The important outcomes were prioritised by the topic-specific members (TSMs) through ranking methods and further confirmed by the standing Committee before the review was carried out.

They thought quality of life to be of particular importance when considering the effectiveness of antidepressant treatment for IBS. Outcomes of symptom response overall and individual symptom response (e.g. bloating, diarrhoea) were also considered important, although the impact of these factors on an individual cannot be assumed. The topic-specific Committee members noted that the improvement in a particular symptom may be viewed differently by the individuals involved. For example, some may consider improvement in their bloating symptoms to be the focal point when considering a treatment, while for others improvement in a different symptom, such as diarrhoea, would be most valuable. The Committee noted the limited reporting of adverse events in the included studies.

The Committee agreed that the outcomes from the included studies should be presented by the class of the drugs involved, that is by TCA and SSRI class. The heterogeneity of the included studies and the differences between the pharmacokinetics and pharmacodynamics of TCAs and SSRIs was further discussed. The Committee concluded that the evidence would be most appropriately presented within their drug class rather than combining the results from all of the included studies together (as had been done previously in CG61). The Committee agreed that the results of the included studies overall showed that antidepressants have an effect to improve the symptoms of IBS. It was agreed that there was more uncertainty with the evidence on SSRIs than with TCAs.

The lack of follow-up within the included studies was discussed by the Committee. It was agreed that the study length in most of the included studies was sufficient to detect a response in patients for the related outcomes. However, for consideration of any adverse effects and longer term symptom control of a fluctuating condition like IBS, further follow-up data would have been needed. The studies that had included adverse events had not reported these in detail.

The Committee noted that there is limited new evidence in the use of antidepressants for those with IBS. It was further discussed that there was no additional evidence that provided any justification for changing the original recommendations on the use of antidepressants developed in CG61. The Committee noted that of the 12 included studies only 2 had reported on the previous IBS treatment that participants had received prior to the study.

The Committee acknowledged that TCAs may cause constipation in some people, therefore clinicians should further consideration this treatment in people with IBS-C.

The Committee agreed that the results of this evidence review were consistent with the results of the evidence reviewed previously in CG61 and with the views of the topic-specific Committee members. Therefore, it was agreed that the existing recommendations for this review question from CG61 would be carried forward into this update.

Finally, as IBS is a chronic condition, the Committee agreed that extra caution should be given if TCAs are to be prescribed for people with IBS and depression, due to the potential harm associated with increased suididal ideation on these drugs. The Committee stated that prescribers should also refer to NICE guideline on depression in adults with a chronic physical health problem and NICE guidance on depression in adults.

The Committee reviewed the evidence identified and noted that there are areas of concern for the applicability of the included studies to the potential users of this guideline update. The majority of the included studies used participants from non-primary care settings. The Committee discussed that this may (but not necessarily) mean that these participants had more severe IBS symptoms than those in primary care.

Nevertheless, the topic-specific Committee members considered that these studies would have included a proportion of participants with symptoms that would be found within primary care; they therefore decided that it was appropriate to extrapolate the evidence.

The Committee noted that the doses used within some of the included studies, particularly for the TCAs, are higher than would (at least initially) be prescribed for IBS treatment. This raised questions about the directness of these studies and this is reflected within the GRADE tables for these studies. This was not applicable with the SSRIs as they are not usually used at low dose in IBS treatment.

The Committee agreed that there is low quality evidence that TCAs and SSRIs have some benefit in the treatment of the symptoms of IBS. Accepting the limitations of this evidence which are noted in the GRADE tables (appendix H), the Committee considered that there was no evidence to contradict or change the recommendations initially developed for CG61.

The Committee discussed the lack of recent published studies in this area. Consequently it was agreed that a modification of the research recommendation in CG61 was justified highlighting the need for further research into the treatment of IBS using antidepressants within primary care.

Furthermore, they noted that the initiation of TCAs and SSRIs for IBS (for analgesic effect) currently often happens within primary care. The topic-specific Committee members considered that referral to secondary care for this therapy is not necessary, that the prescription of antidepressants could be initiated and monitored in primary care. Therefore it was agreed that the included studies have relevance for primary care and that this is an important question to be reviewed within a primary care based guideline.

As currently there is still insufficient evidence on the use of antidepressants for the management of IBS, despite the existing research recommendation on this topic published in the original guideline, the Committee decided and agreed that the research recommendation should be relaunched as part of this update in order to promote more research in this area.

