Review

A systematic review of minimal-contact psychological treatments for symptommanagement in Irritable Bowel Syndrome

Rosanna Pajak a, Jeffrey Lackner b, Sunjeev K. Kamboj a,⁎a Research Department of Clinical, Educational and Health Psychology, University College London, London, UKb Department of Medicine, University at Buffalo School of Medicine, SUNY at Buffalo, DK Miller Bldg, Buffalo, NY, United States

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

Article history:

Received 4 January 2013

Received in revised form 13 May 2013

Accepted 14 May 2013

Keywords:

Irritable Bowel Syndrome

Minimal-contact therapy

Psychological treatment

Cognitive behavioural therapy

Hypnosis

Self-help

Objective: Psychological treatments are effective in alleviating symptoms of IBS but are not widely available.

The need for wider dissemination of treatments has encouraged the development of ‘minimal-contact’

therapies requiring fewer resources than existing psychological treatments which rely on face-to-face contact.

Method: Using comprehensive search terms, the Embase, Medline and PsychInfo databases (all years) were

searched.

Results: Twelve studies – nine RCTs and three non-controlled preliminary studies – meeting inclusion criteria

were reviewed and assessed for quality using objective criteria. Apart from one study of expressive writing, all

interventions were based on cognitive (and/or) behavioural principles or hypnosis and tended to be adaptations

of existing therapist-led interventions. Compared to control conditions, minimal-contact interventions were

efficacious, the majority of studies showing statistically significant improvements by the end of treatment. For

cognitive-behaviour-therapy-based interventions effects sizes were large. The two studies that compared

minimal-contact with therapist-delivered interventions broadly suggest comparable outcomes between these

modalities.

Conclusions: Minimal-contact cognitive–behavioural interventions show promise in the treatment of IBS.

Because of the lower quality of studies of hypnosis and those involving interventions delivered entirely remotely,

further support is needed before such approaches can be recommended for widespread use. More generally,

future research should use representative samples, active control conditions, and intention to treat analysis.

Nonetheless, existing high quality studies suggest that minimal-contact therapies may be a safe, effective

means of achieving scaleability of psychological treatments for IBS.

© 2013 Elsevier Inc. All rights reserved.

Introduction

Irritable Bowel Syndrome (IBS) is a chronic functional gastrointes-

tinal disorder characterised by recurrent episodes of abdominal pain

or discomfort, bloating and altered bowel habits in the absence of

detectible organic disease [1]. Psychiatric comorbidity is common [2].

Up to 11% of the populations of most countries are affected by IBS

symptoms [3]. Of these at least 30% seek medical advice [4]. Since IBS

is a chronic condition for which there is no ‘cure’ [1], this level of

healthcare-need represents a substantial burden on services [5,6]

which is unlikely to be met with existing models of service delivery.

Recent treatment guidance emphasises the need for clinically- and

cost-effective management in primary care through lifestyle/dietary

advice and pharmacotherapy and/or psychological interventions

where appropriate [7]. At primary care level such interventions need

to be acceptable to patients and ideally, easily implemented and highly

scalable.

Psychological approaches to IBS symptom management

In line with the aim of reducing impairment caused by a chronic

condition, self-management approaches are recommended for IBS

patients [7].While some of these emphasize the provision of supportive

counselling or encourage life-style changes, others are more formally

grounded in psychological theories of behaviour change [8]. Given the

high levels of mood and anxiety disorder in IBS [2], treatments that

target psychological distress as well as IBS-specific symptoms of pain

and gut dysfunctionmay be especially appropriate for achieving positive

outcomes in the range of symptoms present in IBS.

Psychological treatments emphasise active coping through system-

atic changes in behaviour and thoughts which are believed to exert a

top-down influence on gastrointestinal functioning [8]. The aims of

treatment include refocusing of, testing IBS-related beliefs and changing

themeaning of symptoms and related sources of distress that contribute

to symptom expression [1,7]. The predominant mode of delivery of

psychological treatments for IBS is the traditional ‘face-to-face’ encounter

involving a variety of therapeutic modalities, including cognitive and/or

behavioural, interpersonal therapy or hypnosis [7,8]. Such approaches

Journal of Psychosomatic Research 75 (2013) 103–112

⁎ Corresponding author. Tel.: +44 20 7679 1958; fax: +44 20 7916 1989.

E-mail address: sunjeev.kamboj@ucl.ac.uk (S.K. Kamboj).

0022-3999/$ – see front matter © 2013 Elsevier Inc. All rights reserved.

http://dx.doi.org/10.1016/j.jpsychores.2013.05.007

Contents lists available at ScienceDirect

Journal of Psychosomatic Research

have considerable advantages over purely self-guided treatments

(encouraging, for example, ongoing symptom monitoring and careful

shaping of between-session self-help behaviour) and may remain

the most desirable for those with the most severe symptoms and

co-morbidities. However, as Wilson and Zandberg [9] note, such a

resource-intensive model of service delivery is, by a considerable

margin, incapable of providing sufficient access to evidence-based

treatments to all those who need them.

Alternative treatment models: minimal-contact psychological treatments

for IBS

Minimal-contact psychological treatments [10–12] place a signifi-

cant emphasis on self-management of symptoms. Contact with health

care professionals varies but is generally limited to a small number of

face-to-face sessions (or possibly, none at all), supplemented or

replaced by computer-assisted therapy, telephone and/or online support.

When optimised, thesemodes of treatment delivery are likely to produce

considerable efficiency savings and allow for wider dissemination, espe-

cially to under-served communities.

The distinction between non-psychologically guided self-

management approaches and those based on formal psychological

models is an important one. For example, while self-guided approaches

lacking a psychological basis – for example stress management and/or

dietary advice – may be efficacious, psychological-based interventions

can ‘add value’ (e.g. [13]). Minimal-contact therapies often form the

first rung within ‘stepped care’models of psychological service delivery

[14] which predominate in the UK and Australia [15]. Stepped care

involves tailoring the ‘intensity’ of the treatment to the presenting

disorder and its severity, with the aim that milder presentations (those

causing relatively mild levels of distress and impairment) with limited

co-morbidity are treated with fewer face-to-face sessions, guided by an

associate-level practitioner (i.e. onewho is not fully accredited in Clinical

Psychology or allied professions). Other important considerations

include choice and convenience for the client: the flexibility engendered

by these approaches mean that clients can often engage in treatment

without significant disruption of their work or other daily activities.

Even 50–60 minute long telephone-based therapy sessions intended to

closely mimic/match face-to-face sessions in content and interaction

with a therapist potentially entail very significant opportunity cost

savings for the client (e.g. in time spent travelling to the therapist’s

office/absence from work).

Recent reviews outlining evidence forminimal-contact psychological

treatments [16–18] conclude that for milder anxiety disorders and

depression such treatments can be as effective as face-to-face treat-

ments. Other reviews [19,20] and meta-analyses [21,22] suggest that

even purely self-administered psychological treatments are effective,

particularly for anxiety problems.

Minimal-contact psychological interventionsmay also be efficacious

for a variety of physical symptoms and their psychological concomi-

tants including tinnitus, [23]; headaches [24]; insomnia [25]; chronic

pain [26] and obesity [27]. Cuijpers et al. [28] reviewed the literature

on internet-delivered treatments for health problems and concluded

that on a variety of outcomes for headaches and chronic pain the effects

for internet-based cognitive behavioural therapy (CBT) were compara-

ble to face-to-face treatments. Even substance use disorders, which

have traditionally relied on case-management ormultimodal treatments

(e.g. pharmacological, and individual/group psychotherapy) are ame-

nable to minimal-contact interventions, which show some promising

outcomes [12].

Given these findings and the fact that existing empirically-

supported face-to-face psychological treatments for IBS already tend

to emphasise self-management strategies [29], it is unsurprising

that minimal-contact treatments have also recently been developed

for IBS. A description of these recent developments will form the

remainder of this paper. In particular, we will systematically review

studies of minimal-contact therapies (published up until 2011)

which are linked by the relatively limited number of direct therapist

contacts compared to predominantly therapist-administered psycho-

logical treatments.

Influenced by Glasgow and Rosen [30] and Newman et al. [12], in

our definition of ‘minimal-contact’ we included studies involving:

(1) pure or predominant self help (with therapist contact for assessment

at most), (2) guided self help, in which limited and/or brief therapist

contact occurred for the purposes of clarification of self-management

strategies/homework assignments, and (3) reduced/remote contact

treatments, in which the ‘dose’ of face-to-face contact is substantially

reduced compared to predominantly therapist-administered treatments

(reviewed in Lackner et al. [8]). The latter category includes studies

whose purpose was to determine whether efficacy is retained through

substitution of direct (face-to-face) contact with remote contact

[11,31]. In addition to service-level efficiency savings, this latter category

of minimal-contact intervention may have high acceptability for clients

for the reasons outlined above.

This is the first review of its kind to specifically focus on minimal-

contact psychological interventions (as defined above) for IBS. A

comprehensive review of psychological treatments by Lackner et al.

