Internet-Based and Other Computerized Psychological Treatments for Adult Depression: A Meta-Analysis

Abstract

Computerized and, more recently, Internet-based treatments for depression have been developed and tested in controlled trials. The aim of this meta-analysis was to summarize the effects of these treatments and investigate characteristics of studies that may be related to the effects. In particular, the authors were interested in the role of personal support when completing a computerized treatment. Following a literature search and coding, the authors included 12 studies, with a total of 2446 participants. Ten of the 12 studies were delivered via the Internet. The mean effect size of the 15 comparisons between Internet-based and other computerized psychological treatments vs. control groups at posttest was d = 0.41 (95% confidence interval [CI]: 0.29-0.54). However, this estimate was moderated by a significant difference between supported (d = 0.61; 95% CI: 0.45-0.77) and unsupported (d = 0.25; 95% CI: 0.14-0.35) treatments. The authors conclude that although more studies are needed, Internet and other computerized treatments hold promise as potentially evidence-based treatments of depression.

Full text

PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by:
[Vrije Universiteit, Library]
On:
26 November 2010
Access details:
Access Details: [subscription number 907218003]
Publisher
Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK
Cognitive Behaviour Therapy
Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713926011
Internet-Based and Other Computerized Psychological Treatments for
Adult Depression: A Meta-Analysis
Gerhard Anderssona; Pim Cuijpersb
a Department of Behavioural Sciences and Learning, Swedish Institute for Disability Research,
Linköping University, Linköping, and Department of Clinical Neuroscience, Psychiatry Section,
Karolinska Institutet, Stockholm, Sweden b Department of Clinical Psychology and EMGO+ Institute,
VU University, Amsterdam, the Netherlands
Online publication date: 15 December 2009
To cite this Article Andersson, Gerhard and Cuijpers, Pim(2009) 'Internet-Based and Other Computerized Psychological
Treatments for Adult Depression: A Meta-Analysis', Cognitive Behaviour Therapy, 38: 4, 196 — 205
To link to this Article: DOI: 10.1080/16506070903318960
URL: http://dx.doi.org/10.1080/16506070903318960
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.

Internet-Based and Other Computerized Psychological
Treatments for Adult Depression: A Meta-Analysis
Gerhard Andersson
1
and Pim Cuijpers
2
1
Department of Behavioural Sciences and Learning, Swedish Institute for Disability Research,
Linko
¨ping University, Linko
¨ping, and Department of Clinical Neuroscience, Psychiatry
Section, Karolinska Institutet, Stockholm, Sweden;
2
Department of Clinical Psychology and
EMGOþInstitute, VU University, Amsterdam, the Netherlands
Abstract. Computerized and, more recently, Internet-based treatments for depression have been
developed and tested in controlled trials. The aim of this meta-analysis was to summarize the effects
of these treatments and investigate characteristics of studies that may be related to the effects. In
particular, the authors were interested in the role of personal support when completing a
computerized treatment. Following a literature search and coding, the authors included 12 studies,
with a total of 2446 participants. Ten of the 12 studies were delivered via the Internet. The mean effect
size of the 15 comparisons between Internet-based and other computerized psychological treatments
vs. control groups at posttest was d¼0.41 (95% confidence interval [CI]: 0.29–0.54). However, this
estimate was moderated by a significant difference between supported (d¼0.61; 95% CI: 0.45 –0.77)
and unsupported (d¼0.25; 95% CI: 0.14–0.35) treatments. The authors conclude that although
more studies are needed, Internet and other computerized treatments hold promise as potentially
evidence-based treatments of depression. Key words: computerized treatment; depression; Internet-
based; role of support
Received 15 August, 2009; Accepted 8 September, 2009
*Correspondence address: Gerhard Andersson, PhD, Department of Behavioural Sciences,
Linko
¨ping University, SE-581 83 Linko
¨ping, Sweden. Tel: þ46 13 28 58 40; Fax: þ46 13 28 21
45. E-mail: Gerhard.Andersson@liu.se
Adult depression is a costly condition for
which numerous different treatments have
been developed (Ebmeier, Donaghey, &
Steele, 2006). Among the psychological treat-
ment options, several have been found to be
effective (Cuijpers, van Straten, Andersson, &
van Oppen, 2008), with no or minor differ-
ences between the main treatment alternatives.
Cognitive behaviour therapy (CBT), however,
has been investigated in by far the most trials
and in different administration formats such
as group (McDermut, Miller, & Brown, 2001)
and telephone-assisted (e.g. Simon, Ludman,
Tutty, Operskalski, & Von Korff, 2004)
treatments. Computer and, more recently,
Internet delivery have become increasingly
common administration formats for
depression treatment in research and slowly
but gradually in clinical settings as well.
Computerized psychological treatments can
be delivered on devices such as stand-alone or
Internet-linked computers, PCs, palmtops,
phone-interactive voice response systems,
CD-ROMS, DVDs, cell phones, and VR
equipment (Marks, Cavanagh, & Gega,
2007). However, during the last 5 years, it is
mostly Internet-delivered treatments that have
been tested in research.