The important outcomes were prioritised by the topic-specific members (TSMs) through ranking methods and further confirmed by the standing Committee before the review was carried out. The Committee reviewed the use of the low FODMAP diet and discussed whether it would be appropriate to consider evidence where components of the diet had been modified. On the advice of the topic-specific members, the Committee concluded that the low FODMAP intervention should be considered as an entity, that it would not be appropriate to consider restriction of the individual short chain carbohydrates that constitute FODMAP. In the dietary and lifestyle advice section of the original CG61, individual components such as sorbitol (which is a polyol) were mentioned and included. This review only considers the low FODMAP diet as an intervention as a whole, and its effectiveness for managing IBS symptoms. This review does not include updating the individual components included in the original guideline.

The Committee considered that the outcomes in the included studies are relevant to those with IBS symptoms, though they noted that no quality of life outcomes were reported and that studies had reported outcomes relating to overall and/or individual symptoms. The Committee further noted that long-term outcomes for the low FODMAP diet, particularly on any potential adverse effects, will be very important. However, current included evidence did not have long enough follow-up period to capture these data.

The Committee agreed that there is some evidence that the low FODMAP diet has an effect on reducing the symptoms of those with IBS. However, this evidence is limited to a small number of localised trials with small participant numbers. The Committee also noted the RCT including participants with diarrhoea had not found an improvement in diarrhoea related symptoms with the low FODMAP diet. The Committee commented that the low FODMAP diet intervention period of these studies did not match current practice in the NHS, which is routinely 8 weeks due to the availability of a dietitian. It was also discussed that the studies did not include further follow-up or the graded re-introduction phase of high FODMAP foods that follows the initial use of the low FODMAP diet in current practice.

In recognition of the limitations of the evidence, the Committee considered whether there was sufficient evidence to enable them to make a specific stand alone recommendation relating to the low FODMAP diet. The Committee discussed that IBS is a common condition and that there may be a considerable impact relating to any recommendation of a dietary intervention. They acknowledged that the low FODMAP diet is currently being used in those with IBS and there is increasing public awareness of this diet. Therefore, guidance in this area would be beneficial and the Committee discussed that there are other dietary and lifestyle changes currently recommended for managing IBS. The Committee commented that any new recommendations relating to the low FODMAP diet should sit within the existing recommendations to ensure that the low FODMAP diet does not get a separate predominance due to its current popularity.

As there is a lack of evidence on the long-term edverse effects of low FODMAP diet, the Committee further discussed the potential harms of following the low FODMAP diet without dietetic support as the diet can be complex to understand and implement. These may include inappropriate or blanket restriction without suitable food replacements, which could lead to nutritional inadequacy (energy or micronutrient, in particular of calcium [Staudacher et al. 2012]) or even deficiency. In addition, in the short-term, a low FODMAP diet modifies faecal microbiota (Staudacher et al. 2012; Halmos et al. 2014). The long term effects of low FODMAP diet on gut microbiota and colonic environment are unknown and warrant further rersearch.

Currently, patients using the low FODMAP diet are usually referred to a dietitian. As the Committee acknowledged the complex nature of this dietary intervention and the need to ensure that those following it have a nutritionally balanced diet, they agreed that the diet should only be undertaken under the advice of a healthcare professional with expertise in dietary management.

The Committee noted the limitations of the evidence base. They also noted the importance of contextualising the low FODMAP diet with other diet and lifestyle interventions for IBS and the need for support by appropriate healthcare professionals. In consideration of these issues the Committee concluded that the optimal recommendation would be to supplement a current recommendation with the option of the low FODMAP diet. Therefore recommendation 1.2.1.8 was adapted to include this option.

The Committee discussed the inherent difficulties with studies that consider dietary intervention, such as the difficulties with blinding the participants. Accounting for this, the Committee agreed that in assessing the studies using GRADE, the evidence was of very low quality. In particular the Committee highlighted the comparison of habitual or standard diet and the likelihood of a lack of consistency in what this entails.

The Committee discussed that the included studies had participants who had been referred to dietitian based clinics and whether they could be considered representative of those with IBS based in primary care. It was agreed that though there may be some differences, these studies have relevance to primary care (accepting the possibility that the study participants may be those with more severe symptoms than those based in primary care).

The Committee noted the difficulties in getting funding for IBS related research in general and the importance of reviewing the low FODMAP diet as it is being discussed both professionally and within patient forums. The the Committee felt that it was important to include all of the identified trial based studies and agreed with the inclusion of the controlled study as well as the two RCT based studies.

The Committee discussed that as the low FODMAP diet is being currently used in those with IBS and that there is very limited evidence of potential benefits and harms, a research recommendation would be appropriate. The Committee considered that this should focus on areas such as patient acceptability of the low FODMAP diet, quality of life, long-term effects and consideration of the re-introduction phase of the low FODMAP intervention.