[8] concluded that psychological treatments as a whole were more

effective in reducing symptom severity than a pooled group of control

conditions. A more recent review examining antidepressants and psy-

chological therapies similarly concluded that psychological treatments

are efficacious compared to control treatments, although concerns

were raised about the quality of studies [32]. Because these reviews

predominantly focused purely on therapist-guided face-to-face inter-

ventions, and taking into account the recent therapeutic and technolog-

ical advances, it is timely to examine the efficacy of minimal-contact

therapies involving a range of psychological therapy modalities in IBS

sufferers. At this relatively early stage of evaluation of these therapies,

it is appropriate to examine both randomised controlled trials (RCTs)

and non-RCT designs.

Method

Search methods for identification of studies

A systematic computer-assisted search of Embase, Medline and

PsychInfo databases (all years) was performed using the search terms

below. The fields “title” and “abstract” were used as limits. No other

limits or filters were used.

Title/abstract search

(IBS OR irritable bowel syndrome OR irritable bowel OR gastroin-

testinal OR bowel disorder OR abdominal OR gastric OR functional gas-

trointestinal disorders OR functional digestive disorders OR functional

GI OR somatisation disorder) AND (CBT OR cognitive behavio(u)ral

therapy OR psychological OR psychology OR psychologic OR mindful-

ness OR psychosocial OR cognitive therapy OR behavio(u)r therapy

OR psychotherapy OR psychoeducational OR psychological treatment

OR counselling OR acceptance OR psychological intervention ORmental

health interventionOR expressivewriting OR intervention) AND (inter-

net self help OR self OR self-help OR self-management OR psycho-

educational OR website OR online OR telephone OR support group OR

psychoeducational OR group cognitive therapy OR internet delivered

OR stepped care OR group cognitive behavio(u)ral therapy OR

internet-therapy).

Titles and/or abstracts of all studies identified by the search strategy

(n = 2334) were screened for relevance. After screening out on the

basis of relevance and removing duplicates, full text articles were

obtained for all potentially eligible studies. Hand searching was

conducted on eligible studies for additional potential papers, and corre-

sponding authors of key research groups were contacted regarding

studies that were in press.

104 R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

Criteria for inclusion of studies for this review

Inclusion criteria were deliberately broad. Thus the review includes

both efficacy studies (randomised controlled trials) as well as non-

randomised studies involving pre-post comparisons. Studies were

included on the basis of the following criteria. (1) Studies primarily

aimed to test treatment effects of an intervention for IBS sufferers

based on a recognised psychological/psychotherapeutic theory. This

included interventions based on (i.e. adapted from) existing bona fide

therapy models as well as those testing new therapeutic strategies

which might normally be integrated into such treatments, e.g. expres-

sivewriting. This excluded those interventionswhich lacked an explicit,

theory-guided focus on emotional, cognitive, behavioural, or interper-

sonal factors (e.g. education, information-giving, peer-support, non-

specific ‘self-help’). (2) Studies presented quantitative data. (3) Studies

involved minimal-contact with some level of clinician involvement, as

outlined above [30]. For the purposes of this review we arbitrarily

defined minimal-contact as involving ≤4 face-to-face sessions, which

is half themodal number of sessions used in the studies of predominantly

therapist administered treatments reviewed by Lackner et al. [8]).

(4) Studies involved adults (≥18 years old) and (5) were published in

English.

Quality assessment

The methodological quality of the included studies was assessed

using a modified version of the Cochrane Collaboration Depression

and Anxiety Neurosis Review Group’s (CCDAN) coding scheme [69]

which has previously been used to evaluate the efficacy and method-

ological quality of psychological interventions for IBS [8]. Although

other scales were reviewed, this remained the most comprehensive

(assessing patient selection, interventions, outcome assessment, data

presentation and statistical analysis) and relevant (in terms of use in

IBS and applicability to minimal-contact treatments). Furthermore,

the modified CCDAN scale contains items recommended by the Rome

committee on design of treatment studies for functional gastrointestinal

disorders [70]. All items on the scale were retained although anchors for

scoring were modified slightly for two items [item 8 on manualisation

and item24 on allegiance] to reflect fundamental differences in research

methodologies employed in minimal-contact therapies compared to

face-to-face therapies (c.f. [8]). In particular, given the purpose of

manualisation (item 8) is to standardise treatment delivery (through

competence and adherence), and therapist contact is by definition

‘minimal’ in predominantly self-help therapies, these treatments were

generally considered highly standardised in their delivery. As such, 2

points were given on the manualisation item to studies which took

steps to standardise therapist contact where it existed, or tested inter-

ventions devoid of therapist contact where self-help instruction was

delivered in written or audio form. For item 24, 1 point was given for

reporting of information that would inform the reader of the authors’

allegiance to the minimal-contact therapy; specifically whether they

had developed the intervention and/or whether it was the only active

treatment tested [71].

Failure to report specific methodological details resulted in a

lower overall quality score.

All included studies (n = 12) were assessed in terms of method-

ological quality independently by two researchers after consulting

with psychotherapy research experts and clarifying how anchors

were to be used. To increase the consistency with which the quality

criteria were applied the scale was first piloted by the first and senior

authors (RP and SK) on two of the studies [47,68]. Following discussion

of items which had the greatest (potential for) divergent scoring and

further consultation with experts, the remaining ten papers were

rated independently by the same two researchers. Discrepancies were

resolved through discussion to reach a consensus score on each item

for each study. Prior to agreement, the independent quality ratings

(including those from the two studies used for pilot scoring) correlated

well between the two raters [r(12) = .830, p b 0.01].

Statistical analysis

Effect size calculations were based on the mean pre-post change in

the treatment group minus the mean pre-post change in the control

group divided by the pooled pre-test standard deviation and multi-

plied by a bias correction term ([92]; Eq. (8)). Such an approach has

been found to produce the most precise and unbiased estimates of

effect size in between groups, repeated measures designs [92]. Differ-

ences between effect sizes reported in the publications that we review

and those presented here are likely due to the use of alternative

methods of calculating effect size (such as using only the post-test

means in reviewed studies). Where insufficient data was provided in

the paper or through contact with authors, studies were not included

in the effect size analysis. We restricted our analysis to the most com-

monly reported outcomes, namely IBS-specific symptom severity (IBS

Severity Scoring System, IBS-SSS; IBS version of the Gastrointestinal

Symptom Rating Scale, GSRS-IBS and GI-specific symptom ratings)

and quality of life (IBS-QoL) scores, where these are correctly reported.

It was thus possible to calculate the effect size for symptom severity for

nine studies [11,13,31,47,53,55,56,60,68] and quality of life for seven

[11,13,31,53,56,60,68].

Results

Search results

Initially a total n = 2334 hits were identified from the three databases (Fig. 1),

of which n = 36 papers were identified that potentially met inclusion criteria

[11,13,31,33–65], after exclusions based on lack of relevance (n = 2264) and duplicates

(n = 34). Two additional papers were found by hand-searching reference lists of

retrieved papers and related reviews [66,67]. A further paper [68] was obtained by

contacting corresponding authors of key research groups.

After reviewing and applying the exclusion criteria to the full manuscripts of these

39 studies, 27were removed. Nineteen of the 27 interventions exceeded the four face-to-

face contacts which formed a basis for our minimal-contact definition (excluded studies

had a modal number of eight sessions) or interventions that were not based on

established psychological/psychotherapeutic theory. Six papers presented the same data

from a former trial and two were only reporting protocols.

This left a final total of 12 studies (Fig. 1).

Quality assessment

Table 1 presents quality scores from the modified-CCDAN [8]. The first column

contains the final summed score for each study, with studies arranged in order of

increasing quality: higher scores indicate the presence of a larger number of important

methodological features within a study. The bottom row shows totals – across studies –

for each assessed methodological element of the studies (i.e. each item from the

modified-CCDAN). Higher scores per item indicate the tendencies for the corresponding

methodological feature to be present across studies (scores increase rightward). A maxi-

mum score of 24 indicates the presence of a highly rated design aspect across all studies

(outcome measures and treatment clearly described: items 17 and 7 respectively).

Across the studies those elements on the right of the Table 1 (shaded) with

scores ≥ 18 (i.e. ≥ 75% of the maximum score for all studies) are taken to represent

methodological elements which are consistently good across studies. There are 14

such elements (Q5, randomised allocation; Q21, appropriate statistical analysis; Q11, ex-

clusion criteria; Q18, information on comparability; Q1, clear objectives/outcomes a

priori; Q12, sample demographics; Q29, consecutive subjects; Q22, conclusions justified;

Q10, inclusion criteria; Q20, presentation of results; Q8, manualisation; Q17, outcome

measures described; and Q7, treatments clearly described; Q26, co-interventions avoided

or equal) in the studies reviewed here. This compares with only five methodological

elements scoring ≥75% across the studies reviewed by Lackner et al. [8].

Descriptions of studies

Table 2 presents a summary of the key characteristics of the twelve included studies.

Study designs

Eight out of the twelve studies were randomised controlled trials (RCTs)

[11,13,31,47,53,55,56,68]. Of these, five indicated that they were registered with an

appropriate oversight body [13,31,55,56,68]. Other studies were non-controlled and

described variously as ‘pilot’, ‘feasibility’ or ‘preliminary’ studies. These used either

105R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

between-groups comparisons based on ‘compliant’ and ‘non-compliant’ participants

[44] or comparisons between active intervention and waitlist or treatment-as-usual

controls [60,66,67].