Internet-delivered treatments of depression
can take on different forms. One approach is
largely based on bibliotherapy, with mainly
text-based materials and guidance by a
therapist via e-mail or phone. On the other
end of the continuum, we find treatments that
are briefer and usually do not target persons
with clinical depression. Moreover, these
q2009 Taylor & Francis ISSN 1650-6073
DOI: 10.1080/16506070903318960
Cognitive Behaviour Therapy Vol 38, No 4, pp. 196–205, 2009
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

treatments are commonly not delivered with
guidance and can hence be reached by more
people at a lower cost. They are, however,
probably less effective (Spek, Cuijpers, et al.,
2007). Regardless of the role of support,
treatment protocols also differ with regard to
how often and how many online activities are
included and how much feedback is
automated.
There are now several reviews and meta-
analyses on computerized (e.g. Cuijpers,
Marks, et al., 2009; Reger & Gahm, 2009)
and Internet-delivered (Barak, Hen, Boniel-
Nissim, & Shapira, 2008; Griffiths & Chris-
tensen, 2006; Spek, Cuijpers, et al., 2007)
treatments. In one previous review, Internet
treatments for depression were covered
(Andersson, 2006), but this was not a meta-
analysis. The other meta-analyses were not
specifically aimed at depression and have
included only a limited portion of currently
available trials. Given the very strong increase
in the number of controlled trials on
computerized and Internet-delivered treat-
ments of depression in the past few years, we
decided to conduct a new meta-analysis in
which we focused on depression only. We were
interested in investigating the role of support
and the overall effects compared with other
treatments. This meta-analysis adds to the
literature by providing an overview of the field
and by contrasting different approaches to
computerized treatment.
Method
Identification and selection of studies
We used several methods to identify studies
for inclusion. First, we used a database of 1036
studies on the psychological treatment of
depression, which includes reports on com-
bined treatments and comparisons with
pharmacotherapy. This database has been
described in detail elsewhere (Cuijpers, van
Straten, Warmerdam, & Andersson, 2008)
and has been used in a series of earlier meta-
analyses (http://www.evidencebasedpsychoth
erapies.org). The database was developed
through a comprehensive literature search
(from 1966 to January 2009), in which we
examined a total of 9011 abstracts: 1629 from
PubMed, 2439 from PsycINFO, 2606 from
Embase, and 2337 from the Cochrane Central
Register of Controlled Trials. These abstracts
were identified by combining terms indicative
of psychological treatment and depression
(both MeSH terms and text words). For this
database, we also collected the primary studies
from 42 meta-analyses of psychological treat-
ment for depression (http://www.
evidencebasedpsychotherapies.org). For the
current study, we examined the full texts of
these 1036 studies, then examined the refer-
ence lists of earlier reviews and meta-analyses
(e.g. Spek, Cuijpers, et al., 2007), and checked
the references of the included primary studies.
We included (a) randomized trials (b) in
which the effects of an Internet-based or
computerized psychological treatment (c)
were compared with a (noncomputerized)
control or comparison group or a face-to-
face psychological treatment (d) in adults (e)
with depression (established by diagnostic
interview or elevated levels of depressive
symptoms based on self-report measures).
We also included studies that were aimed at
adults with depression and anxiety. We
excluded studies on inpatients and those on
adolescents or children (,18 years). We did
not include component studies (e.g. Christen-
sen, Griffiths, Mackinnon, & Brittliffe, 2006).
Comorbid general medical or psychiatric
disorders were not used as an exclusion
criterion. No language restrictions were
applied.
Meta-analyses
For each comparison between Internet-based
or computerized psychological treatment and
control conditions, we calculated the effect
size indicating the difference between the two
conditions at posttest (Cohen’s dor standar-
dized mean difference). We calculated the
effect sizes by subtracting (at posttest)
the average score of the treatment group
from the average score of the control group
and dividing the result by the pooled standard
deviations of the two groups. Effect sizes of
0.8 can be assumed to be large, 0.5 moderate
and 0.2 small (Cohen, 1988).
In the calculations of effect sizes, we only
used those instruments that explicitly
measured symptoms of depression. If more
than one depression measure was used, the
mean of the effect sizes was calculated, so that
each study only provided one effect size.
We only used the effect sizes indicating the
differences between the two types of treatment
VOL 38, NO 4, 2009 Internet-based and computerized treatments for adult depression 197
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

at posttest. We decided not to examine the
differential effects at follow-up because the
numberofeffectsizeswastoolow.In
addition, the follow-up period differed con-
siderably among these studies.
To calculate pooled mean effect sizes, we
used Biostat’s computer program Comprehen-
sive Meta-Analysis (version 2.2.021). Because
we expected considerable heterogeneity
among the studies, we decided to calculate
mean effect sizes using a random-effects
model. In the random-effects model, it is
assumed that the included studies are drawn
from “populations” of studies that differ from
each other systematically (heterogeneity). In
this model, the effect sizes resulting from
included studies differ not only because of the
random error within studies (as in the fixed-
effects model) but also because of true
variation in effect size from one study to the
next.
