A. (2003). The
treatment of anxiety disorders: Clinician guides and patient manuals (2nd ed.). UK: University Press,
Cambridge.; Perini, S., Titov, N., & Andrews, G. (2009). Clinician-assisted Internet-based treatment is
effective for depression: A randomised controlled trial. Australian and New Zealand Journal of Psychi-
atry, 43, 571-8.; Titov, N., Andrews, G., Davies, M., McIntyre, K., Robinson, E., & Solley, K. (2010). Inter-
net treatment for depression: A randomised controlled trial comparing clinician vs. technician assis-
tance. PLoS ONE, 5, e10939.; Robinson, E., Titov, N., Andrews, G., McIntyre, K., Schwencke, G., & Solley,
K. (2010). Internet treatment for generalized anxiety disorder: A randomised controlled trial comparing
clinician vs. technician assistance. PLoS ONE, 5, e10942.; Titov, N., Andrews, G., Schwencke, G., Drobny,
J., & Einstein, D. (2008). Shyness 1: Distance treatment of social phobia over the internet. Australian and
New Zealand Journal of Psychiatry, 42, 585-94.; Wims, E., Titov, N., Andrews, G., & Choi, I. (2010). Clini-
cian-assisted internet-based treatment is effective for panic: A randomised controlled trial. Australian
and New Zealand Journal of Psychiatry, 44, 599-607.
Risk of bias
Random sequence genera- Low risk Quote: "...randomised via a true randomisation process (www.random.org),
tion (selection bias) generated by an independent person..."
Comment: adequate randomisation method"

(selection bias)
and personnel (perfor- to the treatment they delivered (Internet-based CBT vs. waiting list control)
mance bias)
All outcomes
Blinding of outcome as- High risk Quote: "These [diagnostic] assessments were conducted by BFD and GS, who
sessment (detection bias) were not blind to participant's condition."
Observer/Interview-Rated Comment: interviewers were not blind to treatment condition
Outcomes
Incomplete outcome data Low risk Quote: "All post-treatment analyses involved a conservative intention-to-treat
(attrition bias) (ITT) design where missing data was addressed by carrying forward the first
All outcomes available data (i.e., baseline-observation-carried-forward model; BOCF)."
Comment: there were only four formal withdrawals from the study; ITT analy-
ses were used
porting bias) ed
Other bias High risk Treatment group endorsed significantly higher PDSS-SR scores than controls
at baseline
van Ballegooijen 2013

Informed decisions.
van Ballegooijen 2013 (Continued)
Participants Diagnosis: DSM-IV Panic Disorder with or without Agoraphobia (78%), Agoraphobia without Panic
(14%)
N: 126
Age: M = 36.6 (SD = 11.4); range = 18 to 67
Sex: 67.5% women
Country of residence: Netherlands
Co-use of adjunct therapy: included
Baseline depression severity: (CES-D) ICBT M = 20.0 (SD = 9.1); WLC M = 21.6 (SD = 9.0)
Treatment protocol*: participants completed online modules on various cognitive and behavioural
techniques and skills, with e-mail support from a therapist for module activities
Therapists: master's level clinical psychology students; supervised by clinical psychologist
Therapist contact: M time spent by therapist per participant = 1 to 2 hours
Dropout: n = 29; 46%
Duration: 12 weeks
Treatment protocol: participants had access to online non-CBT panic resources
Dropout: n = 24; 38.1%
Primary outcomes:
(1) disorder-specific symptoms: Panic Disorder Severity Scale
Notes *treatment called Don't Panic Online, described in: van Ballegooijen, W., Riper, H., van Straten, A.,
Kramer, J., Conijn, B., & Cuijpers, P. (2011). The effects of an Internet based self-help course for reducing
panic symptoms--Don't Panic Online: Study protocol for a randomised controlled trial. Trials, 12, 75.
Risk of bias
Informed decisions.
van Ballegooijen 2013 (Continued)
Random sequence genera- Low risk Quote: "After the interview, all participants were randomized to 1 of the 2
tion (selection bias) groups. Randomization was stratified for the presence or absence of agora-
phobic symptoms (PDSS-SR item 4 score ≥2) and the use of antidepressants or
sedatives. Randomization lists were generated automatically using a comput-
er program."

(selection bias)
Blinding of participants High risk Not possible to blind participants to the treatment condition they were in (In-
and personnel (perfor- ternet-based CBT or waiting list control)
mance bias)
All outcomes
Incomplete outcome data High risk Significant drop out in both conditions; ITT analyses were employed via multi-
(attrition bias) ple imputation
All outcomes
Selective reporting (re- Unclear risk One outcome - quality of life, as measured by the EuroQol Questionnaire - out-
porting bias) lined in the published study protocol was not reported; results for all other
outcome measures outlined in the study protocol were reported
Wims 2010
Participants Diagnosis: DSM-IV Panic Disorder with or without Agoraphobia
N: 2759
Age: M = 42.08 (SD = 12.29)
Sex: 76% women
Psychiatric co-morbidity: 21% Social Phobia, 31% GAD, 10% OCD, 7% PTSD, 21% Major Depressive
episode
Baseline depression aeverity: (PHQ-9) ICBT M = 10.34 (SD = 4.09); WLC M = 10.24 (SD = 5.93)

Informed decisions.
Wims 2010 (Continued)


turing, exposure, physiological de-arousal, and relapse prevention, with email support from a therapist
for module activities
Therapists: 1 psychiatry registrar
Therapist contact: M e-mails by therapist = 7.5; M time spent by therapist per participant = 75 min
Dropout: n = 10; 31.3%
Duration: 8 weeks
Dropout: n = 5; 18.5%
Primary outcome:
(1) panic and agoraphobia symptoms: Panic Disorder Severity Scale; Body Sensations Questionnaire;
Agoraphobic Cognitions Questionnaire; Mobility Inventory
Secondary outcome:
Notes *online treatment program: Panic Program - Wims E, Titov N, Andrews G. (2008). The Panic program: An
open trial
of Internet-based treatment for panic disorder. Electronic Journal of Applied Psychology, 4, 2.
Risk of bias
Random sequence genera- Low risk Quote: "The 59 people accepted into the program were randomised via a true
tion (selection bias) randomizations process (www.random.org) to either..."

(selection bias)
All outcomes
Blinding of outcome as- High risk Quote: "...the lack of blinding in the administration of the PDSS is a source of
sessment (detection bias) bias, which may account for the larger effect sizes in this domain."
Comment: interviewers were not blind to treatment condition
Informed decisions.
Wims 2010 (Continued)

Observer/Interview-Rated
Outcomes
Incomplete outcome data Low risk Quote: "Two participants found the program required more time than they
(attrition bias) were able to set aside, one person dropped out due to increased severity of
All outcomes their anxiety which required inpatient admission, another became ill, another
found the course too difficult and the final participant moved house during the
program and no longer had internet access."; "Post-treatment data was col-
lected from 44 participants (22/29 treatment group and 22/25 waitlist control
group). In accordance with the intention-to-treat paradigm, the pre-treatment
scores of the participants who did not complete the post-treatment question-
naires were replicated as their post-treatment scores."
porting bias) ed
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Andersson 2006 The intervention involved too much face-to-face contact between therapist and participant (i.e.,
two live exposure sessions)
Andersson 2012c The comparison condition was non-directive supportive therapy and thus too active for the
present comparisons of interest
Andrews 2011 A standardized diagnostic instrument was not used to assess participants for an anxiety disorder
Bell 2012 The intervention did not involve therapist support
Carlbring 2003 The comparison was active applied relaxation and did not fit into one of the present comparisons
Carlbring 2010 The comparator condition included attention bias modification and so did not fit in any of our
comparator categories and was not appropriate for inclusion
Carlbring 2011b Both of the treatment conditions in this study qualified as our intervention of interest, so no appro-
priate comparator
Cunningham 2006 Participants did not meet DSM or ICD criteria for an anxiety disorder diagnosis
Ellis 2011 Participants were not diagnosed with a DSM or ICD anxiety disorder by study investigators
Febbraro 2005 Only a portion of the sample was diagnosed with PD and it could not be separated from the rest of
the sample
Gilson 2006 The comparator condition involved internet-based CBT delivered by a physician (vs. a psycholo-
gist) and so was not sufficiently different from the intervention of interest to be appropriate for in-
clusion
Greist 2002 The intervention was not delivered directly by a therapist but instead used voice response technol-
ogy
Informed decisions.
Study Reason for exclusion
Kenardy 2003 The intervention was computer-augmented; there were six face-to-face sessions between therapist
and participant
Kenwright 2005 The comparator condition was also therapist-delivered distance CBT and so was not sufficiently
different from the intervention of interest to be appropriate for inclusion
Klein 2001 The intervention was not therapist-delivered (i.e., was entirely self-help)
Klein 2006 Participants were randomised with 'sequential randomisation', which is more accurately described
as sequential allocation with no randomisation
Klein 2009 The comparator condition was also therapist-delivered distance CBT and so was not sufficiently
Knaevelsrud 2007 Participants were not diagnosed with a DSM or ICD anxiety disorder by study investigators
Lange 2001 Participants were not diagnosed with a DSM or ICD anxiety disorder by study investigators
Lange 2003 Participants were not diagnosed with a DSM or ICD anxiety disorder by study investigators
Litz 2007 The comparison was active online supportive counselling and thus too active for the present com-
parisons
Marks 2004 The intervention involved too much face-to-face contact between therapist and participant (i.e., in
addition to the computer-based session, each session involved 15 minutes of face-to-face contact)
Newman 1997 The intervention was computer-augmented; there were four face-to-face sessions between thera-
pist and participant
Pittaway 2009 Participants were not diagnosed with a DSM or ICD anxiety disorder by study investigators
Ruwaard 2010 Participants had panic attacks but were not diagnosed with PD
Saul 2007 Participants were not diagnosed with a DSM or ICD anxiety disorder by study investigators
Schneider 2005 The comparator condition was also distance CBT delivered by a physician, just without exposure,
and so was not significantly different from the intervention of interest to be appropriate for inclu-
sion
Shandley 2008 The comparator condition was also distance CBT delivered by a physician and so was not suffi-
ciently different from the intervention of interest to be appropriate for inclusion
Titov 2009b The comparator condition was also therapist-delivered distance CBT and so was not sufficiently
van Straten 2008 Participants were not diagnosed with a DSM or ICD anxiety disorder by study investigators
Wagner 2012 Participants did not have to have a DSM or ICD anxiety disorder diagnosis to participate in this in-
vestigation
Characteristics of studies awaiting assessment [ordered by study ID]
Informed decisions.
Andersson 2013
Participants Diagnosis: DSM-IV Specific Phobia, Snake Type
N: 30
Sex: 84.6% women
Baseline depression severity: (BDI-II) ICBT M = 3.6 (SD = 3.4); Live exposure M = 6.8 (SD = 3.3)
(1) Internet-based BT with e-mail support (n = 15)
Treatment protocol: participants completed online modules on psychoeducation and exposure,

with e-mail support from a therapist for module exercises
Therapists: two clinical psychology students in their last year, one PhD student in clinical psychol-
ogy, and one licensed psychologist
Therapist contact: 25 min per participant
Dropout: n = 2; 13%
(2) Live exposure (n = 15)
Duration: 1 face-to-face session
Treatment protocol*: participants attended an orientation session and 3 hr graded exposure ses-
sion with a therapist
Therapists: two clinical psychology students in their last year, one PhD student in clinical psychol-
ogy, and one licensed psychologist
Therapist, face-to-face contact: 1 x 3 hr exposure session
Dropout: n = 2; 13%
Primary outcomes:
(1) specific phobia symptoms: Behavioural Avoidance Test; Snake Phobia Questionnaire; Fear Sur-
vey Schedule-III
Notes
Informed decisions.
Andrews 2011b
Participants Diagnosis: DSM-IV Panic Disorder with or without Agoraphobia
Age: 18 years and older
Co-use of medication: permitted if stable dose for past month
Interventions Participants randomly assigned to either:
(1) Internet-based CBT with e-mail support
Treatment protocol: participants complete online CBT modules about the management of panic
symptoms with e-mail support from a therapist
Therapists: psychiatrists
Therapist contact: as and when required
(2) Waiting list control
Duration: 8 weeks
Primary outcome:
(1) panic symptoms: Panic Disorder Severity Scale
Secondary outcome:
(1) quality of life: World Health Organization Disability Assessment Scale
Notes ACTRN12611001120965
Andrews 2011c
Informed decisions.
Andrews 2011c (Continued)

Co-use of medication: permitted (except benzodiazepines) if stable dose for past month
(1) Internet-based CBT with e-mail and telephone support
Treatment protocol: participants complete online CBT modules about the management of gener-
alized anxiety symptoms with e-mail or telephone support, or both, from a therapist
Therapists: clinical psychologist
Duration: 10 weeks
Primary outcome:
(1) generalized anxiety symptoms: GAD-7, Penn State Worry Questionnaire
Secondary outcomes:
(1) quality of life: WHO Disability Assessment Scale
(2) treatment satisfaction (at post-treatment): patient satisfaction questionnaire
Andrews 2011d
Participants Diagnosis: DSM-IV Social Phobia, Panic Disorder (with or without agoraphobia), or Generalized
Anxiety Disorder, or both
Psychiatric co-morbidity: included, except Substance Abuse and Dependence, Psychotic Disorder
Informed decisions.
Andrews 2011d (Continued)

Treatment protocol: participants complete online CBT modules about the management of their
anxiety disorder with weekly telephone calls and access to a moderated online discussion forum
Therapists: unknown
Therapist contact: 2 to 7 min/week
Duration: 10 weeks
Primary outcomes:
(1) clinically significant improvement: MINI
(2) general anxiety symptoms: Beck Anxiety Inventory
(3) panic symptoms: Body Vigilance Scale, Agoraphobic Questionnaire
(4) generalized anxiety disorder symptoms: GAD-7
(5) social phobia symptoms: Social Interaction Anxiety Scale
Secondary outcome:
(1) quality of life: Sheehan Disability Scale, World Health Organisation Disability Assessment Sched-
ule II
Andrews 2012a
Participants Diagnosis: DSM-IV Obsessive Compulsive Disorder
Informed decisions.
Andrews 2012a (Continued)

Treatment protocol: participants complete online CBT modules about the management of obses-
sive compulsive disorder with e-mail or telephone support, or both, from a therapist
Duration: 10 weeks
Primary outcomes:
(1) obsessive compulsive disorder symptoms: Obsessive Beliefs Questionnaire, Dimensional Obses-
sive Compulsive Scale
Secondary outcome:
Berger 2012
Methods Randomized controlled trial
Method of Diagnosis: YBOCS
Age: 18-65 years
Country of Residence: Germany
Psychiatric Comorbidity: included (except severe Major Depression)
Co-use of Adjunct Therapy: excluded
Co-use of Medication: permitted (except benzodiazepines) if stable dose for past month
(1) Internet-based CBT with email and telephone support
Duration: 14 online modules
Treatment Protocol: participants complete online modules on exposure and response prevention
including psychoeduction, processing of functionalities, and strategies for relapse prevention with
therapist support via email
Therapists: unknown
Therapist Contact: unknown
Face-to-Face Contact: none

Informed decisions.