While most of RCTs reviewed here effectively employed procedures for ensuring

the treatment allocation code was concealed from those involved in recruitment, all

studies relied on participant self-report and therefore assessments could not be said

to be blind. The difficulty of masking the intention of psychological interventions

from patients is widely acknowledged [72]. Under these circumstances, credibility

assessment is a potentially valuable approach to assessing expectancy effects, although

participants from only two studies [11,13] rated credibility of their allocated treatment.

Sample characteristics

Sample sizes ranged from 28 [60] to 195 [13]. Participants’ mean age across the 12

studies was 40 years old. The percentage of female participants ranged from 71% to

91%, with an average gender ratio across the studies of 4:1 (female:male), consistent

with gender differences in community prevalence of IBS [73].

Recruitment of participants was based on a variety of methods. The majority of the

studies recruited treatment-seeking individuals, who self-referred in response to

adverts via specialist GI services and research websites [13,31,44,47,53,55,67]. Some

obtained additional participants through specialist GI services or primary care services

[11,56,60]. Just two studies relied purely on professional referrals: Forbes et al. [66],

who recruited from participants being considered for hypnotherapy within a specialist

GI service and Ljótsson et al. [68], whose participants originated from a specialist GI

service.

With the exception of Hunt et al. [47], Moss-Morris et al. [55] and Oerlemans et al.

[56], studies referenced either Rome II or Rome III criteria. Moss-Morris et al. [55] used

participants meeting Rome I and/or Rome II criteria (their participants included those

diagnosed in a previous study which used the older criteria) and Oerlemans et al. [56]

used either Rome III or International Classification of Primary Care (IPCP:D93) criteria.

Forbes et al. [66] used only Rome I, whilst Hunt et al. [47] relied on participant reporting

of IBS diagnosis by a medical professional without explicit reference to clinical criteria.

The assessment procedure for determining diagnosis varied across and within

studies (especially in studies that relied on recruitment from more than one source).

Those studies that examined pure self-help relied most heavily on self-report question-

naire data rather than formal research-diagnostic assessment [44,47]. Halpert et al.

[44] used Rome III criteria only for screening (rather than formal assessment) of partic-

ipants on an internet site. Ljótsson et al. [53] and Ljótsson et al. [13] checked Rome II

criteria were met via a telephone interview, whilst Ljótsson et al.’s [68] recruitment

from a gastroenterological clinic enabled specialist assessment by gastroenterologists

as part of the research protocol. Some (33%) of the participants in the study by Sanders

et al. [60] were recruited through their gastroenterologist who confirmed the diagnosis

using Rome II criteria; the remaining participants were diagnosed by their family phy-

sician but assessed further by the research team using Rome II criteria. The participants

in Palsson et al. [67] had their diagnosis confirmed by a physician using Rome II

criteria, although this assessment was not part of the research protocol. Diagnostic

assessment varied depending on the source of the participant in Oerlemans et al. [56]. If

referral was from the family physician, this referrer was asked to confirm International

Classification of Primary Care (IPCP:D93) criteria. Alternatively, diagnosis was assigned

through a review by two psychologists of a self-report questionnaire (using Rome III

criteria) for those responding to advertisements via the Dutch IBS patient association.

As might be expected, more comprehensive assessment for participant inclusion

tended to occur in studies involving some degree of face-to-face contact. For example

participants in Jarrett et al. [31] undertook four weeks of symptom monitoring to

determine whether IBS symptoms (both abdominal pain/discomfort and diarrhoea or

constipation) occurred on at least 25% of the days. Symptoms were then assessed by

a gastroenterologist as part of the research protocol prior to randomization. Inclusion

did not depend on symptom severity in Jarrett et al. [31], although Lackner et al. [11]

used specialist assessment by gastroenterologist as part of the research protocol for

selection of participants with at least moderately-severe symptoms.

Embase

1180

Psych Info

315

Medline

839

31 18 21

36

70

39

Excluded=34 duplicates

Contactingauthors = 1

Hand searching = 2

Included

12

Excluded = 27

Interventions involved more

sessions than our minimal

contact definition (i.e. ≥ 4

face-to-face contacts)

Interventions not based on

established

psychological/psychotherape-

utic theory.

same data from a former trial

only reporting protocols

Excluded = 2264

Lack of relevance

based on

title/abstract

•

•

••

Fig. 1. Flowchart of literature search process.

106 R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

A general feature of trials of hypnosis treatments is that they often include patients

who are described as ‘refractory’ or ‘severe.’ (e.g. [74,75]). While participants in Forbes

et al. [66] were described as “previously unsuccessfully treated” no reference is made

to previous treatment attempts in Palsson et al. [67]. None of the non-hypnosis studies

make reference to previous treatment failure/successes.

With the exception of Hunt et al., [47] –who did not apply any exclusion criteria to

their online sample – exclusion criteria were generally clearly reported in the included

studies. In general exclusions were those: with serious mental or physical health

concerns [11,13,31,44,53,55,56,60,67,68]; receiving psychological or psychiatric treat-

ments [31,56,60,67]; with additional GI pathology, such as inflammatory bowel disease

[11,13,31,53,55,60,66,68]. Only Oerlemans et al. [56] excluded those receiving pharma-

ceutical interventions. Information regarding concurrent drug use was only infrequently

and partially reported [11,66–68]. For two studies the number of participants excluded

on the basis of the stated exclusion criteria remained unclear [66,67].

On the whole, sample demographics were well described across the included studies

as was information regarding the comparability of their treatment and control samples

and any adjustments that were carried out. One study [66] omitted information on

demographic characteristics and one [60] did not provide adequate information on com-

parability of groups.

Interventions

Nine studies examined CBT. Six examined ‘traditional’ CBT, based on the idea that

psychological distress is maintained through interactions between beliefs, affect,

behaviour and physiological symptoms [11,31,47,55,56,60]. Three studies (all originating

from the same team) tested a more recent CBT approach, emphasising mindfulness and

acceptance alongside exposure [13,53,68]. Two studies examined hypnosis [66,67] and

one, the effects of expressive writing [44].

Most of the studies compared a minimal-contact intervention to a waitlist

[11,44,47,53,60,68] or treatment as usual (TAU) control [31,55,56,67]. In some studies, the

intervention was supplemented with a closed online forum [53,68], or regular symptom

monitoring [11,47]. Only Moss-Morris et al. [55] actually explored the level of TAU usage

by participants during the trial period to ensure parity between groups in TAU usage.

Four studies used an active control. Ljótsson et al. [13] used an internet-delivered

stress-management intervention, developed specially as a credible control which

ensured minimal overlap with the active intervention. Three studies addressed the

question: ‘can a significant amount of face-to-face treatment contact be replaced by a

more resource-efficient way of engaging clients?’ [11,31,66]. The active control used

in these studies was the ‘standard’ intervention involving all the active ingredients of

the experimental intervention but a larger number of face-to-face sessions.

CBT studies varied considerably in terms of amount of contact with a therapist. At

one end of the spectrum, Sanders et al. [60] examined a ‘pure’ self-help treatment with

no therapist interaction based on a self-help book – Breaking the Bonds of Irritable Bowel

Syndrome [76] – which included instructions on evidence-based CBT procedures for IBS.

At the other end, Jarrett et al.’s [31] active intervention consisted of nine 60-minute

sessions with a psychiatric nurse (six of which were telephonic), Lackner et al.’s [11]

‘self-administered CBT’ supplemented self-helpmaterialswith four 60-minute clinic visits

and two ten-minute telephone calls and Moss Morris et al.’s [55] ‘cognitive behavioural

self-management program’ was supplemented with three 60-minute sessions (one

face-to-face and two telephonic).

The CBT-based intervention studies generally guided participants to notice and

challenge unhelpful beliefs and behaviours through a variety of means: bibliographic

[60] or online instruction, guided by brief therapist feedback messages [47,56] or occa-

sional face-to-face or telephone sessions [11,31,55].

Alternatively, in line with other acceptance and commitment-based treatments, the

three CBT-based studies by Ljótsson et al. [13,53,68] examined the effects of acceptance

during exposure through mindfulness. These did not directly address ‘cognitive content’

or encourage systematic disputation of unhelpful beliefs. Again, participants received

short encouraging, supporting or corrective messages from their therapists.

In addition to CBT specific procedures, the majority of CBT treatments contained

additional components (education about the GI system, stress and IBS; dietary advice).

Furthermore, relaxation was an explicit component in the six ‘traditional’ CBT studies

[11,31,47,55,56,60] which is also an inherent component of hypnosis.

The two studies that formally describe their intervention as ‘hypnosis’ adapted the

existing treatment protocols used by these research groups in their clinical work with

IBS patients [66,77]. The use of scripted recordings of gut-based hypnotic suggestions

in these studies allows for a high level of standardisation, though no possibility of

personalizing suggestions/imagery based on patient preference. While Forbes et al.