As a test of homogeneity of effect sizes, we
calculated the I
2
statistic, which is an
indicator of heterogeneity in percentages. A
value of 0% indicates no observed heterogen-
eity, and larger values show increasing
heterogeneity, with 25% as low, 50% as
moderate, and 75% as high heterogeneity
(Higgins, Thompson, & Deeks, 2003). We also
calculated the Qstatistic but only report
whether this was significant or not.
Subgroup analyses were conducted accord-
ing to the mixed-effect model. In this model,
studies within subgroups are pooled with the
random-effects model, whereas tests for
significant differences between subgroups are
conducted with the fixed-effects model.
Publication bias was tested by inspecting the
funnel plot on primary outcome measures and
by Duval and Tweedie’s (2000) trim and fill
procedure, which yields an estimate of the
effect size after the publication bias has been
taken into account (as implemented in
Comprehensive Meta-Analysis, version
2.2.021).
Results
Characteristics of included studies
A total of 12 studies, with 2446 participants
(1324 in the Internet-based and computerized
psychological treatment conditions, 996 in the
control conditions, and 126 in the face-to-face
comparison conditions) met all inclusion
criteria. Selected characteristics of these
studies are presented in Table 1.
Ten studies were aimed at adults in general,
and one was aimed at older adults and one at
young adults. All but one study recruited
participants from the community. Only two
studies included participants with depressive
disorder diagnosed in a formal diagnostic
interview. Six studies used a wait list control
group, four a care-as-usual control group, and
the remaining two studies another type of
control group. The 12 studies included 15
comparisons between an Internet-based or
computerized psychological treatment and a
control group (three studies included two
comparisons). In three studies, Internet-based
or computerized psychological treatment was
compared with face-to-face psychological
treatment. Eleven of the 15 comparisons
examined CBT, two problem-solving therapy,
and one psychoeducation. Two studies exam-
ined computerized psychological treatment
and the remaining 10 Internet-based psycho-
logical treatment. Five studies were conducted
in the United States, four in the Netherlands,
and one each in Sweden, United Kingdom,
and Australia. In one study only 50% of the
face-to-face treatment was replaced by a
computerized treatment (Wright et al., 2005).
We decided to include this study and examine
whether removal of this resulted in changes of
the mean effect size (which was not the case;
see later discussion).
Internet-based and computerized
psychological treatment versus control
groups: overall effect size
The mean effect size of the 15 comparisons
between Internet-based and computerized
psychological treatment vs. control groups at
posttest was 0.41 (95% confidence interval
[CI]: 0.29–0.54; Table 2). Heterogeneity was
moderate to high (I
2
¼57.49). The effect sizes
and 95% CIs of the individual contrast groups
are plotted in Figure 1.
Inspection of the forest plot suggested that
two studies were possible outliers (Selmi et al.,
1990; Wright et al., 2005). However, after
removal of these studies, the effect size
remained almost the same (d¼0.37; 95%
CI: 0.26–0.49) and heterogeneity remained at
a moderate level (I
2
¼51.36).
198 Andersson and Cuijpers COGNITIVE BEHAVIOUR THERAPY
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

Table 1.Selected characteristics of randomized controlled studies examining the effects of computerized and Internet-based psychotherapies for adult depression
Study
Target
population Recruitment Inclusion Condition NIntervention
Therapist
support
Internet/
computer Measure
a
Country
Andersson
et al. (2005)
Adults
($18 yrs)
Community p..55 for MDD
(CIDI-SF)
þMADRS-S 15-30
1. iCBT 36 5 modules CBT
and BA
E-mail I, 10 wks BDI;
MADRS-S
Sweden
2. WL/online
discussion
group
39
Christensen
et al. (2004)
Adults
(18– 52 yrs)
Posted
questionnaire
K-10 $22 1. iCBT 182 1. 5 CBT
modules
Telephone calls by
lay interviewers
(all conditions)
I, 6 wks CES-D Australia
2. iPE 165 2. 5 PE modules
3. Attention
placebo
178 –
Clarke
et al. (2002)
Adults Via HMO Depressed,
nondepressed
members
1. iCBT 144 7 chapters CBT None I, NSD CES-D United States
2. CAU 155
Clarke
et al. (2005)
Adults Via HMO Depressed,
nondepressed
members
1. iCBT-a 75 7 chapters CBT None: iCBT-a:
postcard reminders;
iCBT-b: telephone
reminders
I, NSD CES-D United States
2. iCBT-b 80
3. CAU 100
Clarke
et al. (2009)
Young adults
(18– 24 yrs)
Via HMO Depressed,
nondepressed
members
1. iCBT 83 4 sections CBT None I, NSD PHQ-9 United States
2. CAU 77
Proudfoot
et al. (2004)
Adults
(18– 75 yrs)
Primary care GHQ-12 .4
þCIS-R .12
(depression or
anxiety)
1. cCBT 112 8 sessions cCBT 5 min help at
beginning, end of
each session
I, 8 weeks BDI United Kingdom
2. CAU 109
Ruwaard
et al. (2009)
Adults
($18 yrs)
Community BDI 10-29 1. iCBT 36 8 phases CBT,
BA
Asynchronous
contact
I; 11 wks BDI,
SCL-90-R– D
The Netherlands
2. WL 18
Selmi
et al. (1990)
Adults Community SCL-90-R .