Duration: length of treatment (exact duration unknown)
Therapist/Face-to-Face Contact: none
Outcomes Timepoints for Assessment: pre- and post-treatment and 2 and 6 month follow-up
Primary outcome:
(1) obsessive compulsive disorder symptoms: YBOCS, Obsessive Compulsive Inventory Revised
Notes DRKS00004612
Berger 2014
Participants Diagnosis: DSM-IV Social Phobia (n = 113), Panic Disorder with or without Agoraphobia (n = 44),
and Generalized Anxiety Disorder (n = 33)
N: 132
Sex: 56.1% women
Country of residence: Switzerland, Germany, Austria
Psychiatric co-morbidity: 37.1% Anxiety Disorder, 13.6% Mood Disorder, 15.9% Specific Phobia,
5.3% OCD, 12.1% other Axis I disorder
Co-use of adjunct therapy: no
Co-use of medication: yes if stable dose for one month pre-treatment
Method of enrollment: responded to online and media study advertisements
Treatment protocol*: participants completed online modules on motivational enhancement, psy-

choeducation, cognitive restructuring, mindfulness, exposure, lifestyle modification, and relapse
prevention, with email support from a therapist for module exercises
Therapists: 5 Master of Science students in their last term of clinical psychology and psychother-
apy, a psychologist with a master's degree in clinical psychology and in their fourth year of a 5 yr
post-graduate CBT psychotherapy program, and a CBT therapist
Therapist contact: M e-mails written by clients = 6.53 (SD = 7.2), M e-mails written by therapists =
12.6 (SD = 4.6)
Dropout: n = 5; 11%
Informed decisions.

(2) Internet-based CBT with e-mail support tailored to diagnoses (n = 44)
Treatment protocol*: participants completed online modules on motivational enhancement, psy-

choeducation, cognitive restructuring, mindfulness, exposure, lifestyle modification, and relapse
prevention, with email support from a therapist for module exercises; prescribed modules depend-
ed on participants' primary and comorbid diagnoses
Therapists: 5 Master of Science students in their last term of clinical psychology and psychother-
apy, a psychologist with a master's degree in clinical psychology and in their fourth year of a 5 yr
post-graduate CBT psychotherapy program, and a CBT therapist
Therapist contact: M e-mails written by clients = 6.53 (SD = 7.2), M e-mails written by therapists =
12.6 (SD = 4.6)
Dropout: n = 4; 10%
Duration: 8 weeks
Dropout: n = 4; 10%
Primary outcomes:
(1) social phobia symptoms: Social Phobia Scale, Social Interaction Anxiety Scale
(2) generalized anxiety symptoms: Penn State Worry Questionnaire
(3) panic and agoraphobia symptoms: Agoraphobic Cognitions Questionnaire, Bodily Sensations
Questionnaire, Mobility Inventory for Agoraphobia
(4) general anxiety: Beck Anxiety Inventory, Brief Symptom Inventory
(5) clinically significant improvement: SCID-IV
Notes *treatment based on: Clark, D.M., & Wells, A. (1995). A cognitive model of social phobia. New York:
Guilford Press.; Stangier, U., Heidenreich, T., & Peitz, M. (2003). Soziale Phobien. Ein kognitiv-ver-
haltenstherapeutisches Behandlungsmanual. Weinheim: Beltz.; Margraft, J., & Schneider, S. (1989).
Panik: Angstanfalle und ihre Behandlung. Berlin: Springer.; Schneider, S., & Margraf, J. (1998).
Fortschritte der psychotherapie: Agoraphobie und panikstorung. Gottingen: Hogrefe.; Becker, E., &
Margraf, J. (2002). Generalisierte Angststorung. Ein Therapieprogramm. Weinheim: Beltz Verlag.
Carlbring 2012
Method of diagnosis: unknown
Informed decisions.

Co-use of medication: permitted if stable
Duration: 9 weeks
Treatment protocol: participants complete online CBT modules with email support from a thera-
pist for module exercises
Therapists: unknown
Therapist contact: unknown
Duration: 9 weeks
Outcomes Timepoints for assessment: pre- and post-treatment and 6 month, 1 year, and 2 year follow-ups
Primary outcomes:
(1) generalized anxiety symptoms: Penn State Worry Questionnaire; GAD-7
Secondary outcome:
Notes NCT01570374
Greist 2012
Method of diagnosis: YBOCS
Age: 18 years or older
Country of residence: USA
Psychiatric co-morbidity: included (except significant co-morbid Depression)
Co-use of adjunct therapy or medication: unknown
Interventions Participants randomly assigned to one of:
(1) Internet-based CBT with telephone support from a therapist
Duration: 12 weeks
Informed decisions.
Greist 2012 (Continued)

Treatment protocol: participants complete online CBT for OCD modules with weekly telephone
coaching from a therapist
Therapists: 'CBT therapist'
Therapist contact: weekly coaching via telephone
(2) Internet-based CBT with telephone support from a non-therapist
Duration: 12 weeks
Treatment protocol: participants complete online CBT for OCD modules with weekly telephone
coaching from a non-therapist
Therapists: unknown (non-therapist)
Therapist contact: weekly coaching via telephone
(3) Internet-based unguided CBT
Duration: 12 weeks
Treatment protocol: participants complete online CBT for OCD self-help modules
Primary outcome:
(1) obsessive-compulsive symptoms: YBOCS
Notes NCT01522287
Ivarsson 2014
Participants Diagnosis: DSM-IV PTSD
Method of diagnosis: CAPS
N: 62
Age: M = 46 (SD = 11.7); range = 21-67 years
Sex: 82.3% women
Co-use of adjunct therapy: no
Co-use of medication: yes if stable dose for 3 months pre-treatment
Method of Eenrollment: responded to media study advertisements

Informed decisions.
Ivarsson 2014 (Continued)

(1) Internet-based CBT with email support (n=44)
Treatment protocol*: participants completed online modules on psychoeducation, anxiety cop-

ing skill training, exposure, and cognitive restructuring, with email support from a therapist for
module exercises
Therapists: students in a 5 year clinical psychology program
Therapist contact: M 28 mins/week; range = 11-52 mins/week
Dropout: n=3; 10%
(2) Waiting List Control (n=44)
Duration: 8 weeks
Therapist/face-to-face contact: none
Dropout: n=5; 16%
Primary outcomes:
(1) posttraumatic stress symptoms: Impact of Events Scale Revised, Posttraumatic stress Diagnos-
tic Scale
(3) clinically significant improvement: CAPS, Clinical Global Impression - Improvement Scale
Secondary outcome:
Notes *treatment based on Bisson, J. I., Ehlers, A., Matthews, R., Pilling, S., Richards, D., & Turner, S.
(2007). Psychological treatments for chronic post-traumatic stress disorder: Systematic review and
meta-analysis. British Journal of Psychiatry, 190, 97-104.; Harvey, A. G., Bryant, R. A., & Tarrier, N.
(2003). Cognitive behaviour therapy for posttraumatic stress disorder. Clinical Psychology Review,
23, 501-522.
Newby 2013
Participants Diagnosis: DSM-IV GAD and/or MDD (only participants with GAD included in this review)
N: 109
Age: M = 44.3 (SD = 12.2); range = 21-80 years
Sex: 77.8% women
Psychiatric comorbidity: MDD
Informed decisions.
Newby 2013 (Continued)

Co-use of adjunct therapy: yes if stable for 2 months pre-treatment
Co-use of medication: yes if stable dose for 2 months pre-treatment
(1) Internet-based CBT with email and phone support (n=49)
Treatment protocol*: participants completed online modules on psychoeducation, cognitive re-

structuring, mindfulness, exposure, and relapse prevention, with email and phone support from a
therapist for module exercises
Therapists: practice manager, supervised by a clinical psychologist
Dropout: n=3; 6%
(2) Waiting List Control (n=60)
Duration: 10 weeks
Therapist/face-to-face contact: none
Dropout: n=6; 10%
Primary outcome:
(1) generalized anxiety symptoms: GAD-7, Penn State Worry Questionnaire
Secondary outcome:
(1) quality of life: WHO Disability Assessment Schedule - II
Notes *treatment based on Worry and Sadness Program (www.virtualclinic.org.au)
Nordgren 2012
Participants Diagnosis: DSM-IV Panic Disorder with or without Agoraphobia (n = 31), Agoraphobia (n = 8), Social
Phobia (n = 32), Generalized Anxiety Disorder (n = 10), Anxiety Disorder Not Otherwise Specified (n =
19)
N: 100
Age: for ICBT M = 35 (SD = 13); for WLC M = 36 (SD = 12); range = 19 to 68 years
Sex: 63% women
Informed decisions.
Nordgren 2012 (Continued)

Psychiatric co-morbidity: Anxiety Disorders (n = 31), Mood Disorders (n = 43), Hypochondriasis (n
= 1)
Method of enrollment: participants were recruited from a clinical population through referral
from their general practitioner or a nurse, when seeking help at their primary care centre
Baseline depression severity: (MADRS-S) for ICBT, M 19.6 (SD = 1.0); for WLC, M = 17.8 (SD = 1.0)
(1) Internet-based CBT with e-mail support from a therapist
Treatment protocol: participants complete online modules on cognitive restructuring, disor-

der-specific anxiety symptom management, relaxation, behavioural activation, mindfulness, as-
sertiveness, problem solving, stress management, sleep, and relapse prevention with e-mail sup-
port from a therapist for module exercises
Therapists: seven Master's of Science students
Therapist contact: 15 min/week
Dropout: n = 4; 8%
Duration: 10 weeks
Dropout: n = 5; 10%
Primary outcome:
(1) general anxiety symptoms: Clinical Outcomes in Routine Evaluation - Outcome Measure; Beck
Anxiety Inventory
Secondary outcome:
Notes NCT01390168
Richards 2014
Method of Diagnosis: GAD-7
Informed decisions.