[66] described their treatment as hypnosis, modelled on the approach developed by

Whorwell et al. [75], additional non-hypnosis components of their intervention make

it difficult to determine the specific contribution of hypnosis to their reported treat-

ment effects (which were in fact not statistically significant). Alternatively, Palsson

et al.’s [67] study contained a relatively ‘pure’ and specific form of hypnosis along with

standardmedical care. The intervention is clearly defined and describedwith each session

consisting of trance induction, deepening, and gut-related suggestions and imagery aimed

at producing relaxation, reduced pain and attention to gut symptoms.

Lastly, the expressive writing intervention tested by Halpert et al. [44] was based

on principle that expressing traumatic, stressful or emotional events can have impact

on both emotional and physical health [78,79]. The principle theory guiding this

approach is Foa and Kozac’s emotional processing model [80]. Participants were

asked to ‘write freely of their thoughts and feelings about IBS’ (online) for 30 minutes

each day, for four consecutive days.

Compliance with the interventions was assessed in nine of the 12 studies using

online logs [66,67], module completion [13,53,68] and regular submission of home-

work via email or post [31,47,55,60].

Treatment outcomes

Naturally, all measured IBS symptom severity, but many additionally measured

quality of life and psychiatric symptoms (particularly anxiety and depression). IBS

symptom severity was measured using The IBS-SSS [11,44,55,67]; the GSRS-IBS

[13,47,53,68] and GI-symptom diaries [31,53,60,66]. With the exception of Moss-Morris

et al. [55], quality of life was also indexed in all studies, predominantly measured by the

IBS-QOL.

Five of the studies examined changes in IBS-related cognitions, such as catastrophising

[13,31,44,47,56], four examined visceral sensitivity orGI-specific anxiety [13,47,53,68] and

four examined (work and social) functioning [31,53,55,68].

Waitlist or TAU comparisons

All active interventions showed benefit on IBS symptoms severity measures

[11,31,44,47,53,55,56,60,67,68]. However, only five studies found this effect when

intention to treat (ITT) analysis was used [11,31,53,55,68].

Similarly, active interventions led to significantly greater improvements on quality

of life than TAU or a waitlist control in seven of the studies [11,31,47,53,56,67,68], with

the same four studies using ITT. These improvements were maintained at follow-up in

four of the studies [13,31,67,68].

Greater reductions in anxiety were noted in four studies [31,53,55,68] only Jarrett

et al. [31] and Moss-Morris et al. [55] found these reductions at follow up.

Table 1

Consensus quality ratings of studies based on CCDAN scoring.

Study

Forbes et al. [66]

Hunt et al. [47]

Sanders et al. [60]

Halpert et al. [44]

Palsson et al. [67]

Ljotsson et al., [53]

Oerlemans et al. [56]

Ljotsson, et al. [68]

Ljotsson et al. [13]

Lackner et al. [11]

Moss-Morris et al. [55]

Jarrett et al. [31]

Total

25

27

30

32

34

37

41

44

45

46

46

47

Item total

Q13

0

0

0

0

0

0

0

0

0

0

0

0

0

Q28

0

0

0

0

0

0

0

0

1

1

0

0

2

Q15

0

0

0

0

2

0

0

0

0

2

0

0

4

Q27

1

0

0

0

1

0

2

1

0

1

0

0

6

Q25

0

0

0

1

1

0

1

0

1

0

1

2

7

Q19

1

1

0

0

0

1

0

1

1

1

1

2

9

Q4

0

0

0

0

0

0

2

2

2

2

2

2

12

Q24

1

1

1

1

1

1

1

1

1

1

1

1

12

Q6

0

2

2

0

0

2

0

2

2

2

2

0

14

Q16

0

1

1

1

2

1

1

1

1

1

2

2

14

Q2

1

1

0

0

1

2

2

1

2

1

1

2

14

Q9

1

1

1

1

1

1

1

2

1

1

2

2

15

Q14

1

2

1

1

1

1

1

1

1

2

2

2

16

Q23

0

0

0

2

0

2

2

2

2

2

2

2

16

Q3

1

1

1

1

2

1

1

2

2

1

2

2

17

Q26

0

0

2

0

2

2

2

2

2

2

2

2

18

Q5

2

1

2

0

0

2

1

2

2

2

2

2

18

Q21

0

1

1

2

2

1

2

2

2

2

2

2

19

Q11

1

0

2

2

1

2

2

2

2

2

2

2

20

Q18

1

2

0

2

2

1

2

2

2

2

2

2

20

Q1

1

1

1

2

2

2

2

2

2

2

2

2

21

Q12

0

2

2

2

1

2

2

2

2

2

2

2

21

Q29

2

2

2

2

0

2

2

2

2

2

2

2

22

Q22

2

1

1

2

2

2

2

2

2

2

2

2

22

Q10

2

0

2

2

2

2

2

2

2

2

2

2

22

Q20

1

2

2

2

2

1

2

2

2

2

2

2

22

Q8

2

1

2

2

2

2

2

2

2

2

2

2

23

Q17

2

2

2

2

2

2

2

2

2

2

2

2

24

Q7

2

2

2

2

2

2

2

2

2

2

2

2

24

Note. Q13 = blinding of assessor; Q28 = credibility; Q15 = details on side effects; Q27 = assess concurrent drug use; Q25 = follow-up duration; Q26 = co-interventions avoided or

equal; Q19 = inclusion of withdrawals; Q4 = power calculation; Q24 = allegiance to therapy; Q6 = concealment of allocation; Q16 = information on withdrawals; Q2 = sample size;

Q9 = representative subjects and source; Q14 = assessment of compliance; Q23 = declarations of interest; Q3 = duration of trial; Q5 = randomised allocation; Q21 = appropriate

statistical analysis; Q11 = exclusion criteria; Q18 = information on comparability; Q1 = clear objectives/outcomes a priori; Q12 = sample demographics; Q29 = consecutive subjects;

Q22 = conclusions justified; Q10 = inclusion criteria; Q20 = presentation of results; Q8 = manualisation; Q17 = outcome measures described; Q7 = treatments clearly described.

107R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

Table 2

Characteristics of the 12 studies included in the review

Study Study

description

Conditions Source of

participants

IBS

definition

Mean age

(S.D.)

Follow up Main findings

(relative to controls)

Moss-Morris

et al. [55]

RCT 1. Self-administered CBT &

self-management manual (8 weeks):

7 modules, 1 × 60 minute

face-to-face session, 2 × 60 minute

telephone sessions (n = 31)

2. Treatment as usual (TAU): IBS

Factsheet (n = 33)

Primary care Rome I or

Rome II

39.5

(16.8)

3 month,

6 month

Compared to TAU (ITT analysis)

IBS symptom severity

(IBS-SSS)↔. ↑FU

Functioning (W&SAS) ↑FU

Anxiety (HADS)↔, ↑FU

Oerlemans

et al. [56]

Feasibility

trial/open

label RCT

1. CBT e-intervention using digital

assistants (4 weeks): diaries 3 times

daily & daily CBT-based SMS

messages (n = 37)

2. TAU (n = 39)

Primary care &

community

advertising

Rome III 38.2

(13.8)

3 month Compared to TAU

Catastrophising thoughts

(PCS) ↑ FU

Abdominal pain ↑

Quality of life (IBS-QOL) ↑

Dysfunctional cognitions

(CSFBD) ↔

Sanders et

al. [60]

Preliminary

study

1. CBT-based self-help (8 weeks): no

direct or indirect therapist contact

(n = 17)

2. Waitlist (n = 11)

Specialist GI clinic &

community

advertising

Rome II 49.3

(12.3)

3 month Compared to Waitlist

IBS symptom severity (CPRS) ↑

Quality of life (IBS-QOL) ↔

Psychological distress (BSI) ↔

Ljotsson et

al. [13]

RCT 1. Internet-delivered Exposure-based

CBT (10 weeks): online modules &

weekly messages. (n = 98)

2. Internet-delivered Stress-

Management (10 weeks): online

modules & weekly messages (n = 97)

Specialist GI clinic,

CBT therapist, com-

munity advertising

Rome III 38.9

(11.1)

6 month Compared to active control

IBS severity (GSRS-IBS) ↑ FU

Quality of life (IBS-QOL) ↑ FU

GI Anxiety (VSI) ↑ FU

Ljotsson et

al. [53]

RCT 1. Internet-delivered Exposure-based

CBT (10 weeks): online modules &

weekly messages (n = 43)

2. Waitlist: with online discussion

forum& supportivemessages (n = 43)

Specialist GI clinic &

community

advertising

Rome III 34.6

(9.4)

3 month Compared to Waitlist (ITT

analysis)

IBS symptom severity (CPSR;

GSRS-IBS) ↑

Quality of life (IBS-QOL) ↑

Functioning (Sheehan Disabili-

ty Scales), ↑

GI anxiety (VSI) ↑

Depression (MADRD-S) ↑

Lackner et

al. [11]

RCT 1. Minimal contact CBT: self-study

materials, 4 × 60 minute face-to-face

sessions and 2 × 10 minute tele-

phone sessions. (n = 25)

2. Therapist-delivered CBT

(10 × 60 minute sessions) (n = 23)

3. Waitlist (n = 27)

Primary care &

community

advertising

Rome II 46.6

(16.7)

2 week Compared to Waitlist (ITT

analysis)

Adequate relief, IBS symptom

severity (IBS-SSS) ↑

Quality of life (IBS-QOL) ↑

Compared to active control

Adequate relief, IBS symptom

severity (IBS-SS) ↔

Quality of life (IBS-QOL) ↔

Jarrett et al.