65th percentile
þBDI $16
þmajor,
intermittent,
minor
depression
(RDC-SADS)
1. cCBT 12 6 sessions CBT Help at beginning,
end of each session
C; 6 wks BDI; HAM-D;
SCL-90-R– D
United States
VOL 38, NO 4, 2009 Internet-based and computerized treatments for adult depression 199
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

Table 1. Continued
Study
Target
population Recruitment Inclusion Condition NIntervention
Therapist
support
Internet/
computer Measure
a
Country
2. ftf-CBT 12
3. WL 12
Spek,
Nyklı
´cek
et al. (2007)
Older adults
($50yrs)
Community Subthreshold
depression
(EDS $12, no
MDD)
1. iCBT 102 8 weekly
modules
CBT
None I; 8 wks BDI The Netherlands
2. gCBT 99
3. WL 100
Van Straten
et al. (2008)
Adults Community Self-defined
depression or
anxiety
1. iPST 107 5 weekly
modules
PST
E-mail I; 5 wks CES-D;
MDI
The Netherlands
2. WL 106
Warmerdam
et al. (2008)
Adults Community CES-D $16 1. iCBT 88 1. 8 weekly
modules
CBT; 2.5 weekly
modules PST
E-mail I: group
1, 8 wks;
group
2, 5 wks
CES-D The Netherlands
2. iPST 87
3. WL 87
Wright
et al. (2005)
Adults
(18– 65 yrs)
Community MDD (SCID)
þBDI $14
1. cCBT þftf 15 9 sessions CBT 25-min ftf
sessions
þ25-min
cCBT
C; 8 wks BDI;
HAM-D
United States
2. ftf-only
CBT
15
3. WL 15
Note. BA, behavioural activation; BDI, Beck Depression Inventory; C, computerized treatment delivery; CAU, care-as-usual; CBT, cognitive behaviour therapy;
cCBT, computerized CBT; gCBT, group CBT; CES-D, Center for Epidemiological Studies– Depression scale; CIDI-SF, Composite International Diagnostic
Interview Short-Form; EDS, Edinburgh Depression Scale; ftf, face-to-face; GHQ, General Health Questionnaire; HAM-D, Hamilton Depression Rating Scale;
HMO, health maintenance organization; I, Internet treatment delivery; iCBT, Internet-based CBT; iPE, Internet-based psychoeducation; iPST, Internet-based
problem-solving therapy; K-10, Kessler –10; MADRS-S, Montgomery– Asberg Depression Rating Scale; MDD, major depressive disorder; MDI, Major Depression
Inventory; PE, psychoeducation; PHQ-9, Patient Health Questionnaire; PST, problem-solving therapy; RDC, Research Diagnostic Criteria; SADS, Schedule for
Affective Disorders; NSD, no standard duration; SCID, Structured Clinical Interview for DSM-IV; SCL-90-R, Symptom Checklist-90-R; SCL-90-R-D, SCL-90-R
Depression subscale; WL, wait-list.
a
Only the instruments that were used to calculate effect sizes are included.
200 Andersson and Cuijpers COGNITIVE BEHAVIOUR THERAPY
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

In our analyses, we included three studies in
which two psychological treatments were
compared with the same control group. This
means that multiple comparisons from these
three studies were included in the same
analysis. These multiple comparisons, how-
ever, are not independent of each other, which
may have resulted in an artificial reduction of
heterogeneity. Therefore, we conducted
another meta-analysis, in which we included
only one comparison per study (Table 2).
From the three studies with multiple compari-
sons, we first included only the comparison
with the largest effect size. As can be seen in
Table 2, these analyses did indicate that
heterogeneity increased somewhat in some
analyses (I
2
¼65.35), although the effect size
did not differ very much from the overall
analyses. Then we repeated these analyses and
included only the smallest effect size of the
three studies with multiple comparisons.
These analyses also resulted in increased
heterogeneity (I
2
¼64.86) and a comparable
effect size.
Neither the funnel plots nor Duval and
Tweedie’s trim and fill procedure indicated a
significant publication bias.
Subgroup analyses
We conducted several subgroup analyses
(Table 2). These included type of psychologi-
cal treatment (Internet-based vs. computer-
ized), type of control group (care-as-usual,
wait-list, other), content of psychological
treatment (CBT vs. other), whether the study
only included participants with depression or
also persons with anxiety, and whether there
was professional support during the therapy
(yes or no).