Country of residence: Ireland
Co-use of medication: excluded
Duration: 6 online modules completed weekly
Treatment protocol: participants complete online CBT modules about the management of GAD
with e-mail support from a therapist
Therapists: clinical psychology graduate students at the master's level
Therapist contact: 10-15 min/week
Duration: 6 weeks
Primary outcome:
(1) general anxiety symptoms: GAD-7, Penn State Worry Questionnaire
Secondary outcome:
(1) quality of life: EuroQol 5D, Work and Social Adjustment questionnaire
(2) treatment satisfaction: Helpful Aspects of Therapy Form, Satisfaction with Treatment question-
naire
Notes ISRCTN16303842
Schreuders 2008
Participants Diagnosis: DSM-IV Social Phobia, Specific Phobia, or Agoraphobia
Age: unknown
Country of residence: Netherlands
Psychiatric co-morbidity: unknown
Co-use of adjunct therapy: unknown
Co-use of medication: unknown
Informed decisions.
Schreuders 2008 (Continued)
Treatment protocol: participants complete online CBT modules with a focus on exposure with e-
mail support from a therapist
Therapists: unknown
(2) Unclear
Outcomes Timepoints for assessment: pre- and and post-treatment and 3 month follow-up
Primary outcome:
(1) anxiety symptoms: measurement method unknown
Secondary outcome:
(1) treatment satisfaction
Notes NTR 1260
Characteristics of ongoing studies [ordered by study ID]
Andrews 2012b
Trial name or title The Obsessive Compulsive Disorder (OCD) Program - A randomised controlled trial of online versus
face-to-face cognitive behavioural therapy (CBT)
Treatment protocol: participants complete online CBT modules about the management of obses-
sive compulsive disorder with e-mail or telephone support, or both, from a therapist
Informed decisions.
Andrews 2012b (Continued)

Therapist contact: weekly for first two weeks and then when required
(2) Face-to-face individual CBT
Duration: 12 face-to-face individual therapy sessions over 12 weeks
Treatment protocol: individual sessions focused on helping individuals gradually confront feared
situations
Therapists: clinical psychologists
Therapist, face-to-face contact: 12 x 1 hr sessions
Outcomes Timepoints for assessment: pre-, mid-, and post-treatment and 3 month follow-up
Primary outcomes:
(1) obsessive compulsive disorder symptoms: Obsessive Beliefs Questionnaire, Dimensional Obses-
sive Compulsive Scale
Secondary outcome:
Starting date December, 2012
Contact information Professor Gavin Andrews, St Vincent's Hospital, Sydney, 390 Victoria St, Darlinghurst NSW 2010,
Australia; Tel: +612 8382 1400; Fax: +612 8382 1401; gavina@unsw.edu.au
Notes
Bishop 2012
Trial name or title Development of a web-based cognitive behavioral treatment for OEF/OIF veterans with PTSD
symptoms and substance misuse
Duration: 24 brief online intervention modules
Informed decisions.
Bishop 2012 (Continued)

Treatment protocol: participants complete online modules on CBT for PTSD with e-mail support
from a therapist
Therapists: unknown
Duration: unknown
Primary outcome:
(1) post-traumatic stress symptoms: unknown how these symptoms were measured
Starting date Spring 2012
Contact information Kyle Possemato, Ph.D.; Syracuse VA Medical Center, Syracuse, NY 13210; kyle.possemato@va.gov
Notes
Clark 2012
Trial name or title A randomised controlled trial of internet-based cognitive therapy (iCT) and standard cognitive
therapy (CT) for social anxiety disorder
Age: 18 to 65 years
Country of residence: United Kingdom
Co-use of medication: permitted if stable for past two months
(1) Internet-based CT with e-mail and telephone support
Duration: 14 weeks
Treatment protocol: participants complete online CT modules, including video demonstrations

of procedures and virtual audiences to practice real-life tasks, with e-mail and telephone support
from a therapist
Therapists: unknown
Therapist contact: 10 to 15 min phone conversations weekly in addition to e-mail contact
Informed decisions.
Clark 2012 (Continued)

(2) Face-to-face CT
Duration: 14 weeks
Treatment protocol: participants complete 14 weekly individual CT sessions with a therapist
Therapists: unknown
Therapist contact: 14 x 90 min individual sessions
Face-to-face contact: 14 x 90 min individual sessions
Duration: 14 weeks
Outcomes Timepoints for assessment: pre- and post-treatment and 3 month and 1 year follow-ups
Primary outcome:
(1) social phobia symptoms: Anxiety Disorders Interview Schedule (Fear and Avoidance Scale) for
DSM-IV; Social Phobia Weekly Summary Scale; Liebowitz Social Anxiety Scale; Fear of Negative
Evaluation Scale; Social Phobia Scale; Social Interaction and Anxiety Scale
Secondary outcome:
Starting date January, 2013
Contact information Professor David M Clark; Oxford Centre for Anxiety Disorders and Trauma, Department of Experi-
mental Psychology, Tinbergen Building, 9 South Parks Road; david.clark@psy.ox.ac.uk
Notes
Kok 2012
Trial name or title Effectiveness and cost-effectiveness of web-based treatment for phobic outpatients on a waiting
list for psychotherapy: Protocol of a randomised controlled trial
Participants Diagnosis: DSM-IV Social Phobia, Agoraphobia, Specific Phobia
Country of residence: the Netherlands
Psychiatric co-morbidity: included, expect Bipolar Disorder or Psychosis
Co-use of adjunct therapy: unknown
Co-use of medication: included if stable during treatment
Informed decisions.
Kok 2012 (Continued)

(1) Internet-based CBT with e-mail support from a therapist
Duration: 5 weeks
Treatment protocol: participants complete online modules on psychoeducation, exposure, and

relapse prevention with e-mail support from a therapist for module exercises
Therapists: master's level clinical psychology students
Therapist contact: amount unknown; coaching through a secure online message system
(2) Self-help bibliotherapy
Duration: 5 weeks
Treatment protocol: participants receive a self-help book for phobias free in the mail, however,
they are given no instructions or expectations
Outcomes Timepoints for assessment: pre- and post-treatment and 3, 6, 9, and 12 month follow-up
Primary outcome:
(1) phobia symptoms: Fear Questionnaire
Secondary outcome:
(1) quality of life: EuroQol 5-D
(2) treatment satisfaction: Client Satisfaction Questionnaire
Starting date October, 2010
Contact information Robin N Kok, Department of Clinical Psychology and the EMGO institute for Health and Care Re-
search, Faculty of Psychology and Education, VU University Amsterdam, Van der Boechorststraat 1,
1081, BT Amsterdam, The Netherlands; r.n.kok@vu.nl
Notes
Lindner 2013
Trial name or title ACT-smart: Smartphone-supplemented iCBT for social phobia and/or panic disorder
Psychiatric co-morbidity: excluded if another condition requiring specialized treatment
Informed decisions.
Lindner 2013 (Continued)

Co-use of medication: included if stable for past 3 months
(1) Internet-based CBT plus smartphone with e-mail support
Duration: 10 online modules over 10 weeks
Treatment protocol: participants complete online modules on psychoeducation, relaxation, cog-

nitive restructuring, and exposure, with e-mail and Skype support from a therapist for module ex-
ercises
Therapists: unknown
Therapist contact: 15 min/week; feedback and support provided in response to participants'

homework completion
(2) Unguided Internet-based CBT plus smartphone

ercises
Therapists: unknown
Duration: 10 weeks
Outcomes Timepoints for assessment: pre- and post-treatment and 1, 3, 6, and 12 month follow-ups
Primary outcome:
(1) panic disorder symptoms: Panic Disorder Severity Rating Scale
(2) social phobia symptoms: Liebowitz Social Anxiety Scale
(3) general anxiety: GAD-7
Secondary outcome:
Starting date 2013, October
Contact information Per Carlbring, Professor, Stockholm University
Notes
Miclea 2014
Trial name or title PAXonline: A randomized controlled trial assessing the efficacy of an Internet-based cognitive be-
havior intervention for panic disorder
Informed decisions.
Miclea 2014 (Continued)
Age: 18 to 65 years
Country of residence: Romania
Psychiatric co-morbidity: included, except severe Depression, Substance Abuse, Personality Dis-
orders, psychotic disorders, mental retardation
Co-use of medication: excluded if using benzodiazepines

ercises
Therapists: unknown
Therapist contact: feedback and support provided in response to participants' homework com-
pletion
(2) Unguided Internet-based CBT

nitive restructuring, and exposure
Therapists: unknown
Duration: 12 weeks
Outcomes Timepoints for assessment: pre- and post-treatment and 1, 3, 6, and 12 month follow-ups
Primary outcome:
(1) panic disorder symptoms: Panic Disorder Severity Rating Scale, Agoraphobic Cognitions Ques-
tionnaire, Body Sensations Questionnaire
Starting date May, 2014
Contact information Mircea Miclea, Babes-Bolyai University, School of Psychology and Educational Sciences, Depart-
ment of Psychology 37, Republicii Street, Cluj - Napoca, Cluj, Romania, 400015; Tel: +40 753 529
753; liviugcrisan.neuro@gmail.com
Informed decisions.
Miclea 2014 (Continued)
Notes
Rollman 2012
Trial name or title Online treatments for mood and anxiety disorders in primary care
Participants 18 to 75 years of age
Current major depression, panic and/or generalized anxiety disorder on PRIME-MD
Diagnosis: DSM-IV Panic Disorder, Generalized Anxiety Disorder, or MDD (not included in review)
Method of diagnosis: PRIME-MD
Age: 18 to 75 years
Psychiatric co-morbidity: included, except Substance Abuse, Psychosis, Bipolar Disorder
Co-use of medication: unknown
Duration: unknown
Treatment protocol: participants complete online treatment modules for anxiety with weekly e-
Therapist, face-to-face contact: unknown
(2) Internet-based CBT with e-mail support and online support group
Duration: unknown
Treatment protocol: participants complete online treatment modules for anxiety with weekly e-
Therapist, face-to-face contact: unknown
(3) Usual care
Duration: unknown
Therapist, face-to-face contact: variable by type of intervention provided as part of usual care
Outcomes Timepoints for assessment: pre-treatment and 6 or 12 month follow-ups
Primary outcome:
(1) general anxiety symptoms: Hamilton Rating Scale for Anxiety
Secondary outcome:
Informed decisions.
Rollman 2012 (Continued)

(1) quality of life: WHO Health and Work Performance Questionnaire, SF-12
Starting date 2011, November
Contact information Bruce Rollman, University of Pittsburgh
Notes NCT01482806
Titov 2012
Trial name or title A randomized controlled trial of the effects of disorder-specific vs. trans-diagnostic and self-guided
vs. guided Internet-administered treatment on symptoms of social phobia in Australian adults
Age: 18 to 64 years
Psychiatric co-morbidity: included, except Psychosis
Co-use of medication: included if stable dose for past month
(1) Unguided disorder-specific Internet-based CBT
Duration: 5 online modules over 8 week
Treatment protocol: participants complete online treatment modules for social phobia
(2) Unguided Trans-diagnostic Internet-based CBT
Treatment protocol: participants complete online treatment modules for anxiety and depression
(3) Disorder-specific Internet-based CBT with e-mail or phone support, or both
Treatment protocol: participants complete online treatment modules for social phobia with
weekly phone or e-mail support, or both, from a therapist
Therapist contact: weekly
(4) Trans-diagnostic Internet-based CBT with e-mail or phone support, or both
Informed decisions.

Treatment protocol: participants complete online treatment modules for anxiety and depression
with weekly phone or e-mail support, or both, from a therapist
Therapist contact: weekly
Outcomes Timepoints for assessment: pre- and post-treatment and 6, 12, and 24 month follow-ups
Primary outcome:
(1) social phobia symptoms: MINI - Social Phobia Inventory
Secondary outcome:
Starting date 2012, April
Contact information Nickolai Titov, Centre for Emotional Health, Department of Psychology, Building/Room C3A 724
Macquarie University, North Ryde, NSW 2109
Tulbure 2012
Trial name or title Internet treatment for social anxiety disorder in Romania
Country of residence: Romania
Psychiatric co-morbidity: included, except Borderline Personality Disorder or Psychosis
Co-use of medication: included if stable dose for past month
Treatment protocol: participants complete online modules on psychoeducation, cognitive re-

structuring, and exposure, with e-mail support from a therapist for module exercises
Therapist contact: feedback and support provided in response to participants' homework com-
pletion
Informed decisions.
Tulbure 2012 (Continued)

Duration: 9 weeks
Primary outcome:
(1) social phobia symptoms: Liebowitz Social Anxiety Scale; Social Phobia Inventory; Social Interac-
tion Anxiety Scale; Social Phobia Screening Questionnaire
Starting date April, 2012
Contact information Bogdan Tudor Tulbure, Babes-Bolyai University, Cluj - Napoca, Cluj, Romania, 400084; Tel: 0040 745
753061; bogdan.tulbure@ubbcluj.ro
Notes
von Essen 2008

Trial name or title Treatment of post-traumatic stress disorder among parents of children with cancer with cognitive
behavioural therapy over the Internet
Psychiatric co-morbidity: unknown
Duration: unknown
Treatment protocol: participants complete online modules of CBT for PTSD with e-mail support
from a therapist
Therapists: unknown
Duration: unknown
Primary outcome:
Informed decisions.
von Essen 2008 (Continued)

(1) post-traumatic stress symptoms: unknown how these symptoms will be measured
Starting date Unknown
Contact information Louise von Essen, Uppsala universitet
Notes
DATA AND ANALYSES
Comparison 1. Therapist-supported ICBT versus waiting list control
Outcome or subgroup title No. of No. of Statistical method Effect size

studies partici-
pants
1 Clinically Important Improvement in 9 644 Risk Ratio (M-H, Random, 95% CI) 4.18 [2.42, 7.22]
Anxiety at Post-Treatment
2 Anxiety Symptom Severity at Post- 24 1573 Std. Mean Difference (IV, Random, 95% CI) -1.12 [-1.39, -0.85]
Treatment
3 General Anxiety Symptom Severity at 14 1004 Std. Mean Difference (IV, Random, 95% CI) -0.79 [-1.10, -0.48]
Post-Treatment
4 Quality of Life at Post-Treatment 20 1395 Std. Mean Difference (IV, Random, 95% CI) 0.51 [0.40, 0.61]
Analysis 1.1. Comparison 1 Therapist-supported ICBT versus waiting list control,

Outcome 1 Clinically Important Improvement in Anxiety at Post-Treatment.
Study or subgroup ICBT Waiting List Risk Ratio Weight Risk Ratio
n/N n/N M-H, Random, 95% CI M-H, Random, 95% CI
Andersson 2012a 36/102 6/102 14.91% 6[2.64,13.62]
Andersson 2012b 7/20 4/25 11.97% 2.19[0.74,6.43]
Carlbring 2006 23/30 0/30 3.34% 47[2.99,740.03]
Carlbring 2011 16/27 2/27 9.31% 8[2.03,31.48]
Richards 2006 4/12 0/9 3.25% 6.92[0.42,114.19]
Robinson 2010 39/47 19/48 20.3% 2.1[1.44,3.04]
Spence 2011 14/23 4/19 13.59% 2.89[1.14,7.33]
Titov 2009 19/24 3/21 12.08% 5.54[1.91,16.12]
Titov 2010 16/40 3/38 11.25% 5.07[1.6,16.01]
Total (95% CI) 325 319 100% 4.18[2.42,7.22]