[31]

RCT 1. CBT-based self-management:

6 × 60 minute telephone sessions,

3 × 50 minute face-to-face sessions

(n = 58).

2. Therapist-delivered CBT-based

self-management: 9 × 60 minute

sessions (n = 58).

3. TAU (n = 60)

Specialist GI clinic,

community

advertising.

Rome II 44 (14) 6 month,

12 month

Compared to Waitlist (ITT)

IBS symptom severity

(IBS-SSS) ↑ FU

Quality of life (IBS-QOL) ↑ FU

Functioning (WPAI) ↑ FU

Psychological distress (BSI) ↑ FU

Cognitions (CSFBD) ↑

Compared to active control (ITT)

Psychological distress (BSI) ↑ FU

Cognitions (CSFBD) ↓ FU

IBS symptom severity

(IBS-SSS) ↔

Quality of life (IBS-QOL) ↔

Functioning (WPAI) ↔

Hunt et al.

[47]

RCT 1. Internet CBT (5 weeks): online

modules &weeklymessages (n = 28).

2. Waitlist (n = 26).

Community

advertising

Self-report of

medical

diagnosis

38.5 (11) 3 month

(treatment

condition only)

Compared to waitlist

IBS symptom severity

(GSRS-IBS) ↑ Quality of life

(IBS–QOL) ↑α

Visceral Anxiety Sensitivity

(ASI) ↑α Cognitions (CPSQ) ↑α

Halpert et

al., [44]

Exploratory

pilot study

1. Online expressive writing (4 days):

4 × daily 30 minute writing sessions

2. Non-writers (6 weeks)

Community

advertising

Rome III 43 (11.8) 3 month Compared to non-writers

IBS symptom severity

(IBS-SSS) ↑*

Quality of life (IBS-QOL) ↑

Cognitions (CS-FBD) ↑*

Ljotsson et

al. [68]

RCT 1. Internet-delivered Exposure-based

CBT (10 weeks): online modules &

weekly messages. (n = 30)

2. Waitlist: with online discussion

forum (n = 31)

Specialist GI clinic Rome III 34.9

(11.3)

12 month Compared to waitlist

IBS symptom severity

(GSRS-IBS) ↑*α

Quality of life (IBS-QOL) ↑*α

Functioning (Sheehan

Disability Scales) ↑*

Visceral sensitivity (VSI) ↑*α

108 R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

Active control comparisons

Ljótsson et al. [13] found participants in the ‘internet CBT’ condition experienced

greater improvements in terms of IBS symptom severity, anxiety, quality of life and

cognitions than those in their ‘internet stress-management’ condition. They utilised

an ITT analysis and the findings were broadly maintained at follow up.

The other three studies which utilised an active control generally found no statis-

tically significant differences in IBS-symptom outcomes between their minimal-

contact and therapist-delivered conditions [11,31,66]. The only differences noted

were on secondary measures: Jarrett et al. [31] found that in comparison to therapist

delivered CBT, their minimal-contact intervention led to stronger improvements in

psychological distress but weaker improvements in IBS-related cognitions, whilst

Forbes et al. [66] noted a slight increase in anxiety in their hypnosis audiotape

condition.

Inclusion of withdrawals in analysis. Four studies [44,56,60,67] did not include

those who dropped out of the treatment in their analysis. Of the eight studies which

did so, six [13,47,55,53,66,68] estimated all such outcomes based on last observations

carried forward, whilst only two [11,31] actually attempted to obtain outcome data for

those participants who had not completed the intervention. In addition, whilst the

number of withdrawals was reported in all studies except one [66], only three studies

[31,55,67] reported the reasons for such withdrawals; data which provides important

insights into the acceptability of minimal-contact interventions.

Follow up data

Whilst ten of the twelve studies did include some form of extended follow up data,

it was only possible to compare the different conditions at follow up in six studies. This

was because two of the studies [53,68] only followed up their intervention group;

further two studies [47,60] utilised a cross-over design so control group data was

combined with data from the intervention condition. Including the follow up period,

only five studies had a total duration of more than six months [13,31,55,67,68].

None of the studies reported adverse effects of the active treatment.

Effect sizes

Table 3 summarises the effect sizes for studies that contained CBT-based minimal-

contact therapy and waitlist control or TAU conditions. This indicated a large overall

mean effect size for symptom severity favouring minimal-contact therapy at the end

of treatment. Ljótsson et al. [13] compared a minimal-contact treatment with an active

control condition, yet this study still produced a medium effect size (d = 0.50) effect

favouring the ‘active’ minimal contact CBT condition. The two studies [11,31] which

compared minimal-contact with therapist-delivered treatment showed similar effect

sizes in the two conditions relative to control conditions. Thus, the medium to large

effects found in the minimal-contact therapy conditions in Lackner et al (d = 1.58;

[11]) and Jarrett et al (d = 0.72; [31]; see Table 3) compared well with the

therapist-delivered conditions in these studies (d = 1.29 [11]; d = 0.92 [31]).

Overall, a medium effect size was found for quality of life at the end of treatment.

Use of an active treatment condition in Ljótsson et al. [13] again favoured the active

minimal-contact condition, producing a medium effect. The medium effect sizes

found for the minimal-contact condition in Lackner et al. [11] and Jarrett et al. [31]

for quality of life again compared well against the therapist-delivered conditions

(d = 0.76 [11]; d = 0.30 [31]).

Discussion

This paper is the first systematic review assessing the quality of

current literature on minimal-contact psychological interventions

for IBS. Although this is a relatively novel area of outcome research

there have been rapid advances, with all but two of the reviewed stud-

ies being published after 2008. Furthermore, most studies are RCTs.

These suggest that minimal-contact interventions for IBS symptom

management are a promising approach to reducing IBS symptom sever-

ity and improving quality of life. It remains unclear which other symp-

tom (or well-being) domains are most affected by treatment. Those

studies which compared minimal-contact approaches to therapist-

delivered treatments indicate that minimal-contact interventions can

achieve similar outcomes [11,31,66].

Formal assessment of the methodological quality suggests common

strengths across studies. These relate especially to clear descriptions

of methodology (clear descriptions of nature and details of interven-

tions, sample characteristics, outcomemeasures, inclusion/exclusion

criteria, appropriate statistical analysis and randomization proce-

dure, if applicable). Other strengths related to a priori reporting of

study objectives and outcomes as well as general reporting of results

and reaching justified conclusions. Reduced or absent therapist contact

minimises the possibility of ‘therapist drift’ and increases standardisation

of treatment across participants, another notable strength across studies.

The qualitative assessment of studies (Table 1) suggests that methodo-

logical quality has improved since the publication of a general review

of psychological interventions for IBS [8]. The improved scores are likely

to reflect widespread adoption of CONSORT [81] guidelines since that

review.

Given the limited number of studies and the varied modes of

intervention (hypnosis, expressive writing, ‘traditional-,’ and ‘accep-

tance and commitment-based-’ CBT) it is not yet possible to comment

on whether any particular treatment (or treatment component) is

superior to any other. Furthermore, only one study [13] specifically test-

ed a psychologically-based (CBT) intervention against an alternative

psychological approach, although the comparison condition – stress

management – might be considered a non-bona fide therapy. As

would be expected, given the more stringent control of ‘non-specific

treatment factors,’ this study [13] showed a lower effect size favouring

the active minimal contact CBT relative to the mean of the other CBT

studies. However, amediumeffect [13] in the context of an active control

condition, with evidence of durability over time is certainly promising.

Furtherwork using active controls is needed to determine the robustness

of such an effect. The current reliance on waitlist controls does not allow

us to untangle the potentially potent effects of non-specific factors

(e.g. clear explanation for symptoms, heightened expectation of im-

provement) present in all minimal-contact interventions. Furthermore,

given that the placebo effect for psychological interventions tends to

last approximately 12 weeks [3], longer-term follow up periods for all

conditions is also an important methodological consideration for future

research.

No study examined outcomes using blind assessors. Such blind

assessment is considered critical for establishing credible effect sizes

[82]. However, one of the strengths of minimal-contact therapies –

especially those based exclusively on self-management via internet

instruction – is that, relative to fully therapist-guided, or reduced

Table 2 (continued)

Study Study

description

Conditions Source of

participants

IBS

definition

Mean age

(S.D.)

Follow up Main findings

(relative to controls)

Palsson et al.

[67]

Pilot study 1. Hypnosis audiotape (12 weeks):

bi-weekly 30 minute sessions & daily

13 minute sessions (n = 19)

2. TAU (n = 57)

Community

advertising

Rome II 43.6

(16.4)

3 month Compared to TAU

IBS symptom severity

(IBS-SSS) ↑*

Quality of life (IBS-QOL) ↑*

Psychological distress ↔

Forbes et al.