Table 2. Meta-analyses of studies examining the effects of computerized and Internet-based psychological
treatments for adult depression
Study N
comp
d95% CI ZI
2a
p
b
Overall effects
All studies 15 0.41 0.29– 0.54 6.47**** 57.49***
One ES per study (highest) 12 0.45 0.29– 0.61 5.39**** 65.35***
One ES per study (lowest) 12 0.43 0.27– 0.59 5.21**** 64.86***
Two possible outliers removed
c
13 0.37 0.26– 0.49 6.41**** 51.36**
Subgroup analyses
d
Type .107
Internet-based 12 0.37 0.24– 0.49 5.84**** 53.83**
Computerized 3 0.85 0.27– 1.43 2.89*** 61.18*
Control group .038
Care-as-usual 5 0.23 0.06– 0.40 2.68*** 46.34
Wait-list 7 0.56 0.37 –0.76 5.60**** 43.51
Other 3 0.45 0.21– 0.69 3.64**** 59.38*
Psychological treatment .923
CBT 12 0.42 0.26 –0.59 5.09**** 64.82***
Other 3 0.41 0.27– 0.56 5.51**** 0
Anxiety allowed .234
Only depression 13 0.38 0.25– 0.51 5.75**** 57.26***
Depression or anxiety 2 0.64 0.24– 1.04 3.13*** 42.73
Professional support .000
Support 8 0.61 0.45– 0.77 7.67**** 23.74
No professional support 7 0.25 0.14– 0.35 4.75**** 10.41
C or I psychotherapy vs. ftf
psychotherapy (all studies)
320.05 20.29–0.20 20.36 0
Note. CBT, cognitive behaviour therapy; C, computerized treatment delivery; CI, confidence interval; ES, effect
size; ftf, face to face; I, Internet treatment delivery; N
comp
, number of comparisons.
a
The pvalues indicate
whether the Qstatistic is significant (the I
2
statistics does not include a test of significance).
b
The pvalues
indicate whether the difference between the effect sizes in the subgroups is significant.
c
Selmi et al., 1990; Wright
et al., 2005.
d
All subgroup analyses were conducted with mixed-effects analyses. *p,.10. **p,.05.
***p,.01. ****p,.001.
VOL 38, NO 4, 2009 Internet-based and computerized treatments for adult depression 201
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

As can be seen in Table 2, the type of
control group was significantly associated
with the effect sizes ( p,.05). Studies with a
wait-list control had higher effect sizes than
care-as-usual and other control groups.
However, heterogeneity remained high in
these subgroups. We also found that studies
in which no professional support was given
had lower effect sizes than those in which
support was given ( p,.000). The heterogen-
eity in these two groups was low (I
2
,25%).
Because the study by Christensen, Griffiths,
and Jorm (2004) was complex to interpret
(there was some support by telephone but
no clear help in working through the treat-
ment), we repeated this subgroup analysis
without this study. The results were, however,
comparable (unguided: d¼0.18; 95%
CI ¼0.05–0.30, I
2
¼0; guided: d¼0.61,
95% CI ¼0.45 – 0.77, I
2
¼23.74; p
difference ¼.000).
Discussion
The aim of this meta-analysis was to
summarize the literature on Internet-based
and other computerized psychological treat-
ments for adult depression. We found an
overall effect size of d¼0.41, but that
estimate is probably not meaningful because
it hides the finding that interventions in which
support is provided to the participant are
more effective. Indeed, the computerized
interventions with support showed an average
between-group effect size of d¼0.61, whereas
the unsupported treatments had a much
smaller effect of d¼0.25. This is similar to
the findings by Spek, Cuijpers, et al. (2007),
who found that interventions without support
had an average effect size of d¼0.24, whereas
Internet interventions with support had a
large mean effect size of d¼1.0. Although the
meta-analyses overlap in terms of studies
included, we included more recent studies and
also other computerized treatments not
covered in the Spek et al. study (e.g. not
Internet-delivered). Overall, it appears that
computerized treatments with therapist sup-
port are much more effective than unsup-
ported treatments, and this has been
confirmed in open studies as well (e.g.
Christensen, Griffiths, Groves, & Korten,
2006). However, the concept of support is
not fully investigated in the literature, and we
cannot exclude the possibility that some forms
of support can be automated or that other
factors, such as having a clear deadline for
completion of a treatment with a scheduled
follow-up (e.g. a telephone interview), would
make less-supported treatments more effective
(Nordin, Carlbring, Cuijpers, & Andersson, in
press). Clearly, there is a need to investigate
Figure 1. Internet-based and other computerized treatments of depression compared to control groups:
standardized effect sizes.The letters A and B in the above figure refer to different subsamples within each study.
202 Andersson and Cuijpers COGNITIVE BEHAVIOUR THERAPY
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

the concept of support further and the role of
therapist factors in computerized treatments
(e.g. Almlo
¨v, Carlbring, Berger, Cuijpers, &
Andersson, 2009).
There were few studies available for the
contrast between Internet-based or other
computerized treatments and face-to-face
treatments. Here we found no difference,
which is in line with other studies in the field of
anxiety disorders (e.g. Carlbring et al., 2005;
Kiropoulos et al., 2008). There is a need to
further test whether computerized treatment,
and Internet-delivered treatment in particular,
can be as effective as face-to-face treatments in
depression. These studies need to be designed
as equivalence studies, and indeed in our
meta-analysis the finding of no difference
could be regarded as a promising sign of
equivalence between the treatment formats.