Total events: 174 (ICBT), 41 (Waiting List)
Heterogeneity: Tau2=0.35; Chi2=19.62, df=8(P=0.01); I2=59.23%
Test for overall effect: Z=5.13(P<0.0001)
Favours Waiting List 0.01 0.1 1 10 100 Favours ICBT
Informed decisions.
Analysis 1.2. Comparison 1 Therapist-supported ICBT versus waiting

list control, Outcome 2 Anxiety Symptom Severity at Post-Treatment.
Study or subgroup ICBT Waiting List Std. Mean Difference Weight Std. Mean Difference
N Mean(SD) N Mean(SD) Random, 95% CI Random, 95% CI
Andersson 2012a 102 29.9 (8.3) 102 42.4 (8.1) 4.92% -1.52[-1.83,-1.21]
Andersson 2012b 23 34.7 (4.9) 26 35.7 (5.5) 4.29% -0.2[-0.76,0.37]
Berger 2009 31 36.8 (8.9) 21 48.3 (7.1) 4.13% -1.39[-2,-0.77]
Carlbring 2001 21 33.9 (4.6) 20 45.4 (6) 3.66% -2.13[-2.91,-1.35]
Carlbring 2006 30 13.8 (1.7) 30 22 (2.1) 3.2% -4.32[-5.27,-3.38]
Carlbring 2007 29 22.6 (6) 28 37.1 (6.5) 3.96% -2.28[-2.96,-1.6]
Carlbring 2011 27 1 (0.6) 27 1.5 (0.4) 4.28% -0.97[-1.54,-0.4]
Furmark 2009a 40 34.4 (7.2) 40 45.5 (9.5) 4.51% -1.31[-1.8,-0.83]
Johnston 2011 46 7.5 (5.7) 42 11.8 (4.6) 4.64% -0.81[-1.25,-0.37]
Paxling 2011 44 32.1 (6.4) 45 39.6 (3.9) 4.56% -1.4[-1.86,-0.93]
Richards 2006 12 13.5 (5.5) 9 17.5 (3.6) 3.31% -0.8[-1.7,0.11]
Robinson 2010 47 51.5 (12.3) 48 64.2 (11.8) 4.65% -1.05[-1.48,-0.62]
Silfvernagel 2012 29 6.5 (5) 28 13.8 (5.5) 4.24% -1.37[-1.95,-0.79]
Spence 2011 23 44.8 (17.3) 19 51.8 (12.5) 4.14% -0.45[-1.06,0.17]
Titov 2008a 50 29.9 (9.1) 49 42.5 (11.5) 4.65% -1.2[-1.63,-0.77]
Titov 2008b 41 29 (11.5) 40 44.8 (10.2) 4.49% -1.44[-1.93,-0.94]
Titov 2008c 31 29.8 (8) 34 44.3 (13.6) 4.37% -1.27[-1.81,-0.73]
Titov 2009 24 56.8 (10.8) 21 66.1 (8.7) 4.13% -0.93[-1.55,-0.32]
Titov 2010 GAD 18 60.9 (9.4) 16 61.9 (11.2) 3.97% -0.1[-0.77,0.58]
Titov 2010 Panic 10 7.7 (4) 11 15 (6.7) 3.18% -1.25[-2.21,-0.3]
Titov 2010 Social Phobia 12 13.3 (10.7) 11 18.4 (11.9) 3.52% -0.44[-1.27,0.39]
Titov 2011 19 7.6 (5.3) 17 8.9 (4.1) 4.02% -0.26[-0.91,0.4]
van Ballegooijen 2013 63 5.8 (4.9) 63 7.3 (4.9) 4.84% -0.3[-0.66,0.05]
Wims 2010 29 41.6 (6.3) 25 45.1 (6.2) 4.34% -0.55[-1.1,-0.01]
Total *** 801 772 100% -1.12[-1.39,-0.85]

Heterogeneity: Tau2=0.36; Chi2=133.74, df=23(P<0.0001); I2=82.8%
Favours ICBT -5 -2.5 0 2.5 5 Favours Waiting List
Analysis 1.3. Comparison 1 Therapist-supported ICBT versus waiting list

control, Outcome 3 General Anxiety Symptom Severity at Post-Treatment.
Andersson 2012a 102 9.5 (6.4) 102 14 (8.4) 8.62% -0.61[-0.89,-0.33]
Andersson 2012b 23 38.5 (3.5) 26 43.5 (5.4) 6.83% -1.07[-1.67,-0.47]
Carlbring 2001 21 9.8 (8.4) 20 21.2 (10.4) 6.43% -1.19[-1.86,-0.52]
Carlbring 2006 30 8.5 (5.5) 30 19.6 (9.9) 7.05% -1.37[-1.93,-0.8]
Carlbring 2007 29 8.2 (7.9) 28 14.5 (9) 7.22% -0.73[-1.27,-0.2]
Carlbring 2011 27 13.9 (9.2) 27 17.2 (8) 7.22% -0.38[-0.91,0.16]
Furmark 2009a 40 10.2 (6.3) 40 15.3 (9.3) 7.74% -0.64[-1.09,-0.19]
Johnston 2011 46 34.9 (24) 42 48.5 (20.4) 7.86% -0.6[-1.03,-0.18]
Paxling 2011 44 35.5 (4.7) 45 45.8 (2.5) 6.93% -2.76[-3.34,-2.17]
Richards 2006 12 18.4 (11.9) 9 17.6 (8) 5.33% 0.07[-0.79,0.94]
Silfvernagel 2012 29 17.9 (8.5) 28 23 (9.4) 7.26% -0.57[-1.1,-0.04]
Informed decisions.
Spence 2011 23 7.9 (6) 19 10.6 (3.5) 6.73% -0.53[-1.15,0.09]
Titov 2011 19 37.1 (23.2) 17 43.1 (20.7) 6.5% -0.27[-0.92,0.39]
van Ballegooijen 2013 63 17 (12.7) 63 22 (12.7) 8.27% -0.39[-0.74,-0.04]
Total *** 508 496 100% -0.79[-1.1,-0.48]

Heterogeneity: Tau2=0.26; Chi2=66.04, df=13(P<0.0001); I2=80.32%
Analysis 1.4. Comparison 1 Therapist-supported ICBT versus

waiting list control, Outcome 4 Quality of Life at Post-Treatment.
Andersson 2012a 102 1.3 (2) 102 0.8 (1.7) 15.05% 0.28[0.01,0.56]
Andersson 2012b 23 1.6 (1.5) 26 1 (1.6) 3.56% 0.41[-0.16,0.98]
Carlbring 2001 21 2.2 (1.2) 20 1.3 (1.2) 2.84% 0.74[0.1,1.37]
Carlbring 2006 30 1.8 (1.6) 30 1.1 (1.6) 4.36% 0.43[-0.08,0.94]
Carlbring 2007 29 1.4 (1.8) 28 0.7 (1.8) 4.17% 0.38[-0.14,0.91]
Carlbring 2011 27 2 (1.7) 27 0.8 (1.6) 3.79% 0.68[0.13,1.23]
Furmark 2009a 40 1.3 (2) 40 0.4 (1.6) 5.8% 0.47[0.02,0.91]
Johnston 2011 46 18.2 (7.9) 42 14.1 (7.8) 6.33% 0.51[0.09,0.94]
Paxling 2011 44 1.3 (1.7) 45 0.4 (1.8) 6.42% 0.51[0.09,0.93]
Richards 2006 12 59.1 (9.1) 9 51.7 (9.8) 1.41% 0.75[-0.15,1.65]
Robinson 2010 47 20.6 (9.4) 48 14.3 (7.7) 6.62% 0.73[0.32,1.15]
Silfvernagel 2012 29 1.7 (1.5) 28 0.7 (1.7) 4.05% 0.6[0.07,1.13]
Spence 2011 23 16.8 (9.4) 19 11.9 (6.7) 2.97% 0.58[-0.04,1.2]
Titov 2008a 50 33.2 (12.9) 49 23.9 (15) 6.98% 0.66[0.26,1.07]
Titov 2008b 41 33.8 (12.9) 40 25.1 (13) 5.71% 0.66[0.21,1.11]
Titov 2008c 31 21.2 (6.6) 34 18.5 (7.5) 4.75% 0.37[-0.12,0.86]
Titov 2009 24 22.2 (7.9) 21 15 (10.3) 3.09% 0.77[0.17,1.38]
Titov 2010 40 19.5 (7.7) 38 13.6 (9.1) 5.46% 0.7[0.24,1.16]
Titov 2011 19 16 (10.3) 17 14.5 (8) 2.67% 0.15[-0.5,0.81]
Wims 2010 29 15.6 (8.4) 25 14.5 (5.4) 3.99% 0.16[-0.38,0.7]
Total *** 707 688 100% 0.51[0.4,0.61]

Heterogeneity: Tau2=0; Chi2=10.94, df=19(P=0.93); I2=0%
Favours Waiting List -4 -2 0 2 4 Favours ICBT
Comparison 2. Therapist-supported ICBT versus unguided CBT

studies partici-
pants
1 Clinically Important Improve- 1 Risk Ratio (M-H, Random, 95% CI) Totals not selected
ment in Anxiety at Post-Treatment
Informed decisions.

studies partici-
pants
2 Anxiety Symptom Severity at 4 253 Std. Mean Difference (IV, Random, 95% CI) -0.24 [-0.69, 0.21]
Post-Treatment
3 Anxiety Symptom Severity at Fol- 3 192 Std. Mean Difference (IV, Random, 95% CI) -0.30 [-0.58, -0.01]
low-up
4 General Anxiety Symptom Sever- 2 138 Mean Difference (IV, Random, 95% CI) 0.28 [-2.21, 2.78]
ity at Post-Treatment
5 General Anxiety Symptom Sever- 2 138 Mean Difference (IV, Random, 95% CI) 0.72 [-2.12, 3.57]
ity at Follow-up
6 Quality of Life at Post-Treatment 3 199 Std. Mean Difference (IV, Random, 95% CI) 0.07 [-0.37, 0.50]
7 Quality of Life at Follow-up 2 138 Std. Mean Difference (IV, Random, 95% CI) -0.19 [-0.53, 0.14]
Analysis 2.1. Comparison 2 Therapist-supported ICBT versus unguided CBT,

Study or subgroup ICBT Unguided CBT Risk Ratio Risk Ratio
Berger 2011 16/27 15/27 1.07[0.67,1.69]
Favours Unguided CBT 0.01 0.1 1 10 100 Favours ICBT
Analysis 2.2. Comparison 2 Therapist-supported ICBT versus unguided

CBT, Outcome 2 Anxiety Symptom Severity at Post-Treatment.
Study or subgroup Guided ICBT Unguided CBT Std. Mean Difference Weight Std. Mean Difference
Berger 2011 27 31.1 (8.9) 27 34.7 (7.8) 23.88% -0.43[-0.97,0.11]
Furmark 2009a 40 34.4 (7.2) 40 33.7 (9.1) 27.09% 0.09[-0.34,0.53]
Furmark 2009b 29 30.2 (7.9) 29 28.9 (9.1) 24.64% 0.15[-0.37,0.66]
Titov 2008c 31 29.8 (8) 30 38.2 (11.9) 24.38% -0.82[-1.34,-0.29]
Total *** 127 126 100% -0.24[-0.69,0.21]

Test for overall effect: Z=1.05(P=0.3)
Favours ICBT -5 -2.5 0 2.5 5 Favours Unguided CBT
Informed decisions.

unguided CBT, Outcome 3 Anxiety Symptom Severity at Follow-up.
Study or subgroup ICBT Unguided CBT Std. Mean Difference Weight Std. Mean Difference
Berger 2011 27 30.8 (9.2) 27 33.1 (7.9) 28.27% -0.26[-0.8,0.27]
Furmark 2009a 40 28.3 (7.5) 40 32 (8.3) 41.12% -0.46[-0.91,-0.02]
Furmark 2009b 29 28.4 (8.7) 29 29.4 (8.8) 30.6% -0.11[-0.63,0.4]
Total *** 96 96 100% -0.3[-0.58,-0.01]


CBT, Outcome 4 General Anxiety Symptom Severity at Post-Treatment.
Study or subgroup ICBT Unguided CBT Mean Difference Weight Mean Difference
Furmark 2009a 40 10.2 (6.8) 40 10.5 (7.5) 62.95% -0.32[-3.46,2.82]
Furmark 2009b 29 10.9 (7.9) 29 9.6 (8.1) 37.05% 1.31[-2.79,5.41]
Total *** 69 69 100% 0.28[-2.21,2.78]

Favours ICBT -10 -5 0 5 10 Favours Unguided CBT

CBT, Outcome 5 General Anxiety Symptom Severity at Follow-up.
Study or subgroup ICBT Unguided CBT Mean Difference Weight Mean Difference
Furmark 2009a 40 9.1 (9.6) 40 9.1 (8) 53.92% 0.02[-3.86,3.9]
Furmark 2009b 29 11.1 (9.8) 29 9.6 (6.1) 46.08% 1.55[-2.64,5.74]
Total *** 69 69 100% 0.72[-2.12,3.57]


unguided CBT, Outcome 6 Quality of Life at Post-Treatment.
Furmark 2009a 40 1.3 (2) 40 1.4 (1.7) 36.27% -0.08[-0.51,0.36]
Furmark 2009b 29 1.1 (1.9) 29 1.6 (1.8) 31.71% -0.23[-0.75,0.28]
Favours Unguided CBT -4 -2 0 2 4 Favours ICBT
Informed decisions.
Titov 2008c 31 21.2 (6.6) 30 17.1 (8.5) 32.02% 0.52[0.01,1.03]
Total *** 100 99 100% 0.07[-0.37,0.5]