[66]

RCT 1. Hypnosis audiotape (12 weeks):

daily 30 minute sessions

2. Individual hypnotherapy

(12 weeks): 6 × 30 minute face-to

face-sessions

Specialist GI clinic Rome I 37

(median)

None Compared to active control

IBS Symptomseverity (diary)↔

Quality of life (SF-36) ↔

Psychological distress (GHQ,

HADS) ↓

↑ = significantly improvement at end of treatment; ↔ = no significant difference between groups at end of treatment (or follow up); ↓ = significantly less improvement (at end

or follow up); ↑ FU = improvement significant at follow-up. IBS-SSS = Irritable Bowel Syndrome Severity Scoring System; GSRS-IBS = Gastrointestinal Symptom Rating Scale;

CPRS = Composite Primary SymptomReduction Score; IBS-QOL = IBS Quality of Life;W&SAS = Work & Social Adjustment Scale;WPAI = Work Productivity and Activity Impairment

Questionnaire; CS-FBD = Cognitive Scale for Functional Bowel Disorders; VSI = Visceral Sensitivity Index; CPSQ: The Consequences of Physical Sensations Questionnaire; ASI: Anxiety

Sensitivity Index; HADS = Hospital Anxiety and Depression Scale; BSI = Brief Symptoms Index; MADRD-S = Montgomery Asberg Depression Rating Scale — Self Report;

109R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

contact therapies, they are relatively inexpensive to evaluate.While fea-

sible, the use of blind assessors is likely to increase the cost of treatment

evaluation substantially. Furthermore, direct evaluation of outcome by a

researcher entails reduced ecological validity of assessment for thera-

pies intended to be delivered exclusively on the internet.

A number of advantages of minimal-contact treatments were

outlined in the introduction section including efficiency (for services)

and convenience (for clients). From the clients’ perspective, many of

the traditional barriers to psychological therapy do indeed seem to be

minimised. However, whilst minimal-contact treatments potentially

enable greater access through flexible engagement, these modes of

treatment also require significant levels of motivation and time com-

mitment in the absence of – or with reduced amounts of – therapist

support. The wider literature on minimal-contact interventions has

noted reductions in compliance when therapist contact is reduced

[18,83] and higher dropout rates from those interventions which do

not involve therapist contact [84]. Although Lackner et al. [11], Jarrett

et al. [31] and Forbes et al. [66] report similar levels of attrition from

their minimal-contact and full-contact conditions, the lack of therapist

supportwas cited as an important reason for dropout from the hypnosis

audiotape condition in Palsson et al.’s [67] study. Clearly this is an

important issue for the design of services and selection of clients for

minimal-contact treatments. The issues around the acceptability of

minimal-contact therapies also highlight the importance of including

withdrawals in analyses (intention to treat or endpoint analysis) in

future research.

Ljótsson et al., [68] suggest acceptability of these treatments may be

a particularly salient issue in more representative IBS samples. In their

study of minimal contact internet-based exposure Ljótsson et al., [68]

found significantly higher levels of attrition (and poorer outcomes) in

a representative sample of IBS patients recruited from a specialist GI

compared to thosewho self-referred [13,53]. Themajority of the studies

included in this review relied on self-referral to some extent although

such individuals are not likely to be highly representative of typical

IBS patients encountered in gastroenterologist clinics [85–87]. Another

issue related to representativeness is the inconsistency in the use of

diagnostic procedures which varied both within and between studies.

Revisions of the Rome diagnostic criteria have led to improved specific-

ity and sensitivity of diagnosis [88,89] although this remains less reli-

able when performed by family doctors [90].

The issue of ‘representativeness’ remains an importantmethodolog-

ical and definitional issue for studies evaluating interventions which

will potentially be available to anyone, regardless of symptom charac-

teristics and severity, without evaluation by a healthcare practitioner.

Those with relatively mild symptoms – those who are perhaps most

likely to use minimal contact therapies – tend not to be specifically

recruited for efficacy studies of these interventions. In our evaluation

of ‘representativeness’ in this study (Table 1) we also have tended to

favour trials that used ‘typical’ IBS patients rather than those who

might be recommended for minimal contact therapies at the initial

stage of a stepped care process. It remains to be seen how treatment

efficacy interacts with initial IBS severity in future studies designed to

address this issue.

Attrition varied throughout the studies and may be attributed to

limited (or absent) therapist contact and symptom severity. In the

reviewed studies non-completers had higher symptom severity scores

[11,60], had higher levels of IBS-related impairment [68], were more

likely to meet criteria for an Axis I disorder [60] and provided lower

credibility ratings at the start of the intervention [53]. Conversely,

Oerlemans et al. [56] found their dropouts had higher baseline levels

of IBS-related quality of life. From a clinical perspective these findings

generally highlight the need for placing minimal-contact treatments

for IBS within a stepped care framework allowing for referral to more

intensive face-to-face treatment depending on the presence of severe

symptoms and comorbidity [10,14].

It has also been argued that at least some therapist feedback is

necessary to enhance efficacy. For example, Palmqvist et al. [91] found

evidence for a strong correlation between therapist input and outcome

across a range of disorders. However, the effect of mode of therapist

feedback (face-to-face, telephonic, video conferencing, text messaging,

email) is less clear. On the basis of the two studies reviewed here in

which face-to-face contact was substantially replaced by indirect con-

tact (telephonic), high levels of efficacy seem to be retained [11,31].

The relative effect on outcome of a complete replacement of face-to-

face with indirect modes of contact remains untested.

In addition to the degree of contact (either face-to-face or indirect)

the necessity for credentialed/accredited therapists rather than more

minimally-trained practitioners in achieving positive outcomes is

unclear. Within a stepped care context it may be possible that contact

with credentialed therapists could be replaced by minimally-trained

professionals with appropriate supervision, supplementing self-

guided activities. Of the 12 studies outlined above, seven provided

information about the credentials of those who administered guided

interventions (or provided participants with feedback). In four studies

[11,31,55,68] these were experienced and highly-trained professionals,

whilst in three studies [13,53,56] the remote therapists were graduate

psychology students or junior researchers. It therefore remains an im-

portant empirical question as to the level of accreditation or experience

required for safe and effective minimal-contact treatment delivery.

Similarly, there is currently little evidence regarding the economic

aspects of minimal-contact interventions for IBS. Only one of the

reviewed studies [68] directly addressed the issue of cost-effectiveness.

Building on previous work [33], this study [68] noted that internet-CBT

treatment produced societal cost-savings compared to waiting list

control, offsetting short- and long-term treatment costs despite high

dropout levels.

One potential concern in relation to widespread availability of

minimal contact therapies is that this will lead to treatment over-use,

with the cost implications that attend this. As already, the commitment

and motivation required to engage in the types of therapy reviewed

here would be expected to guard against over-treatment. However, it

remains an open question as to whether greater dissemination will

lead to significant levels of waste.

A number of systematic reviews have now been published on

minimal-contact therapies across a variety of disorders although

these have used rather ‘minimalist’ quality-assessment instruments.

Our decision to use a comprehensive assessment tool was driven by

the continued need to formally assess the full range of relevantmethod-

ological features of studies of minimal-contact therapies. While the

study-quality items of the CCDAN measure were specifically designed

for studies of traditional face-to-face therapies, most were applicable

to minimal-contact therapies. On the other hand a small number of

items on the CCDAN exhibited limited sensitivity and did not appear

to distinguish between studies very well. This suggests an on-going

Table 3

Overall treatment effect sizes of CBT-based interventions: symptomseverity and quality of

life

Study Severity Quality of life

End of

treatment

Follow up End of

treatment

Follow up

Moss-Morris et al. [55] 0.53 0.96 ∫ N/A N/A

Oerlemans et al. [56] 0.30 0.19 # 0.18 0.15 #

Sanders et al. [60] 0.33 0.31

Ljotsson et al. [13] 0.50† 0.61 ∫ † 0.44† 0.24 ∫ †

Ljotsson et al. [53] 1.30 1.10

Lackner et al. [11] 1.58 0.92

Jarrett et al. [31] 0.72 0.49 ∫ 0.26 0.21 ∫

Hunt et al. [47] 0.93 N/A

Ljotsson et al. [68] 1.25 1.19

Mean effect size 0.83 0.56 0.63 0.20

# 3 month FU; ∫ 6 month FU; † Comparison with active control not used to calculate

overall mean.

110 R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

need for a quality-assessment instrument to be developed for evaluat-

ing such minimal-contact therapies for IBS and other disorders.

To summarise, minimal-contact therapies for IBS appear to be

superior to usual care and a small number of studies have demonstrated

equivalence to fully therapist-delivered interventions. The quality of

research in this area is varied but improving. We have tentatively

recommended areas for improvement of future studies, particularly in

relation to the use of active (non waiting list) control groups and long

follow up times.

Conflict of interest statement

The authors have no conflicts of interests to declare.

Acknowledgments

No external funder had a role in the study design or collection,

analysis or interpretation of data, writing of the report or decision

to submit the article for publication.

References

[1] Zijdenbos IL, De Wit NJ, Van Der Heijden GJ, Rubin G, Quartero AO. Psychologicaltreatments for the management of irritable bowel syndrome. Cochrane DatabaseSyst Rev 2009;1.