Although the effects of supported computer-
ized treatments in this meta-analysis are in line
with those of previous meta-analyses for
psychological treatments (e.g. Cuijpers, van
Straten, Andersson, et al., 2008), they are
somewhat lower. However, the effects are not
low when compared with the effects of
psychological treatments in primary care
(Cuijpers, van Straten, van Schaik, & Anders-
son, 2009). Several other findings are worth
commenting on. We did not find a significant
difference between computerized and Inter-
net-delivered treatments, but this could be due
to power problems. In line with findings of
other studies, the type of control group
influenced the effect size estimate, with lower
effects when treatment as usual was the
comparison group. This was expected because
no treatment is worse than some treatment.
We did not find any differences between CBT-
oriented and other forms of computerized
treatments. This comparison was unbalanced,
with most of the studies being done from a
CBT perspective, but we welcome more
studies on computerized treatments from a
non-CBT perspective, because it is not settled
whether other evidence-based psychological
treatments such as interpersonal psychother-
apy are possible to transfer to the computer
medium or the Internet. We also did not find
any effects of comorbid anxiety on the effect
size estimate, but this needs to be further
explored in future studies, because there are
very few studies on computer interventions for
depression in which other conditions than
depression have been assessed.
In this meta-analysis, we did not present
data on quality assessment. However, all
studies were checked using the four basic
criteria, as suggested in the Cochrane Hand-
book for Systematic Reviews of Interventions
(Higgins & Green, 2005): allocation to
conditions conducted by an independent
(third) party, blinding of assessors of out-
comes, completeness of follow-up data, and
adequacy of random allocation concealment
to respondents. Overall, the study quality was
not satisfactory, but studies are increasingly
following the CONSORT guidelines (e.g.
Boutron, Moher, Altman, Schulz, & Ravaud,
2008).
There are several possible challenges for
future research. First, Internet-delivered and
other computerized psychological interven-
tions are still not supported by a solid
database, and many studies have failed to
perform a proper diagnostic assessment
(Andersson & Cuijpers, 2008). There is a
need for accurate diagnostic procedures in
future trials. Second, most studies have been
done either in an academic setting with
participants recruited via advertisement or
by an epidemiological screening approach.
Only one of the studies in our meta-analysis
included patients recruited from primary care.
There is a need for effectiveness studies of
Internet and other computerized treatments
involving patients from psychiatric settings.
Third, studies on long-term effects of treat-
ment are largely lacking. There is a 1-year
follow-up of the trial by Christensen et al.
(2004), which showed some remaining benefits
(Mackinnon, Griffiths, & Christensen, 2008),
but more research is needed, in particular,
because help-seeking and health care use
might be affected as an effect of treatment
(Christensen, Leach, Barney, Mackinnon, &
Griffiths, 2006). Fourth, our meta-analysis did
not cover the issue of whom Internet and other
computerized treatments are suitable for. For
example, one study found an indication that
number of previous depression episodes was
related to worse treatment outcome (Anders-
son, Bergstro
¨m, Holla
¨ndare, Ekselius, &
Carlbring, 2004). It is also important to
investigate differential predictors of outcome
for different treatment formats (Spek, Nyklı
´-
cek, Cuijpers, & Pop, 2008).
VOL 38, NO 4, 2009 Internet-based and computerized treatments for adult depression 203
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

This meta-analytic review has some limi-
tations. First, we included very heterogeneous
treatments and samples. This can be seen as an
advantage for a meta-analysis because we then
could investigate differences between studies.
However, because relatively few studies could
be located, we were underpowered to detect
effects for some contrasts. A second limitation
has to do with the selection of adult samples
only. Our impression is that the available
studies on adolescents do not alter our main
finding that support may be needed to achieve
good outcomes (e.g. O’Kearney, Gibson,
Christensen, & Griffiths, 2006). A third
limitation concerns the methodological qual-
ity of the studies that we did not fully report.
For example, we did not include analyses on
the drop-out rate for different treatments,
which is a known problem in Internet
interventions.
Despite the limitations of this meta-anal-
ysis, we believe there is emerging evidence that
Internet and other computerized intervention
can be helpful in reducing symptoms of
depression.
Acknowledgments
The preparation of this study was supported in
part by a grant from the Swedish Research
Council.
References
References marked with an asterisk indicate studies
included in the meta-analysis..
Almlo
¨v, J., Carlbring, P., Berger, T., Cuijpers, P., &
Andersson, G. (2009). Therapist factors in
Internet-delivered CBT for major depressive
disorder. Cognitive Behaviour Therapy,38(4),
247–254.
Andersson, G. (2006). Internet based cognitive
behavioral self-help for depression. Expert
Review of Neurotherapeutics,6, 1637–1642.
*Andersson, G., Bergstro
¨m, J., Holla
¨ndare, F.,
Carlbring, P., Kaldo, V., & Ekselius, L. (2005).