Analysis 2.7. Comparison 2 Therapist-supported ICBT versus unguided CBT, Outcome 7 Quality of Life at Follow-up.
Furmark 2009a 40 1.6 (1.6) 40 1.7 (1.4) 58.43% -0.06[-0.5,0.38]
Furmark 2009b 29 1 (1.8) 29 1.7 (2) 41.57% -0.38[-0.9,0.14]
Total *** 69 69 100% -0.19[-0.53,0.14]

Comparison 3. Therapist-supported ICBT versus face-to-face CBT

studies partici-
pants
1 Clinically Important Improvement 4 365 Risk Ratio (M-H, Random, 95% CI) 1.09 [0.89, 1.34]
in Anxiety at Post-Treatment
2 Clinically Important Improvement 3 279 Risk Ratio (M-H, Random, 95% CI) 1.10 [0.94, 1.27]
in Anxiety at Follow-up
3 Anxiety Symptom Severity at Post- 6 424 Std. Mean Difference (IV, Random, 95% CI) 0.09 [-0.26, 0.43]
Treatment
4 Anxiety Symptom Severity at Fol- 5 341 Std. Mean Difference (IV, Random, 95% CI) -0.21 [-0.42, 0.00]
low-Up
5 General Anxiety Symptom Severity 5 317 Std. Mean Difference (IV, Random, 95% CI) 0.17 [-0.35, 0.69]
at Post-Treatment
6 General Anxiety Symptom Severity 4 237 Std. Mean Difference (IV, Random, 95% CI) -0.16 [-0.42, 0.09]
at Follow-up
7 Quality of Life at Post-Treatment 5 392 Std. Mean Difference (IV, Random, 95% CI) 0.26 [0.06, 0.45]
8 Quality of Life at Follow-up 4 316 Std. Mean Difference (IV, Random, 95% CI) 0.33 [0.11, 0.55]
Informed decisions.
Analysis 3.1. Comparison 3 Therapist-supported ICBT versus face-to-face CBT,

Study or subgroup ICBT Face-to- Risk Ratio Weight Risk Ratio
Face CBT
Bergstrom 2010 30/50 34/54 45.02% 0.95[0.7,1.29]
Carlbring 2005 20/25 16/24 35.37% 1.2[0.85,1.69]
Hedman 2011 18/64 12/62 10.18% 1.45[0.77,2.76]
Kiropoulos 2008 14/46 11/40 9.43% 1.11[0.57,2.15]
Total (95% CI) 185 180 100% 1.09[0.89,1.34]

Total events: 82 (ICBT), 73 (Face-to-Face CBT)
Favours ICBT 0.01 0.1 1 10 100 Favours Face-to-Face CBT
Analysis 3.2. Comparison 3 Therapist-supported ICBT versus face-to-face

CBT, Outcome 2 Clinically Important Improvement in Anxiety at Follow-up.
Study or subgroup ICBT Face-to- Risk Ratio Weight Risk Ratio
Face CBT
Bergstrom 2010 35/50 32/54 27.47% 1.18[0.89,1.57]
Carlbring 2005 23/25 21/24 62.05% 1.05[0.87,1.27]
Hedman 2011 25/64 21/62 10.48% 1.15[0.73,1.83]
Total (95% CI) 139 140 100% 1.1[0.94,1.27]

Total events: 83 (ICBT), 74 (Face-to-Face CBT)
Favours ICBT 0.01 0.1 1 10 100 Favours Face-to-Face CBT
Analysis 3.3. Comparison 3 Therapist-supported ICBT versus face-

to-face CBT, Outcome 3 Anxiety Symptom Severity at Post-Treatment.
Study or subgroup ICBT Face-to-Face CBT Std. Mean Difference Weight Std. Mean Difference
Andersson 2009 13 10.7 (6.8) 14 10.1 (5.6) 11.69% 0.09[-0.66,0.85]
Bergstrom 2010 50 12.2 (5.5) 54 10.9 (5.2) 19.89% 0.24[-0.14,0.63]
Carlbring 2005 25 14.7 (2) 24 14.6 (1.7) 15.61% 0.05[-0.51,0.61]
Hedman 2011 64 27.9 (6.7) 62 31.7 (7.9) 20.68% -0.51[-0.86,-0.15]
Kiropoulos 2008 44 11.5 (4.1) 37 9.4 (4.3) 18.42% 0.49[0.05,0.93]
Tillfors 2008 19 25.7 (6.3) 18 24 (6.7) 13.71% 0.26[-0.39,0.9]
Total *** 215 209 100% 0.09[-0.26,0.43]

Favours ICBT -4 -2 0 2 4 Favours Face-to-Face CBT
Informed decisions.
Analysis 3.4. Comparison 3 Therapist-supported ICBT versus face-

to-face CBT, Outcome 4 Anxiety Symptom Severity at Follow-Up.
Andersson 2009 13 10.8 (5.3) 12 11.3 (5.4) 7.4% -0.09[-0.88,0.69]
Bergstrom 2010 50 10.4 (5.2) 54 10.6 (5.4) 30.81% -0.04[-0.42,0.35]
Carlbring 2005 25 14.6 (2.1) 24 14.6 (2) 14.53% -0.04[-0.6,0.52]
Hedman 2011 64 23.5 (7.8) 62 27.2 (7.4) 36.31% -0.48[-0.83,-0.13]
Tillfors 2008 19 23 (5.8) 18 23.8 (7.5) 10.95% -0.11[-0.76,0.53]
Total *** 171 170 100% -0.21[-0.42,0]

Favours ICBT -5 -2.5 0 2.5 5 Favours Face-to-Face CBT
Analysis 3.5. Comparison 3 Therapist-supported ICBT versus face-to-

face CBT, Outcome 5 General Anxiety Symptom Severity at Post-Treatment.
Andersson 2009 13 9.7 (8.2) 12 7.2 (6.1) 16.3% 0.33[-0.46,1.12]
Carlbring 2005 25 10.9 (7.1) 24 12.3 (7.7) 20.05% -0.19[-0.75,0.38]
Hedman 2011 64 12.1 (8.6) 62 14.2 (11.3) 23.39% -0.21[-0.56,0.14]
Kiropoulos 2008 42 9.9 (1.9) 38 8 (1.9) 21.6% 1.01[0.54,1.48]
Tillfors 2008 19 5.8 (5.3) 18 6.2 (4.8) 18.66% -0.08[-0.72,0.57]
Total *** 163 154 100% 0.17[-0.35,0.69]

Heterogeneity: Tau2=0.27; Chi2=19.1, df=4(P=0); I2=79.05%
Analysis 3.6. Comparison 3 Therapist-supported ICBT versus face-to-

face CBT, Outcome 6 General Anxiety Symptom Severity at Follow-up.
Andersson 2009 13 6.8 (3.8) 12 8.5 (6.6) 10.43% -0.31[-1.1,0.48]
Carlbring 2005 25 10.7 (7.9) 24 12.3 (10.1) 20.68% -0.17[-0.74,0.39]
Hedman 2011 64 10.6 (10) 62 11.8 (9.2) 53.29% -0.12[-0.47,0.23]
Tillfors 2008 19 6.1 (3.7) 18 6.8 (3.6) 15.6% -0.19[-0.83,0.46]
Total *** 121 116 100% -0.16[-0.42,0.09]

Informed decisions.

face-to-face CBT, Outcome 7 Quality of Life at Post-Treatment.
Bergstrom 2010 50 7.9 (1.4) 54 7.2 (1.8) 26.09% 0.47[0.08,0.86]
Carlbring 2005 25 2 (1.4) 24 1.7 (1.5) 12.59% 0.2[-0.36,0.77]
Hedman 2011 64 1.6 (1.6) 62 1.1 (1.7) 32.19% 0.3[-0.05,0.65]
Kiropoulos 2008 40 66.6 (7.1) 36 66 (7.5) 19.58% 0.08[-0.37,0.53]
Tillfors 2008 19 2 (1.2) 18 2.1 (1.9) 9.55% -0.06[-0.71,0.58]
Total *** 198 194 100% 0.26[0.06,0.45]

Favours Face-to-Face CBT -4 -2 0 2 4 Favours ICBT

face-to-face CBT, Outcome 8 Quality of Life at Follow-up.
Bergstrom 2010 50 8.3 (1.4) 54 7.8 (1.7) 32.91% 0.35[-0.04,0.74]
Carlbring 2005 25 1.9 (1.4) 24 1.7 (1.3) 15.73% 0.15[-0.42,0.71]
Hedman 2011 64 1.8 (1.5) 62 1.1 (1.5) 39.46% 0.46[0.11,0.82]
Tillfors 2008 19 2.1 (1.3) 18 2 (1.8) 11.9% 0.06[-0.58,0.71]
Total *** 158 158 100% 0.33[0.11,0.55]

Test for overall effect: Z=2.9(P=0)
Favours Face-to-Face CBT -4 -2 0 2 4 Favours ICBT
ADDITIONAL TABLES
Table 1. Summary of included studies table
Study Diagnosis and Co-mor- Participant Charac- Co-Use N Intervention Comparison Assess- Outcomes
bidity teristics of Med- ment
Library
Cochrane
ication Type & Therapist Points
(M age, age range,
sex, country of resi- Duration Contact
dence)
Anders- Specific Phobia, Spider M age=25.6 (4.1) Not re- 27 IBT with email: M to- Orientation post- specific phobia sx;
son et al Type ported 4 wks; 5 online tal time and 1 3-hour treat- general anxiety sx
Better health.
Informed decisions.
Trusted evidence.
(2009) 18-65 years modules spent live exposure ment
co-morbidity not reported per par- session
84.8% women 12-
ticipant
= 25 min month
Sweden
fol-
low-up
Anders- Social Phobia ICBT M age=38.1 13.7% 204 ICBT with M time Online Discus- post- diagnostic status;
son et al (11.3) using email: 9 wks; 9 spent sion Group treat- social phobia sx;
(2012a) co-morbidity included but medica- online modules per par- ment QOL; general anxi-
not reported WLC M age=38.4 tion ticipant ety sx;
(10.9) per week
= 15 min
19-71 years
61% women
Sweden
Anders- GAD ICBT M age=44.4 32.1% 81 ICBT with M to- (1) Waiting post- diagnostic status,
son et al (12.8) using email: 8 wks; 8 tal time List Control treat- GAD sx; general
(2012b) 22.2% Social Phobia, medica- online modules spent ment anxiety sx; QOL
19.8% PD, 3.7% OCD, IPDTM age=36.4 (9.7) tion per par- (2) IPDT: 8
23.5% MDD ticipant wks; 8 online
WLC M age=39.6 modules
= 92 min

(13.7)
(SD=61)
19-66 years
76.5% women
Sweden
Berg- Social Phobia M age=28.9 (5.3) Exclud- 52 ICBT with M=5.5 Waiting List post- social phobia sx;
er et al ed email: 10 wks; 5 emails Control treat- treatment satisfac-
(2009) 26.9% co-morbid Axis I 19-43 years online modules from ment tion
disorder partici-
44.2% women
pant
144
Table 1. Summary of included studies table (Continued)
Switzerland, France, weekly
Belgium emails
Library
Cochrane
from
therapist
Berg- Social Phobia M age=37.2 (11.2) 7.4% us- 81 ICBT with M=6.16 (1) Unguided post- diagnostic status;
er et al ing med- email: 10 wks; 5 (SD=4.56; ICBT treat- social phobia sx;
(2011) 38% co-morbid Axis I dis- 19-62 years ication online modules range=1-17) ment treatment satisfac-
Better health.
Informed decisions.
Trusted evidence.
order; 12% PD, 10% Spe- emails 10 weeks; 5 tion
cific Phobia, 2% GAD, 22% 53.1% women online mod- 6-month
from
MDD/Dysthymia, 2% ED partici- ules fol-
Switzerland
pant low-up
(2) Step-up on
M=12.44 demand ICBT
(SD=2.85;
range=6-17)
emails
from
therapist
Bergstrom 15.4% PD ICBT M age=33.8 (9.7) 45% us- 104 ICBT with M=11.3 10 weekly 2- post- diagnostic status;
et al GCBT M age=34.6 ing med- email: 10 wks; (SD=4.3) hour sessions treat- PD sx; QOL
(2010) 84.6% PD with Agorapho- (9.2) ication; 10 online mod- emails of GCBT ment
bia 34% ules from
18 years or older SSRI/SN- therapist 6 month
co-morbidity not reported fol-
RIs, 13%
61.5% women M to- low-up
BZ, 24%
BZ de- tal time
Sweden
rivatives spent
or neu- per par-
rolep- ticipant
tics; 5% = 35.4
TCAs min