[2] Drossman D, Creed F, Olden K, Svedlund J, Toner B, Whitehead W. Psychosocialaspects of the functional gastrointestinal disorders. Gut 1999;45:1125–30.

[3] Spiller R, Aziz Q, Creed F, Emmanuel A, Houghton L, Hungin P, et al. Guidelines onthe irritable bowel syndrome: mechanisms and practical management. Gut2007;56:1770.

[4] Jones R, Lydeard S. Irritable bowel syndrome in the general population. BMJ1992;304:87–90.

[5] Talley NJ, Gabriel SE, Harmsen WS, Zinsmeister AR, Evans RW. Medical costs incommunity subjects with irritable bowel syndrome. Gastroenterology 1995;109:1736–41.

[6] Wilson S, Roberts L, Roalfe A, Bridge P, Singh S. Prevalence of irritable bowelsyndrome: a community survey. Br J Gen Pract 2004;54:495–502.

[7] Irritable bowel syndrome in adults: diagnosis and management of irritable bowelsyndrome in primary care.NICE; 2008 [http://www.nice.org.uk/. Accessed:29-Dec-2012].

[8] Lackner JM, Mesmer C, Morley S, Dowzer C, Hamilton S. Psychological treatmentsfor irritable bowel syndrome: a systematic review and meta-analysis. J ConsultClin Psychol 2004;72:1100.

[9] Wilson GT, Zandberg LJ. Cognitive-behavioral guided self-help for eating disorders:effectiveness and scalability. Clin Psychol Rev 2012;32:343–57.

[10] Richards DA, Suckling R. Improving access to psychological therapy: the Doncasterdemonstration site organisational model’, presented at the Clinical PsychologyForum New Series, 181; 2008. p. 9.

[11] Lackner JM, Jaccard J, Krasner SS, Katz LA, Gudleski GD, Holroyd K.Self-administered cognitive behavior therapy for moderate to severe irritablebowel syndrome: clinical efficacy, tolerability, feasibility. Clin GastroenterolHepatol 2008;6:899–906.

[12] Newman MG, Szkodny LE, Llera SJ, Przeworski A. A review of technology-assistedself-help and minimal-contact therapies for drug and alcohol abuse and smokingaddiction: Is human contact necessary for therapeutic efficacy? Clin Psychol Rev2011;31:178–86.

[13] Ljótsson B, Hedman E, Andersson E, Hesser H, Lindfors P, Hursti T, et al.Internet-delivered exposure-based treatment vs. stress management for irritablebowel syndrome: a randomized trial. Am J Gastroenterol 2011;106:1481–91.

[14] Haaga DAF. Introduction to the special section on stepped care models in psycho-therapy. J Consult Clin Psychol 2000;68:547.

[15] DH/Mental Health Programme/Improving Access to Psychological Therapies. IAPTImplementation Plan: National Guidelines for Regional Delivery; 2008.

[16] Bower P, Richards D, Lovell K. The clinical and cost-effectiveness of self-help treat-ments for anxiety and depressive disorders in primary care: a systematic review.Br J Gen Pract 2001;51:838.

[17] Kaltenthaler E, Brazier J, De Nigris E, Tumur I, Ferriter M, Beverley C, et al.Computerised cognitive behaviour therapy for depression and anxiety update: asystematic review and economic evaluation. Health Technol Assess 2006;10.

[18] Newman MG, Erickson T, Przeworski A, Dzus E. Self‐help and minimal‐contacttherapies for anxiety disorders: Is human contact necessary for therapeutic efficacy?J Clin Psychol 2003;59:251–74.

[19] Cuijpers P. Bibliotherapy in unipolar depression: a meta-analysis’. J Behav TherExp Psychiatry 1997;28:139–47.

[20] McKendreeSmith NL, Floyd M, Scogin FR. Self‐administered treatments fordepression: A review. J Clin Psychol 2003;59:275–88.

[21] Gould RA, Clum GA. A meta-analysis of self-help treatment approaches. ClinPsychol Rev 1993;13:169–86.

[22] Marrs RW. A meta-analysis of bibliotherapy studies. Am J Community Psychol1995;23:843–70.

[23] Kaldo V, Levin S, Widarsson J, Buhrman M, Larsen HC, Andersson G. Internetversus group cognitive-behavioral treatment of distress associated with tinnitus:a randomized controlled trial. Behav Ther 2008;39:348–59.

[24] Andersson G, Lundström P, Ström L. Internet‐based treatment of headache: doestelephone contact add anything? Headache J Head Face Pain 2003;43:353–61.

[25] Ström L, Pettersson R, Andersson G. Internet-based treatment for insomnia: acontrolled evaluation. J Consult Clin Psychol 2004;72:113–20.

[26] Buhrman M, Fältenhag S, Ström L, Andersson G. Controlled trial of internet-basedtreatment with telephone support for chronic back pain’. Pain 2004;111:368–77.

[27] Tate DF, Wing RR, Winett RA. Using internet technology to deliver a behavioralweight loss program. J Am Med Assoc 2001;285:1172–7.

[28] Cuijpers P, Straten A, Andersson G. Internet-administered cognitive behaviortherapy for health problems: a systematic review. J Behav Med 2007;31:169–77.

[29] Toner BB, Segal ZV, Emmott SD, Myran D. Cognitive-behavioral treatment ofirritable bowel syndrome: The brain–gut connection. The Guilford Press; 2000 .

[30] Glasgow RE, Rosen GM. Behavioral bibliotherapy: A review of self-help behaviortherapy manuals. Psychol Bull 1978;85:1.

[31] Jarrett ME, Cain KC, Burr RL, Hertig VL, Rosen SN, Heitkemper MM. Comprehensiveself-management for irritable bowel syndrome: Randomized trial of in-person vs.combined in-person and telephone sessions. Am J Gastroenterol 2009;104:3004–14.

[32] Ford AC, Talley NJ, Schoenfeld PS, Quigley EMM, Moayyedi P. Efficacy of antide-pressants and psychological therapies in irritable bowel syndrome: systematicreview and meta-analysis. Gut 2009;58:367–78.

[33] Andersson G. Cost-effectiveness of internet-based cognitive behavior therapy forirritable bowel syndrome: results from a randomized controlled trial. BMC PublicHealth 2011;11:215.

[34] Arn I. Psychodrama group therapy for patients with functional gastrointestinaldisorders - A controlled long-term follow-up study. Psychother Psychosom1989;51:113–9.

[35] Blanchard EB. Prediction of outcome from the self-regulatory treatment of irritablebowel syndrome. Behav Res Ther 1988;26:187–90.

[36] Blanchard EB. A controlled evaluation of group cognitive therapy in the treatmentof irritable bowel syndrome. Behav Res Ther 2007;45:633–48.

[37] Blanchard EB. Psychological changes associated with self-regulatory treatments ofirritable bowel syndrome. Biofeedback Self Regul 1987;12:31–7.

[38] Bogalo L. ‘The effectiveness of homework tasks in an irritable bowel syndromecognitive behavioural self-management programme. N Z J Psychol 2006;35:120–5.

[39] Everitt HM. Management of irritable bowel syndrome in primary care: feasibilityrandomised controlled trial of mebeverine, methylcellulose, placebo and a patientself-management cognitive behavioural therapywebsite. (MIBS trial). Gastroenterol2010;10:136.

[40] Gaylord S. Therapeutic impact of mindfulness meditation on Irritable Bowel Syndrome(IBS): Results of a randomized controlled trial. Gastroenterology 2011;140:S50.

[41] Gaylord S. Mindfulness for irritable bowel syndrome: protocol development for acontrolled clinical trial. BMC Complement Altern Med 2009;9:24.

[42] Halpert A. Patient centered educational intervention in irritable bowel syndrome(IBS). Gastroenterology 2010;138:S164.

[43] Halpert A, Rybin D. Expressive writing is a promising therapeutic modality forirritable bowel syndrome (IBS). Gastroenterology 2009;136:A118.

[44] Halpert A, Rybin D, Doros G. Expressive writing is a promising therapeutic modalityfor the management of IBS: a pilot study. Am J Gastroenterol 2010;105:2440–8.

[45] Heitkemper MM. Self-management for women with irritable bowel syndrome.Clin Gastroenterol Hepatol 2004;2:585–96.

[46] Hislop IG. Effect of very brief psychotherapy on the irritable bowel syndrome.Med J Aust 1980;2:620–3.

[47] Hunt G, Moshier S, Milonova M. Brief cognitive-behavioral internet therapy forirritable bowel syndrome. Behav Res Ther 2009;47:797–802.

[48] Keefer L. Behavioral interventions may prolong remission in patients with inflam-matory bowel disease. Behav Res Ther 2011;49:145–50.

[49] Lackner JM, Krasner SS, Quigley B, Gudleski D, Katz A, Powell C, et al. Who dropsout from cognitive behavior therapy (CBT) for IBS?: Results of an NIH randomizedclinical trial. Gastroenterology 2009;136:A379.

[50] Leibbrand R. Cognitive behavior therapy for functional gastrointestinal disorders:Is group treatment effective? Acta Neuropsychiatrica 2003;15:242–8.