Internet-based self-help for depression:
A randomised controlled trial. British Journal
of Psychiatry,187, 456–461.
Andersson, G., Bergstro
¨m, J., Holla
¨ndare, F.,
Ekselius, L., & Carlbring, P. (2004). Delivering
CBT for mild to moderate depression via the
Internet. Predicting outcome at 6-months
follow-up. Verhaltenstherapie,14, 185–189.
Andersson, G., & Cuijpers, P. (2008). Pros and cons
of online cognitive-behavioural therapy. British
Journal of Psychiatry,193, 270–271.
Barak, A., Hen, L., Boniel-Nissim, M., & Shapira,
N. (2008). A comprehensive review and a meta-
analysis of the effectiveness of Internet-based
psychotherapeutic interventions. Journal of
Technology in Human Services,26, 109–160.
Boutron, I., Moher, D., Altman, D. G., Schulz,
K. F., & Ravaud, P. (2008). Extending the
CONSORT statement to randomized trials of
nonpharmacologic treatment: Explanation and
elaboration. Annals of Internal Medicine,148,
295–309.
Carlbring, P., Nilsson-Ihrfelt, E., Waara, J.,
Kollenstam, C., Buhrman, M., Kaldo, V., et al.
(2005). Treatment of panic disorder: Live
therapy vs. self-help via Internet. Behaviour
Research and Therapy,43, 1321–1333.
Christensen, H., Griffiths, K., Groves, C., &
Korten, A. (2006). Free range users and one
hit wonders: Community users of an Internet-
based cognitive behaviour therapy program.
Australian and New Zealand Journal of Psy-
chiatry,40, 59–62.
*Christensen, H., Griffiths, K. M., & Jorm, A.
(2004). Delivering interventions for depression
by using the Internet: Randomised controlled
trial. British Medical Journal,328, 265–268.
Christensen, H., Griffiths, K.M., Mackinnon, A.J.,
& Brittliffe, K. (2006). Online randomized trial
of brief and full cognitive behaviour therapy for
depression. Psychological Medicine,36,
1737–1746.
Christensen, H., Leach, L. S., Barney, L., Mack-
innon, A. J., & Griffiths, K. M. (2006). The
effect of web based depression interventions on
self reported help seeking: Randomised con-
trolled trial. BMC Psychiatry,6, 13.
*Clarke, G., Eubanks, D., Reid, E., Kelleher, C.,
O’Connor, E., DeBar, L. L., et al. (2005).
Overcoming depression on the Internet (ODIN)
(2): A randomized trial of a self-help depression
skills intervention program with reminders.
Journal of Medical Internet Research,7, e16.
*Clarke, G., Kelleher, C., Hornbrook, M., DeBar,
L., Dickerson, J., & Gullion, C. (2009).
Randomized effectiveness trial of an Internet,
pure self-help, cognitive behavioral intervention
for depressive symptoms in young adults.
Cognitive Behaviour Therapy,38(4), 222–234.
*Clarke, G., Reid, E., Eubanks, D., O’Connor, E.,
DeBar, L. L., Kelleher, C., et al. (2002).
Overcoming depression on the Internet
(ODIN): A randomized controlled trial of an
Internet depression skills intervention program.
Journal of Medical Internet Research,4, e14.
Cohen, J. (1988). Statistical power analysis for the
behavioral sciences (2nd ed.). Hillsdale, NJ:
Erlbaum.
Cuijpers, P., Marks, I., van Straten, A.-M.,
Cavanagh, K., Gega, L., & Andersson, G.
(2009). Computer-aided psychotherapy for
anxiety disorders: A meta-analytic review.
Cognitive Behaviour Therapy,38, 66–82.
Cuijpers, P., van Straten, A., Andersson, G., & van
Oppen, P. (2008). Psychotherapy for depression
in adults: A meta-analysis of comparative
204 Andersson and Cuijpers COGNITIVE BEHAVIOUR THERAPY
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010

outcome studies. Journal of Consulting and
Clinical Psychology,76, 909–922.
Cuijpers, P., van Straten, A., van Schaik, A., &
Andersson, G. (2009). Psychological treatment
of depression in primary care: A meta-analysis.
British Journal of General Practice,59, e51–60.
Cuijpers, P., van Straten, A., Warmerdam, L., &
Andersson, G. (2008). Psychological treatment
of depression: A meta-analytic database of
randomized studies. BMC Psychiatry, 8, 36.
Duval, S., & Tweedie, R. (2000). Trim and fill: A
simple funnel-plot-based method of testing and
adjusting for publication bias in meta-analysis.
Biometrics,56, 455–463.
Ebmeier, K. P., Donaghey, C., & Steele, J.D.
(2006). Recent developments and current con-
troversies in depression. Lancet,367, 153 –167.
Griffiths, K. M., & Christensen, H. (2006). Review
of randomised controlled trials of Internet
interventions for mental disorders and related
conditions. Clinical Psychologist,10, 16–29.