(SD=19)
Carl- PD M age=34 (7.5) 64% us- 41 ICBT with M reci- Waiting List post- diagnostic status;
bring et ing med- email: 7-12 wks; procal Control treat- PD sx; QOL; gener-
al (2001) co-morbidity included but 21-51 years ication; 6 online mod- emails ment al anxiety sx; treat-
not reported 44% ules = 7.5 ment satisfaction
71% women
SSRIs, (SD=1.2;
Sweden 10% range=6-15)
BZ, 5%
TCAs, M to-
5% be- tal time
ta-block- spent
145
ers per par-

ticipant
= 90 min
Library
Cochrane
Carl- 49% PD M age=35.0 (7.7) 30.6% 49 ICBT with M reci- 10 weekly post- diagnostic status;
bring et SSRIs, email: 10 wks; procal 45-60 min ses- treat- PD and agorapho-
al (2005) 51% PD with Agoraphobia 18-60 years old 8.2% 10 online mod- emails sions of indi- ment bia sx; general anx-
BZ, 6.1% ules =15.4 vidual CBT iety sx; QOL
49% Anxiety Disorder; 6% 71% women 12-
TCAs, (SD=5.5;
MDD month
Sweden 6.1% be- range=4-31)
Better health.
Informed decisions.
Trusted evidence.
ta block- fol-
ers M to- low-up
tal time
spent
per par-
ticipant
=150 min
Carl- PD M age=36.7 (10) 54% us- 60 ICBT with email M reci- Waiting List post- diagnostic status;
bring et ing med- & phone: 10 procal Control treat- PD and agorapho-
al (2006) co-morbidity included but 18-60 years ication wks; 10 online contacts ment bia sx; general anx-
not reported modules = 13.5 iety sx; QOL; treat-
60% women
(SD =4.4; ment satisfaction
Sweden range=7-29)
M time
spent
per par-
ticipant
per week
= 12 min
M length
phone
call =

11.8 min
(range=
9.6-15.6)
Carl- Social Phobia ICBT M age=32.4 (9.1) Included 60 ICBT with email M time Waiting List post- social phobia sx;
bring et but not & phone: 9 wks; spent Control treat- general anxiety sx;
al (2007) co-morbidity included but WLC M age=32.9 (9.2) reported 9 online mod- per par- ment QOL
not reported ules ticipant
18-60 years
per week
64.9% women = 22 min
146
Sweden
M length
phone
Library
Cochrane
call =
10.5 min
(SD= 3.6)
Carl- 9% PD M age=38.8 (10.7) 26% us- 54 ICBT with M time Attention post- diagnostic status;
bring et ing an email: 10 wks; spent Control treat- anxiety sx (broad-
al (2011) 22% PD with Agoraphobia 22-63 years antide- 6-10 online per par- ment ly); QOL; general
Better health.
Informed decisions.
Trusted evidence.
pressant modules ticipant 10 wks of anxiety sx
39% Social Phobia 76% women posts in an
or anxi- per week
olytic = 15 min online sup-
20% GAD Sweden
port forum
13% ADNOS
2% OCD, 2% PTSD, 20%

MDD, 7% mild depression;
15% Dysthymia
Furmark Social Phobia ICBT M age=35 (10.2) 13.9% 120 ICBT with M time (1) Bib- post- social phobia sx;
et al using email: 9 wks; 9 spent lio-therapy: 9 treat- general anxiety sx;
(2009a) co-morbidity not reported WLC M age=35.7 medica- online modules per par- wks; 9 lessons ment QOL
(10.9) tion ticipant
per week (2) Waiting
Bib M age=37.7 (10.3) List Control
= 15 min
18 years or older
67.5% women
Sweden
Furmark Social Phobia ICBT M age=34.9 (8.4) 6.7% us- 115 ICBT with M time (1) Bib- post- social phobia sx;
et al ing med- email: 9 wks; 9 spent lio-therapy: 9 treat- general anxiety sx;
co-morbidity not reported Bib M age=32.5 (8.5)

(2009b) ication online modules per par- wks; 9 lessons ment QOL
ticipant
Applied Relaxation M (2) Bib-
per week
age=36.4 (9.8) lio-therapy
= 15 min
and discus-
18 years or older
sion group: 9
67.8% women wks; 9 lessons
Sweden (2) Inter-

net-based ap-
plied relax-
ation: 9 wks;
147
9 online mod-
ules
Library
Cochrane
Hed- Social Phobia ICBT M age=35.2 19.8% 126 ICBT with M=17.4 15 weekly 2.5- post- diagnostic status;
man et (11.1) GCBT M SSRIs, email: 15 wks; emails hour sessions treat- social phobia sx;
al (2011) 47.5% Anxiety Disorder, age=35.5 (11.6) 4.8% SN- 15 online mod- per par- of GCBT ment QOL; general anxi-
15.1% MDD RIs ules ticipant ety sx
18-64 years 6 month
M time fol-
Better health.
Informed decisions.
Trusted evidence.
35.7% women spent low-up
per par-
Sweden
ticipant
per week
= 5.5 min
(SD=3.6)
John- 20.6% PD with or without M age=41.62 (12.83) 29% us- 139 ICBT with email M=8.83 Waiting List post- disorder-specific
ston et Agoraphobia ing med- & phone: 10 (SD=3.29) Control treat- sx; general anxiety
al (2011) 19-79 years ication wks; 8 online emails ment sx; QOL; treatment
34.4% Social Phobia modules per par- satisfaction
58.8% women
ticipant
45% GAD
Australia
M=7.54
29% Anxiety Disorder,
(SD=2.43)
9.2% Affective Disorder,
phone
32.1% both disorders
calls per
partici-
pant
M to-
tal time
spent
per par-
ticipant

= 69.09
min
(SD=32.29)
Kiropou- 41.9% PD M age=38.96 (11.13) 47.7% 86 ICBT with M=18.24 12 weekly 1- post- diagnostic status;
los et al using email: 6 wks, (SD=9.82) hour sessions treat- PD and agorapho-
(2008) 58.1% PD with Agorapho- 20-64 years medica- 6 required & 2 emails of individual ment bia sx; general anx-
bia tion optional online from CBT iety sx; QOL; treat-
72.1% women
modules therapist ment satisfaction
72.1% co-morbid Mood,
Anxiety, Somatoform, or Australia
M=10.64
Substance Disorder (SD=8.21)
148
emails
from
Library
Cochrane
partici-
pant
M to-
tal time
spent
per par-
Better health.
Informed decisions.
Trusted evidence.
ticipant
= 352
min
(SD=240)
Paxling GAD M age=39.3 (10.8) 37.1% 89 ICBT with M to- Waiting List post- GAD sx; general
et al using email: 8 wks; 8 tal time Control treat- anxiety sx; QOL
(2012) co-morbidity included but 18-66 years medica- online modules spent ment
not fully reported; 22.5% tion per par-
MDD 79.8% women
ticipant
Sweden = 97 min
(SD=52)
Richards 21.9% PD M age=36.59 (9.9) 15.6% 23 ICBT with M=18 Information post- diagnostic status;
et al anti-de- email: 8 wks, 6 (SD=6.5) Only Control treat- PD and agorapho-
(2006) 78.1% PD with Agorapho- 18-70 years pres- online modules emails ment bia sx; general anx-
bia sants, from Weekly sta- iety sx; QOL
68.8% women tus updates to
12.5% therapist
22% Social Phobia, 13% clinician and
Australia BZ, 9.4%
GAD, 9% Specific Pho- M=15.3 access to on-
antide-
bia, 6% PTSD, 9% MDD, (SD=12.8) line non-CBT
pres-
6% Hypochondriasis, 3% emails info
sants
Somatization from
and BZ
partici-
pant

M to-
tal time
spent
per par-
ticipant
=376.3
min
(SD=156.8)
149
Robin- GAD M age=46.96 (12.7) Included 101 ICBT with email M =33.2 Waiting List post- diagnostic status;
son et al but not and phone: 10 (SD=4) Control treat- GAD sx; QOL; treat-
Library
Cochrane
(2010) co-morbidity included but 18-80 years reported wks; 6 online emails/ ment ment satisfaction
not reported modules calls per
68.3% women
partici-
Australia pant
M to-
tal time
Better health.
Informed decisions.
Trusted evidence.
spent
per par-
ticipant
= 80.8
min
(SD=22.6)
Silfver- 7% PD M age=32.4 (6.9) 47% us- 57 ICBT with M time Waiting List post- diagnostic status;
nagel et ing med- email: 8 wks; spent Control treat- PD sx; general anxi-
al (2012) 83% PD with Agoraphobia 20-45 years ication 6-8 online mod- per par- ment ety sx; QOL
ules ticipant
16% Social Phobia 65% women
= 15
19% GAD Sweden min/
week
2% ADNOS
32% co-morbid disorder
Spence PTSD M age=42.6 (13.1) 60% us- 44 ICBT with email M=5.39 Waiting List post- diagnostic remis-
et al ing med- & phone: 8 wks; (SD=3.54) Control treat- sion; PTSD sx; QOL;
(2011) 62% MDD, 33% Social 21-68 years ication 7 online mod- emails ment general anxiety sx;
Phobia, 31% PD with or ules per par- treatment satisfac-
without Agoraphobia, 81% women
ticipant tion
26% GAD; 17% OCD
Australia

M=7.87
(SD=2.56)
phone
calls per
partici-
pant
M to-
tal time
spent
per par-
ticipant
150
= 103.91
min
(SD=96.53)
Library
Cochrane
Tillfors Social Phobia ICBT M age=32.3 (9.7) Included 38 ICBT with M=35 ICBT with post- social phobia sx;
et al but not email: 9 wks; 9 min per email (9 on- treat- general anxiety sx;
(2008) co-morbidity included but ICBT+exposure M reported online modules partici- line modules) ment QOL; treatment
not reported age= 30.4 (6.3) pant per + 5 live 2.25- satisfaction
week hour expo- 12-
19-53 years month
sure sessions;
Better health.
Informed decisions.
Trusted evidence.
9 wks fol-
78.9% women
low-up
Sweden
Titov Social Phobia M age=38.13 (12.24) 29% us- 105 ICBT with M to- Waiting List post- social phobia sx;
et al ing med- email: 10 wks; 6 tal time Control treat- QOL; treatment
(2008a) co-morbidity included but 18-72 years ication online modules spent ment satisfaction
not reported per par-
59% women
ticipant
Australia = 125
min
(SD=25)
Titov Social Phobia M age=36.79 (10.93) 25.9% 88 ICBT with M to- Waiting List post- social phobia sx;
et al using email: 10 wks; 6 tal time Control treat- QOL; treatment
(2008b) co-morbidity included but 20-61 years medica- online modules spent ment satisfaction
not reported tion per par-
62.96% women
ticipant
Australia = 126.76
min
(SD=30.89)
Titov Social Phobia M age=37.97 (11.29) 25.9% 98 ICBT with M to- (1) Unguided post- social phobia sx;
et al using email: 10 wks; 6 tal time ICBT treat- QOL; treatment

(2008c) co-morbidity included but 18-64 years medica- online modules spent ment satisfaction
not reported tion per par- 10 wks; 6 on-
61.05% women line modules
ticipant
Australia = 168
(2) Waiting
min
List Control
(SD=40)
Titov et GAD M age=44 (12.98) 29% us- 48 ICBT with email M =23.7 Waiting List post- diagnostic status;
al (2009) ing med- & phone: 9 wks, emails, Control treat- GAD sx; QOL; treat-
co-morbidity included but 18 years or older ication 6 online mod- 5.5 in- ment ment satisfaction
not reported ules stant
76% women
mes-
151
Australia sages,
and 4.1
Library
Cochrane
calls per
partici-
pant
M to-
tal time
spent
Better health.
Informed decisions.
Trusted evidence.
per par-
ticipant
= 130
min
Titov et 26.9% PD with Agorapho- M age=39.5 (13) 47.4% 86 ICBT with email M Waiting List post- diagnostic status;
al (2010) bia using & phone: 8 wks; =23.6emails Control treat- disorder-specific
18 years or older medica- 6 online mod- from ment anxiety sx; QOL;
29.5% Social Phobia tion ules therapist treatment satisfac-
67.9% women
tion
43.6% GAD M to-
Australia
tal time
28.2% Anxiety Disorder,
spent
20.5% Affective Disorder,
per par-
26.9% both disorders
ticipant
= 46 min
(SD=16)
Titov et 10% PD with or without M age=43.9 (14.6) 54% us- 74 ICBT with email M=5.45 Waiting List post- disorder-specific
al (2011) Agoraphobia ing med- & phone: 10 (SD=3.57) Control treat- sx; general anxiety
18-79 years ication wks; 8 online emails ment sx; QOL; treatment
11% Social Phobia modules per par- satisfaction
73% women
ticipant
28% GAD
Australia
M=9.35