[51] Ljótsson B, Andréewitch S, Hedman E, Ruck C, Andersson G, Linfors N. Exposureand mindfulness based therapy for irritable bowel syndrome: An open pilotstudy. J Behav Ther Exp Psychiatry 2010;41:185–90.

[52] Ljótsson B, Hedman E, Lindfors P, Hursti T, Lindfors N, Andersson G, et al.Long-term follow-up of internet-delivered exposure and mindfulness basedtreatment for irritable bowel syndrome. Behav Res Ther 2011;49:58–61.

[53] Ljótsson B, Falk L, Vesterlund AW, Hedman E, Lindfors, Ruck PC, et al.Internet-delivered exposure and mindfulness based therapy for irritable bowelsyndrome — a randomized controlled trial. Behav Res Ther 2010;48:531–9.

[54] Mikocka-Walus A. Cognitive-behavioural therapy (CBT) to manage inflammatorybowel disease (IBD): Preliminary findings. J Crohns Colitis 2011;5:S142.

[55] Moss-Morris R, McAlpine L, Didsbury LP, Spence MJ. A randomized controlled trialof a cognitive behavioural therapy-based self-management intervention for irritablebowel syndrome in primary care. Psychol Med 2010;40:85–94.

[56] Oerlemans S, van Cranenburgh O, Herremans PJ, Spreeuwenberg P, van Dulmen S.‘Intervening on cognitions and behavior in irritable bowel syndrome: A feasibilitytrial using PDAs. J Psychosom Res 2011;70:267–77.

[57] Payne A, Blanchard EB. A controlled comparison of cognitive therapy andself-help support groups in the treatment of irritable bowel syndrome. J ConsultClin Psychol 1995;63:779.

111R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112

[58] RiberdyPoitrasMR, Verrier P, So C, Paquet S, BouinM, Poitras P. Group counselling psy-chotherapy for patients with functional gastrointestinal disorders: Development ofnewmeasures for symptom severity and quality of life. Dig Dis Sci 2002;47:1297–307.

[59] Robinson A, Lee V, Kennedy A, Middleton L, Rogers A, Thompson DG, et al. Arandomised controlled trial of self-help interventions in patients with a primarycare diagnosis of irritable bowel syndrome. Gut 2006;55:643–8.

[60] Sanders KA, Blanchard EB, Sykes MA. Preliminary study of a self-administeredtreatment for irritable bowel syndrome: Comparison to a wait list controlgroup. Appl Psychophysiol Biofeedback 2007;32:111–9.

[61] Sanders KA, Blanchard EB, Sykes MA. Self-administered treatment for irritablebowel syndrome: Comparison to a wait list control group. Diss Abstr Int B SciEng 2006;66:6292.

[62] Tkachuk GA. Controlled trial of a multicomponent cognitive-behavioral grouptreatment for irritable bowel syndrome. Diss Abstr Int B Sci Eng 2002;62:4807.

[63] Tkachuk GA, Graff LA, Martin GL, Bernstein CN. Randomized controlled trial ofcognitive-behavioral group therapy for irritable bowel syndrome in a medicalsetting. J Clin Psychol Med Settings 2003;10:57–69.

[64] Turner SM. Process variables in small-group cognitive therapy for the irritablebowel syndrome. Diss Abstr Int B Sci Eng 1998;58:4476.

[65] WangW. Cognitive therapy for patients with refractory irritable bowel syndrome.Zhonghua nei ke za zhi [Chinese Journal of Internal Medicine] 2002;41:156–9.

[66] Forbes A, MacAuley S, Chiotakakou-Faliakou E. Hypnotherapy and therapeuticaudiotape: effective in previously unsuccessfully treated irritable bowel syndrome?Int J Colorectal Dis 2000;15:328–34.

[67] Palsson OS, Turner MJ, Whitehead WE. Hypnosis home treatment for irritablebowel syndrome: a pilot study. Int J Clin Exp Hypn 2006;54:85–99.

[68] Ljótsson B, Andersson G, Andersson E, Hedman E, Lindfors P, Andreewitch S, et al.Acceptability, effectiveness, and cost-effectiveness of internet-based exposuretreatment for irritable bowel syndrome in a clinical sample: a randomizedcontrolled trial. BMC Gastroenterol 2011;11:110.

[69] Moher D, Cook DJ, Jadad AR, Tugwell, Moher PM, Jones A, et al. Assessing the qualityof reports of randomised trials: implications for the conduct ofmeta-analyses. HealthTechnol Assess 1999;3:1–98.

[70] van Zanten SJOV, Talley N, Bytzer P, Klein K, Whorwell P, Zinsmeister A. Design oftreatment trials for functional gastrointestinal disorders. Gut 1999;45:II69–77.

[71] Gaffan EA, Tsaousis I, Kemp-Wheeler SM. Researcher allegiance and meta-analysis:the case of cognitive therapy for depression’. J Consult Clin Psychol 1995;63:966–80.

[72] Turk DC, Rudy TE, Sorkin BA. Neglected topics in chronic pain treatment outcomestudies: determination of success. Pain 1993;53:3–16.

[73] Chang L, Toner BB, Fukudo S, Guthrie E, Locke GR, Norton NJ, et al. Gender, age,society, culture, and the patient’s perspective in the functional gastrointestinaldisorders. Gastroenterology 2006;130:1435–46.

[74] Whorwell P, Prior A, Faragher E. Controlled trial of hypnotherapy in the treatmentof severe refractory irritable-bowel syndrome. Lancet 1984;2:1232.

[75] Whorwell P, Prior A, Colgan S. Hypnotherapy in severe irritable bowel syndrome:further experience. Gut 1987;28:423–5.

[76] Bolen BB. Breaking the bonds of irritable bowel syndrome: A psychological approachto regaining control of your life. New Harbinger Publications, Incorporated; 2000 .

[77] Palsson OS, Turner MJ, Johnson DA, Burnett CK, Whitehead WE. Hypnosis treat-ment for severe irritable bowel syndrome: investigation of mechanism and effectson symptoms. Dig Dis Sci 2002;47:2605–14.

[78] Pennebaker JW. Opening up: The healing power of expressing emotions. TheGuilford Press; 1997 .

[79] Pennebaker JW, Beall SK. Confronting a traumatic event: toward an understanding ofinhibition and disease. J Abnorm Psychol 1986;95:274.

[80] Foa EB, Kozak MJ. Emotional processing of fear: exposure to corrective information.Psychol Bull 1986;99:20.

[81] Boutron I, Moher D, Altman DG, Schulz KF, Ravaud P. Extending the CONSORTstatement to randomized trials of nonpharmacologic treatment: explanationand elaboration. Ann Intern Med 2008;148:295–309.

[82] Lynch D, Laws K, McKenna P. Cognitive behavioural therapy for major psychiatricdisorder: does it really work? A meta-analytical review of well-controlled trials.Psychol Med 2010;40:9–24.

[83] Newman MG, Szkodny LE, Llera SJ, Przeworski A. A review of technology-assistedself-help and minimal-contact therapies for anxiety and depression: Is humancontact necessary for therapeutic efficacy? Clin Psychol Rev 2011;31:89–103.

[84] Scogin F, Bynum J, Stephens G, Calhoon S. Efficacy of self-administered treatmentprograms: Meta-analytic review. Prof Psychol Res Pract 1990;21:42.

[85] Hilsden RJ, Meddings JB, Verhoef MJ. Complementary and alternative medicineuse by patients with inflammatory bowel disease: An Internet survey. Can JGastroenterol 1999;13:327.

[86] Jones MP, Bratten J, Keefer L. Quality of life in patients with inflammatory boweldisease and irritable bowel syndrome differs between subjects recruited fromclinic or the internet. Am J Gastroenterol 2007;102:2232.

[87] Soetikno RM, Mrad R, Pao V, Lenert LA. Quality-of-life research on the Internet:feasibility and potential biases in patients with ulcerative colitis. J AmMed InformAssoc 1997;4:426–35.

[88] Lea R, Hopkins V, Hastleton J, Houghton L, Whorwell P. Diagnostic criteria for ir-ritable bowel syndrome: utility and applicability in clinical practice. Digestion2004;70:210–3.

[89] Sperber AD, Shvartzman P, Friger M, Fich A. A comparative reappraisal of theRome II and Rome III diagnostic criteria: arewe getting closer to the ‘true’ prevalenceof irritable bowel syndrome? Eur J Gastroenterol Hepatol 2007;19:441.

[90] Thompson W, Heaton K, Smyth G, Smyth C. Irritable bowel syndrome in generalpractice: prevalence, characteristics, and referral. Gut 2000;46:78–82.

[91] Palmqvist B, Carlbring P, Andersson G. Internet-delivered treatments with orwithout therapist input: does the therapist factor have implications for efficacyand cost? Expert Rev Pharmacoecon Outcomes Res 2007;7:291–7.

[92] Morris SB. Estimating effect sizes from pretest-post test-control group designs.Organ Res Methods 2008;11:364–86.

112 R. Pajak et al. / Journal of Psychosomatic Research 75 (2013) 103–112