Higgins, J. P. T., & Green, S. (2005). Cochrane
handbook for systematic reviews of interven-
tions 4.2.5. In The Cochrane Library, Issue 3.
Chichester, UK: Wiley.
Higgins, J. P., Thompson, S. G., & Deeks, J. J.
(2003). Measuring inconsistency in meta-ana-
lyses. British Medical Journal,327, 557–560.
Kiropoulos, L. A., Klein, B., Austin, D. W., Gilson,
K., Pier, C., Mitchell, J., et al. (2008). Is
Internet-based CBT for panic disorder and
agoraphobia as effective as face-to-face CBT?
Journal of Anxiety Disorders,22, 1273–1284.
Mackinnon, A. J., Griffiths, K. M., & Christensen,
H. (2008). Comparative randomised trial of on-
line cognitive behaviour therapy and an
information website for depression: 12 month
outcomes. British Journal of Psychiatry,192,
130–134.
Marks, I. M., Cavanagh, K., & Gega, L. (2007).
Hands-on help. Maudsley Monograph no. 49.
Hove, UK: Psychology Press.
McDermut, W., Miller, I. W., & Brown, R. A.
(2001). The efficacy of group psychotherapy for
depression: A meta-analysis and review of the
empirical research. Clinical Psychology: Science
and Practice,8, 98–116.
Nordin, S., Carlbring, P., Cuijpers, P., & Anders-
son, G. (in press). Expanding the limits of
bibliotherapy for panic disorder. Randomized
trial of self-help without support but with a
clear deadline. Behavior Therapy.
O’Kearney, R., Gibson, M., Christensen, H., &
Griffiths, K. M. (2006). Effects of a cognitive-
behavioural Internet program on depression,
vulnerability to depression and stigma in
adolescent males: A school-based controlled
trial. Cognitive Behaviour Therapy,35, 43–54.
*Proudfoot, J., Ryden, C., Everitt, B., Shapiro,
D. A., Goldberg, D., Mann, A., et al. (2004).
Clinical efficacy of computerised cognitive-
behavioural therapy for anxiety and depression
in primary care: Randomised controlled trial.
British Journal of Psychiatry,185, 46–54.
Reger, M. A., & Gahm, G. A. (2009). A meta-
analysis of the effects of Internet- and compu-
ter-based cognitive-behavioral treatments for
anxiety. Journal of Clinical Psychology,65,
53–75.
*Ruwaard, J., Schrieken, B., Schrijver, M., Broek-
steeg, J., Dekker, J., Vermeulen, H., et al.
(2009). Standardized web-based CBT of mild to
moderate depression: A randomized controlled
trial with a long-term follow-up. Cognitive
Behaviour Therapy,38(4), 206–221.
*Selmi, P. M., Klein, M. H., Greist, J. H., Sorrell, S.
P.,&Erdman,H.P.(1990).Computer-
administrated cognitive behavioral therapy for
depression. American Journal of Psychiatry,
147, 51–56.
Simon, G. E., Ludman, E. J., Tutty, S., Oper-
skalski, B., & Von Korff, M. (2004). Telephone
psychotherapy and telephone care management
for primary care patients starting antidepress-
ant treatment: A randomized controlled trial.
Journal of the American Medical Association,
292, 935–942.
Spek, V., Cuijpers, P., Nyklı
´cek, I., Riper, H.,
Keyzer, J., & Pop, V. (2007). Internet-based
cognitive behaviour therapy for symptoms of
depression and anxiety: A meta-analysis.
Psychological Medicine,37, 319–328.
Spek, V., Nyklı
´cek, I., Cuijpers, P., & Pop, V.
(2008). Predictors of outcome of group and
Internet-based cognitive behavior therapy.
Journal of Affective Disorders,105, 137–145.
*Spek, V., Nyklı
´cek, I., Smits, N., Cuijpers, P.,
Riper, H., Keyzer, J., et al. (2007). Internet-
based cognitive behavioural therapy for sub-
threshold depression in people over 50 years
old: A randomized controlled clinical trial.
Psychological Medicine,37, 1797–1806.
*van Straten, A., Cuijpers, P., & Smits, N. (2008).
Effectiveness of a web-based self-help interven-
tion for symptoms of depression, anxiety, and
stress: Randomized controlled trial. Journal of
Medical Internet Research,10(1), e7.
*Warmerdam, L., van Straten, A., Twisk, J., Riper,
H., & Cuijpers, P. (2008). Internet-based
treatment for adults with depressive symptoms:
Randomized controlled trial. Journal of Medical
Internet Research,10, e44.
*Wright, J. H., Wright, A. S., Albano, A. M.,
Basco, M. R., Goldsmith, L. J., Raffield, T.,
et al. (2005). Computer-assisted cognitive
therapy for depression: Maintaining efficacy
while reducing therapist time. American Journal
of Psychiatry,162, 1158–1164.
VOL 38, NO 4, 2009 Internet-based and computerized treatments for adult depression 205
Downloaded By: [Vrije Universiteit, Library] At: 11:10 26 November 2010