51% MDD
(SD=2.96)
81% had a co-morbidity phone
calls per
partici-
pant
M to-
tal time
spent
per par-
ticipant
= 84.76
152
min
(SD=50.37)
Library
Cochrane
Van Bal- 78% PD with or without M age=36.6 (11.4) Included 126 ICBT with M to- Waiting List post- PD sx; general anx-
legooi- Agoraphobia but not email: 12 wks; 6 tal time Control treat- iety sx
jen et al 18-67 years reported online modules spent ment
(2013) 14% Agoraphobia per par-
67.5% women
ticipant
co-morbidity included but
Netherlands = 1 to 2
Better health.
Informed decisions.
Trusted evidence.
not reported
hours
Wims et PD with or without Agora- M age=42.08 (12.29) 31% us- 59 ICBT with M =7.5 Waiting List post- diagnostic status;
al (2010) phobia ing med- email: 8 wks; 6 emails Control treat- PD & agoraphobia
18 years or older ication online modules from ment sx; QOL
21% Social Phobia, 31% therapist
GAD, 10% OCD, 7% PTSD, 76% women
21% MDD M to-
Australia
tal time
spent
per par-
ticipant
= 75 min
Notes: All data in the above table represent only that included in/relevant to the present review.
ADNOS = anxiety disorder, not otherwise specified; Bib = Bibliotherapy; BZ = benzodiazepine; ED = eating disorder; GAD = generalized anxiety disorder; GCBT = group cognitive
behavioural therapy; IBT = internet-based behavioural therapy; ICBT = internet-based cognitive behavioural therapy; IPDT = internet-based psychodynamic therapy; MDD = major
depressive disorder; PD = panic disorder; QOL = quality of life; SNRI = serotonin-norepinephrine re-uptake inhibitor; SSRI = selective serotonin re-uptake inhibitor; sx = symptoms;
TCA = tricyclic antidepressant; VCBT = videoconferencing cognitive-behavioural therapy; WLC = waiting list control.
Table 2. Subgroup analyses. Comparison 1: therapist-supported ICBT versus waiting list control
Outcome and Subgroup No. of No. of Participants Statistical Method Effect I2

Studies Size
ICBT Comparator
Clinically Important Improvement in Anxiety at Post-Treatment
a. By Disorder
i) Panic 2 42 39 RR, M-H, Random 18.32 3

[2.50,
134.18]
153
Table 2. Subgroup analyses. Comparison 1: therapist-supported ICBT versus waiting list control (Continued)
ii) Social Phobia 1 102 102 RR, M-H, Random 6.00 --

[2.64,
Library
Cochrane
13.62]
iii) GAD 3 91 94 RR, M-H, Random 2.58 36

[1.48,
4.51]
Better health.
Informed decisions.
Trusted evidence.
iv) PTSD 1 23 19 RR, M-H, Random 2.89 --
[1.14,
7.33]
v) Specific Phobia 0 -- -- -- -- --
vi) Trans-diagnostic 2 67 65 RR, M-H, Random 6.12 0

[2.54,
14.77]
b. By Therapist Contact
i) High 1 12 9 RR, M-H, Random 6.92 --

[0.42,
114.19]
ii) Medium 6 226 224 RR, M-H, Random 4.74 35

[2.66,
8.46]
iii) Low 2 87 86 RR, M-H, Random 2.80 57

[1.17,
6.66]

c. By Research Group
i) Sweden 4 179 184 RR, M-H, Random 5.76 53

[2.26,
14.70]
ii) Australia 1 4 134 126 RR, M-H, Random 3.06 43

[1.79,
5.23]
154
iii) Australia 2 1 47 48 RR, M-H, Random 2.10 --

[1.44,
Library
Cochrane
3.04]
Anxiety Symptom Severity at Post-Treatment
Better health.
Informed decisions.
Trusted evidence.
a. By Disorder
i) Panic 5 155 147 SMD, Random -1.58 94

[-2.79,
-0.37]
ii) Social Phobia 7 324 314 SMD, Random -1.44 25

[-1.65,
-1.23]
iii) GAD 4 138 140 SMD, Random -0.91 72

[-1.40,
-0.43]
iv) PTSD 1 23 19 SMD, Random -0.45 --

[-1.06,
0.17]
vi) Trans-diagnostic 5 161 152 SMD, Random -0.75 52

[-1.10,
-0.40]

i) High 1 12 9 SMD, Random -0.80 --

[-1.70,
0.11]
ii) Medium 14 556 552 SMD, Random -1.30 87

[-1.68,
-0.92]
155
iii) Low 7 233 211 SMD, Random -0.87 68

[-1.23,
Library
Cochrane
-0.50]
i) Sweden 10 408 409 SMD, Random -1.51 91

[-2.05,
Better health.
Informed decisions.
Trusted evidence.
-0.97]
ii) Australia 1 10 350 333 SMD, Random -0.85 53

[-1.09,
-0.61]
iii) Australia 2 1 12 9 SMD, Random -0.80 --

[-1.70,
0.11]
General Anxiety Symptom Severity at Post-Treatment
a. By Disorder
i) Panic 4 126 122 SMD, Random -0.74 78

[-1.35,
-0.13]
ii) Social Phobia 3 171 170 SMD, Random -0.64 0

[-0.85,
-0.42]

iii) GAD 2 67 71 SMD, Random -1.91 94
[-3.57,
-0.26]
iv) PTSD 1 23 19 SMD, Random -0.53 --

[-1.15,
0.09]
156
vi) Trans-diagnostic 4 121 114 SMD, Random -0.49 0

[-0.75,
Library
Cochrane
-0.23]
i) High 1 12 9 SMD, Random 0.07 --

[-0.79,
Better health.
Informed decisions.
Trusted evidence.
0.94]
ii) Medium 10 410 408 SMD, Random -0.89 85

[-1.27,
-0.50]
iii) Low 3 86 79 SMD, Random -0.67 48

[-1.13,
-0.21]
i) Sweden 10 408 409 SMD, Random -0.95 85

[-1.33,
-0.56]
ii) Australia 1 3 88 78 SMD, Random -0.51 0

[-0.82,
-0.20]
iii) Australia 2 1 12 9 SMD, Random 0.07 --

[-0.79,
0.94]

Table 3. Subgroup analyses. Comparison 2: therapist-supported ICBT versus unguided CBT
Outcome and Subgroup No. of No. of Participants Statistical Method Effect Size I2
Studies
ICBT Comparator
a. By Disorder
157
Table 3. Subgroup analyses. Comparison 2: therapist-supported ICBT versus unguided CBT (Continued)
i) Panic 0 -- -- -- -- --
Library
Cochrane
ii) Social Phobia 1 27 27 RR, M-H, Random 1.07 [0.67, 1.69] --
iii) GAD 0 -- -- -- -- --
iv) PTSD 0 -- -- -- -- --
Better health.
Informed decisions.
Trusted evidence.
vi) Trans-diagnostic 0 -- -- -- -- --
i) High 0 -- -- -- -- --
ii) Medium 0 -- -- -- -- --
iii) Low 1 27 27 RR, M-H, Random 1.07 [0.67, 1.69] --
i) Sweden 1 27 27 RR, M-H, Random 1.07 [0.67, 1.69] --
ii) Australia 1 0 -- -- -- -- --
iii) Australia 2 0 -- -- -- -- --

a. By Disorder
i) Panic 0 -- -- -- -- --
ii) Social Phobia 4 127 126 SMD, Random -0.24 [-0.69, 0.21] 68
iii) GAD 0 -- -- -- -- --
iv) PTSD 0 -- -- -- -- --
158
Library
Cochrane
i) High 0 -- -- -- -- --
Better health.
Informed decisions.
Trusted evidence.
ii) Medium 3 100 99 SMD, Random -0.18 [-0.78, 0.41] 77
iii) Low 1 27 27 SMD, Random -0.43 [-0.97, 0.11] --
i) Sweden 3 96 96 SMD, Random -0.04 [-0.38, 0.30] 30
ii) Australia 1 1 31 30 SMD, Random -0.82 [-1.34, -0.29] --
iii) Australia 2 0 -- -- -- -- --
a. By Disorder
i) Panic 0 -- -- -- -- --
ii) Social Phobia 2 69 69 MD, Random 0.28 [-2.21, 2.78] 0
iii) GAD 0 -- -- -- -- --

iv) PTSD 0 -- -- -- -- --
i) High 0 -- -- -- -- --
159
ii) Medium 2 69 69 MD, Random 0.28 [-2.21, 2.78] 0
Library
Cochrane
iii) Low 0 -- -- -- -- --
i) Sweden 2 69 69 MD, Random 0.28 [-2.21, 2.78] 0
Better health.
Informed decisions.
Trusted evidence.
ii) Australia 1 0 -- -- -- -- --
iii) Australia 2 0 -- -- -- -- --
Clinically Important Improvement in Anxiety at Follow-up
a. By Disorder
i) Panic 0 -- -- -- -- --
ii) Social Phobia 0 -- -- -- -- --
iii) GAD 0 -- -- -- -- --
iv) PTSD 0 -- -- -- -- --

i) High 0 -- -- -- -- --
ii) Medium 0 -- -- -- -- --
iii) Low 0 -- -- -- -- --
i) Sweden 0 -- -- -- -- --
160
ii) Australia 1 0 -- -- -- -- --
Library
Cochrane
iii) Australia 2 0 -- -- -- -- --
Anxiety Symptom Severity at Follow-up
Better health.
Informed decisions.
Trusted evidence.
a. By Disorder
i) Panic 0 -- -- -- -- --
ii) Social Phobia 3 96 96 SMD, Random -0.30 [-0.58, -0.01] 0
iii) GAD 0 -- -- -- -- --
iv) PTSD 0 -- -- -- -- --
b. By Therapist Contact at Follow-up
i) High 0 -- -- -- -- --
ii) Medium 2 69 69 SMD, Random -0.31 [-0.65, 0.03] 3
iii) Low 1 27 27 SMD, Random -0.26 [-0.80, 0.27] --

i) Sweden 3 96 96 SMD, Random -0.30 [-0.58, -0.01] 0
ii) Australia 1 0 -- -- -- -- --
iii) Australia 2 0 -- -- -- -- --
General Anxiety Symptom Severity at Follow-up

161
a. By Disorder
Library
Cochrane
i) Panic 0 -- -- -- -- --
ii) Social Phobia 2 69 69 MD, Random 0.72 [-2.12, 3.57] 0
iii) GAD 0 -- -- -- -- --
Better health.
Informed decisions.
Trusted evidence.
iv) PTSD 0 -- -- -- -- --
i) High 0 -- -- -- -- --
ii) Medium 2 69 69 MD, Random 0.72 [-2.12, 3.57] 0
iii) Low 0 -- -- -- -- --
i) Sweden 2 69 69 MD, Random 0.72 [-2.12, 3.57] 0
ii) Australia 1 0 -- -- -- -- --
iii) Australia 2 0 -- -- -- -- --

Table 4. Subgroup analyses. Comparison 3: therapist-supported ICBT versus face-to-face CBT
Outcome and Subgroup No. of No. of Participants Statistical Method Effect Size I2
Studies
ICBT Comparator
a. By Disorder
162
Table 4. Subgroup analyses. Comparison 3: therapist-supported ICBT versus face-to-face CBT (Continued)
i) Panic 3 121 118 RR, M-H, Random 1.06 [0.85, 1.32] 0
Library
Cochrane
ii) So- 1 64 62 RR, M-H, Random 1.45 [0.77, 2.76] --
cial
Phobia
iii) GAD 0 -- -- -- -- --
Better health.
Informed decisions.
Trusted evidence.
iv) 0 -- -- -- -- --
PTSD
v) Spe- 0 -- -- -- -- --
cific
Phobia
vi) 0 -- -- -- -- --
Trans-
diag-
nostic
i) High 1 46 40 RR, M-H, Random 1.11 [0.57, 2.15] --
ii) 1 25 24 RR, M-H, Random 1.20 [0.85, 1.69] --

Medi-
um
iii) Low 2 114 116 RR, M-H, Random 1.08 [0.72, 1.60] 34

i) Swe- 3 139 140 RR, M-H, Random 1.09 [0.88, 1.36] 0
den
ii) Aus- 0 -- -- -- -- --
tralia 1
iii) Aus- 1 46 40 RR, M-H, Random 1.11 [0.57, 2.15] --

tralia 2
163
Library
Cochrane
a. By Disorder
i) Panic 3 119 115 SMD, Random 0.29 [0.03, 0.54] 0
ii) So- 2 83 80 SMD, Random -0.18 [-0.92, 0.57] 76

cial
Better health.
Informed decisions.
Trusted evidence.
Phobia
iii) GAD 0 -- -- -- -- --
iv) 0 -- -- -- -- --
PTSD
v) Spe- 1 13 14 SMD, Random 0.09 [-0.66, 0.85] --

cific
Phobia
vi) 0 -- -- -- -- --
Trans-
diag-
nostic
i) High 2 63 55 SMD, Random 0.42 [0.05, 0.78] 0
ii) 1 25 24 SMD, Random 0.05 [-0.51, 0.61] --

Medi-
um

iii) Low 3 127 130 SMD, Random -0.08 [-0.63, 0.46] 76
i) Swe- 5 171 172 SMD, Random -0.01 [-0.36, 0.35] 59

den
ii) Aus- 0 -- -- -- -- --
tralia 1
164
iii) Aus- 1 44 37 SMD, Random 0.49 [0.05, 0.93] --

tralia 2
Library
Cochrane
a. By Disorder
Better health.
Informed decisions.
Trusted evidence.
i) Panic 2 67 62 SMD, Random 0.42 [-0.75, 1.60] 90
ii) So- 2 83 80 SMD, Random -0.18 [-0.49, 0.13] 0

cial
Phobia
iii) GAD 0 -- -- -- -- --
iv) 0 -- -- -- -- --
PTSD
v) Spe- 1 13 12 SMD, Random 0.33 [-0.46, 1,12] --

cific
Phobia
vi) 0 -- -- -- -- --
Trans-
diag-
nostic
b. By Therapist Con-
tact

i) High 2 61 56 SMD, Random 0.49 [-0.57, 1.56] 86
ii) 1 25 24 SMD, Random -0.19 [-0.75, 0.38] --

Medi-
um
165
i) Swe- 4 121 116 SMD, Random -0.13 [-0.38, 0.13] 0

den
Library
Cochrane
ii) Aus- 0 -- -- -- -- --
tralia 1
iii) Aus- 1 42 38 SMD, Random 1.01 [0.54, 1.48] --

tralia 2
Better health.
Informed decisions.
Trusted evidence.
Clinically Important Improvement in Anxiety at Follow-up
a. By Disorder
i) Panic 2 75 78 RR, M-H, Random 1.09 [0.93, 1.28] 0
ii) So- 1 64 62 RR, M-H, Random 1.15 [0.73, 1.83] --

cial
Phobia
iii) GAD 0 -- -- -- -- --
iv) 0 -- -- -- -- --
PTSD
v) Spe- 0 -- -- -- -- --
cific
Phobia
vi) 0 -- -- -- -- --
Trans-

diag-
nostic
i) High 0 -- -- -- -- --
ii) 1 25 24 RR, M-H, Random 1.05 [0.87, 1.27] --

Medi-
um
166
iii) Low 2 114 116 RR, M-H, Random 1.17 [0.92, 1.50] 0
Library
Cochrane
i) Swe- 3 139 140 RR, M-H, Random 1.10 [0.94, 1.27] 0

den
ii) Aus- 0 -- -- -- -- --
Better health.
Informed decisions.
Trusted evidence.
tralia 1
iii) Aus- 0 -- -- -- -- --
tralia 2
Anxiety Symptom Severity at Follow-up
a. By Disorder
i) Panic 2 75 78 SMD, Random -0.04 [-0.36, 0.28] 0
ii) So- 2 83 80 SMD, Random -0.39 [-0.71, -0.08] 0

cial
Phobia
iii) GAD 0 -- -- -- -- --
iv) 0 -- -- -- -- --
PTSD
v) Spe- 1 13 12 SMD, Random -0.09 [-0.88, 0.69] --

cific

Phobia
vi) 0 -- -- -- -- --
Trans-
diag-
nostic
i) High 1 19 18 SMD, Random -0.11 [-0.76, 0.53] --

167
ii) 1 25 24 SMD, Random -0.04 [-0.60, 0.52] --

Medi-
Library
Cochrane
um
Better health.
Informed decisions.
Trusted evidence.
i) Swe- 5 171 170 SMD, Random -0.21 [-0.42, 0.00] 0
den
ii) Aus- 0 -- -- -- -- --
tralia 1
iii) Aus- 0 -- -- -- -- --
tralia 2
General Anxiety Symptom Severity at Follow-up
a. By Disorder
i) Panic 1 25 24 SMD, Random -0.17 [-0.74, 0.39] --
ii) So- 2 83 80 SMD, Random -0.14 [-0.45, 0.17] 0

cial
Phobia
iii) GAD 0 -- -- -- -- --
iv) 0 -- -- -- -- --

PTSD
v) Spe- 1 13 12 SMD, Random -0.31 [-1.10, 0.48] --

cific
Phobia
vi) 0 -- -- -- -- --
Trans-
diag-
nostic
168
b. By Therapist Con-
tact
Library
Cochrane
i) High 1 19 18 SMD, Random -0.19 [-0.83, 0.46] --
ii) 1 25 24 SMD, Random -0.17 [-0.74, 0.39] --

Medi-
um
Better health.
Informed decisions.
Trusted evidence.
i) Swe- 4 121 116 SMD, Random -0.16 [-0.42, 0.09] 0

den
ii) Aus- 0 -- -- -- -- --
tralia 1
iii) Aus- 0 -- -- -- -- --
tralia 2

169
Informed decisions.
APPENDICES
Appendix 1. CCDANCTR update search (25 September 2014)

#1 (internet* or online or web*):ti
#2 (*phobi* or panic or "anxiety disorder*" or (anxiety and depression) or GAD or "general* anxiety" or OCD or obsess* or PTSD or *trauma*
or "stress disorder*"):ti
#3 (assisted or administer* or coach* or guided or guidance or *therapist* or ((telephone or email) next (support or assist*))):ti,ab
#4 (#1 and #2 and #3)
#5 (2012* or 2013* or 2014*):yr,xdd
[ti:title; ab:abstract; yr:year; xdd:record entry date]
CONTRIBUTIONS OF AUTHORS
This review was prepared primarily by Dr Olthuis in close collaboration with respect to its content (search criteria, search methods, data
analysis and interpretation) with Dr Watt and Dr Stewart. Data extraction and ROB assessment were completed by Dr Olthuis and Ms Bailey.
Dr Hayden provided an extensive contribution with respect to Cochrane protocol and methods throughout the review process and the
preparation of the review manuscript.
DECLARATIONS OF INTEREST
None of the authors have known competing interests.
SOURCES OF SUPPORT
Internal sources
• Dalhousie University Department of Psychology and Neuroscience and Department of Psychiatry, Canada.
Dr Stewart is supported by the Department of Psychology and Neuroscience and the Department of Psychiatry at Dalhousie University.
• Saint Francis Xavier University Department of Psychology, Canada.
Dr Watt is supported by the Department of Psychology at Saint Francis Xavier University.

• Killam Graduate Student Scholarship (Dalhousie University), Canada.
Dr Olthuis's graduate studies at the time of completion of this review were supported by a Killam Graduate Student Scholarship.
• Dalhousie University Department of Community Health and Epidemiology, Canada.
Dr Hayden is supported by the Department of Community Health and Epidemiology at Dalhousie University for administrative support
and office space.
• Dalhousie University Department of Psychology, Canada.
Ms Bailey is supported by the Department of Psychology at Dalhousie University.
External sources
• Nova Scotia Health Research Foundation, Canada.
This project was supported by a Knowledge Transfer/Exchange Systematic Review Grant to Dr Watt, Dr Olthuis, and Dr Stewart from the
Nova Scotia Health Research Foundation.
• Canadian Institutes of Health Research, Vanier Canada Graduate Scholarship, Canada.
Dr Olthuis's graduate studies at the time of completion of this review were supported by a Vanier Canada Graduate Scholarship from
the Canadian Institutes of Health Research.
• Canadian Institutes of Health Research, Pain in Child Health Initiative, Canada.
Ms Bailey is supported by Pain in Child Health, a strategic training initiative by the Canadian Institutes of Health Research.
Informed decisions.
• Nova Scotia Health Research Foundation, Canada.
The Nova Scotia Health Research Foundation provides infrastructure funding to Dr Hayden to support to the Nova Scotia Cochrane
Resource Centre.
• Canadian Chiropractic Research Foundation/Dalhousie University, Canada.
Dr Hayden is supported by a Research Professorship Award from the Canadian Chiropractic Research Foundation.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

Several changes were made to our protocol (Olthuis 2011) during the transition from protocol to full review. They are listed here.
1. In the protocol, we planned to assess the efficacy of a broader range of distance delivery treatments including Internet-supported
CBT and also CBT delivered by phone or videoconferencing. A reviewer commented that these different types of distance delivery were
too broad and dissimilar to be included in the same review. As such, we have now focused the review only on Internet-supported CBT
interventions to increase homogeneity across included studies and to improve interpretation of findings. With the exception of the
characteristics of the experimental intervention, the protocol remains largely unchanged.
2. In the protocol we stated that 'relaxation' could qualify as a CBT intervention. This was an oversight; on further consideration the review
team decided there were significant differences between relaxation alone and the key components of CT, BT, and CBT. Thus, while relaxation
could qualify as part of an intervention of interest if it was presented as a component of a more comprehensive CBT package, we did not
include therapist-supported Internet-based relaxation as an intervention of interest.
3. We originally planned to include quasi-RCTs, as stated in our protocol. However, the field was more developed than we anticipated.
Thus, in order to increase the strength of the evidence within the review, we elected to exclude quasi-RCTs and include only RCTs.
4. In the original protocol we had designated the first primary outcome as the efficacy of therapist-supported ICBT in reducing anxiety,
as measured by either remission of anxiety disorder diagnosis or reduction in anxiety symptom severity. With respect to the latter, we
specified that a reduction in anxiety symptom severity could be indexed by measures of either disorder-specific anxiety symptoms or
anxiety symptoms in general. On further consideration, we decided amalgamating these two types of measures resulted in lost information
about the efficacy of the intervention. Thus, in the review we indexed the efficacy of therapist-supported ICBT in reducing anxiety as
measured by (a) remission of anxiety disorder diagnosis, (b) a reduction in disorder-specific anxiety symptoms, and (c) a reduction in
anxiety symptoms in general.
5. The protocol listed our time periods for outcome assessment as short-term (less than 12 months) and long-term (12 months or greater).
Later, we decided that we wanted to select one time period that would maximize the number of studies that could be included and would
be clinically meaningful. Thus, we consolidated our follow-up assessment to one time point, 6 to 12 months.
6. In the original protocol, we planned to assess dropout and treatment adherence as a secondary outcome. After reviewing the included
studies, we observed that so many different methods of indexing dropout were used (e.g., number of participants not completing entire
treatment protocol, number of participants not completing 75% of treatment protocol, number of participants not completing follow-up
questionnaires) that combining these measures across studies did not lend itself to any type of meaningful interpretation. As such, rather
than examine dropout and treatment adherence as a separate outcome, we elected to assess this outcome via risk of bias and sensitivity
analyses. More specifically, in the risk of bias evaluation, we identified studies that did not use an adequate ITT paradigm in their data
analytic procedure; then we excluded these studies using sensitivity analyses.
7. We removed the originally planned sensitivity analysis which would have excluded cross-over trials with carry-over effects. The inclusion
of this sensitivity analysis in our protocol was an oversight as we had elected to only include data from participants before they crossed
over to their second treatment condition.
8. In response to suggestions by review editors, we added in a subgroup analysis (by research group) and two sensitivity analyses (exclusion
of studies with an active waiting list control; assuming that treatment dropouts were responders on dichotomous outcomes) that were
not proposed in the original published protocol.
NOTES
An updated search conducted in September 2014 identified four new completed studies, seven previously ongoing studies that have now
been completed, and three new ongoing studies that should be included in the present review. This is a fast-moving area of research and
as such we plan to conduct an update of this review after its publication, in which these new studies will be fully incorporated.
Informed decisions.
INDEX TERMS
Medical Subject Headings (MeSH)

*Internet; Agoraphobia [therapy]; Anxiety Disorders [*therapy]; Depressive Disorder [therapy]; Phobic Disorders [therapy];
Randomized Controlled Trials as Topic
MeSH check words

Adult; Aged; Humans; Middle Aged

Cochrane (19)

Uploaded by

Copyright:

Available Formats

Cochrane (19)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cochrane (19)

Uploaded by

Copyright:

Available Formats

Cochrane

Therapist-supported Internet cognitive behavioural therapy for

Olthuis JV, Watt MC, Bailey K, Hayden JA, Stewart SH

Olthuis JV, Watt MC, Bailey K, Hayden JA, Stewart SH.

Therapist-supported Internet cognitive behavioural therapy for anxiety

Editorial group: Cochrane Common Mental Disorders Group

Data collection and analysis

PLAIN LANGUAGE SUMMARY

Who may be interested in this review?

People who suffer from anxiety and their families.

Professionals working in psychological therapy services.

Developers of Internet-based therapies for mental health problems.

Why is this review important?

What questions does this review aim to answer?

The review aims to answer the following questions:

Which studies were included in the review?

What does the evidence from the review tell us?

Patient or population: patients with anxiety disorders

Waiting list, Therapist-supported ICBT

Clinically im- Study population RR 4.18 644 ⊕⊕⊕⊝

Cochrane Database of Systematic Reviews

Cochrane Database of Systematic Reviews

GRADE Working Group grades of evidence

Therapist-supported ICBT compared to unguided CBT for anxiety disorders in adults

Patient or population: patients with anxiety disorders

Unguided Therapist-supported ICBT

Cochrane Database of Systematic Reviews

Cochrane Database of Systematic Reviews

GRADE Working Group grades of evidence

Therapist-supported ICBT compared to face-to-face CBT for anxiety disorders in adults

Cochrane Database of Systematic Reviews

Cochrane Database of Systematic Reviews

GRADE Working Group grades of evidence

Cochrane Database of Systematic Reviews

BACKGROUND Unfortunately, certain barriers (for example, time constraints,

health clinics, or private practice clinics. Participants could be Comparator interventions

Summary of findings Results of the search

Figure 1. PRISMA diagram of the search process.

E-mail correspondence to collect information to supplement Participants

Sample sizes Twenty-eight trials included participants who were using

Interventions Four studies compared therapist-supported ICBT to unguided CBT

Allocation As participants were not blind to their treatment condition in

1.2 Reduction in disorder-specific anxiety symptom severity

Primary outcomes difficult to estimate heterogeneity (I2 = 0%). No sensitivity analyses

Newman 1997 {published data only}

Lindner P, Ivanova E, Ly KH, Andersson G, & Carlbring P. Guided Andersson 2012

Beck 1991 Deacon 2004

Casey 2004 First 2002

Choy 2007 Foa 2010

Clark 1986 Frisch 1992

in the 1990s. Journal of Clinical Psychiatry 1999;60:427-35. [DOI: Higgins 2011b

Olatunji 2010 Robins 1991

Ost 1997 Salkovskis 1985

Otto 2000 Slade 2007

and anxiety: A meta-analysis. Psychological Medicine Westra 1998

* Indicates the major publication for the study

Characteristics of included studies [ordered by study ID]

Participants Diagnosis: DSM-IV Specific Phobia, Spider Type

Method of diagnosis: SCID-IV

Age: M = 25.6 (SD = 4.1); range = 18 to 65 years