In addition, the Committee also acknowledged that the current dietary resources available for implementing the low FODMAP diet only includes a list of foods that are common in a typical western diet, and that information on culturally specific foods are very limited. Therefore, the Committee emphasized that healthcare professionals need to have an appropriate discussion with people with IBS who wish to go on a dietary intervention, particular people who consumed culturally specific foods. Full information on available food sources for low FODMAP diet needs to be provided and discussed with people with IBS so that an informed decision could be made.

804 participants meeting Rome II criteria for IBS-C. 18+.

Eligibility for randomisation: average score of ≥3 for daily abdominal pain at its worst (11 point rating scale) and an average of <3 Complete Spontaneous Bowel Movements (CSBMs) per week and ≤5 Spontaneous Bowel Movements (SBMs)/week during the baseline period (12 weeks) not necessarily consecutive, in the 12 months before the screening visit.

Mean age 44yrs, Female 90%, White 78%.

Significantly higher proportion of men in placebo arm than the linaclotide arm (12.7 vs 8.2% p=0.037).

Linaclotide 290µg orally once daily, 30 mins before breakfast.

N=401

Placebo

12 weeks and 26 weeks respectively.

1. FDA Responder (Pain ≥50% of weeks).
2. FDA Responder (Stool frequency ≥50% of weeks).
3. FDA Combined responder pain and stool frequency (≥50% of weeks)
4. FDA Pain Responder (≥30% improvement 75% of weeks)
5. FDA Combined responder Pain and stool frequency 75% of weeks
6. Constipation Responder (improvement in stool consistency ≥1 point on BSFS)
7. Bloating Responder (improvement ≥50% wks) Bloating severity (5 point scale).

800 participants

As above (Chey 2012)

Mean age 44 years, 90.5% female.

Linaclotide 290µg once daily. Timing not specified.

N=405

Placebo

1. FDA Responder (Pain ≥50% of weeks).
2. FDA Responder (Stool frequency ≥50% of weeks).
3. FDA Combined responder pain and stool frequency (≥50% of weeks)
4. FDA Pain Responder (≥30% improvement 75% of weeks)
5. FDA Combined responder Pain and stool frequency 75% of weeks
6. Constipation Responder (improvement in stool consistency ≥1 point on BSFS)
7. Bloating Responder (improvement ≥50% wks)
8. Constipation severity (5 point scale).

420 participants

18+ Rome II criteria

<3 SBMs per week and ≥1 of the following for at least 12 wks in the preceding 12 months:

1)

Straining during ≥25% of bowel movements

2)

Lumpy or hard stools during ≥25% of bowel movements

3)

Sensation of incomplete evacuation during ≥25% of bowel movements, plus

Mean score of ≥2 for abdominal (non-menstrual) pain or discomfort on 5 point scale 1=none, 5=very severe) and

Mean of <3 CSBMs and ≤6 SBMs per week.

Discontinuation of ineligible medication (e.g. anticholinergic agents, opiods).

Mean Age 44. Female 92%.

Linaclotide once daily BEFORE first meal. 290µg dose arm reported only.

N=84

Placebo

1. QOL (IBS QOL scale) >14 point change.
2. Mean change from baseline (QOL scale)
3. IBS degree of relief responders (Equivalent to EMA recommended outcome).
4. Constipation Severity

803 participants (Trial 1, Rao (2012) as above.

805 participants (Trial 2, Chey (2012) as above,

Linaclotide 290µg (as above)

Placebo

1. IBS QOL Mean change from baseline* (improvement) by week 12.
2. EMA 12-week abdominal pain/discomfort responders (Pain rated on 11 point NRS. Responder = those with an improvement of ≥30% for at least 6/12 weeks).
3. EMA 26-week abdominal pain/discomfort responders (as above but for 13/26 weeks)
4. EMA 12 week degree of relief responders
5. EMA 26-week degree of relief responders (as above but for at least 13/26 weeks)

62 patients with physician diagnosis of IBS and Rome III criteria for IBSC.

Age 18+

Baseline Characteristics: (not reported by arm)

Mean age (SD) 41.95 (13.56), 85.5% Female.

Average IBS Severity Score at baseline was 296 (95% CI 274,317).

Percentage per score category:

Mild (score<175) - 8.1%

Moderate (175–300) - 46.7%

Severe (>300) - 45.2%

Lubiprostone 48µg, one capsule twice daily.

(n=62 or 60)

Placebo

After treatment period 2

1. Life interference, mean difference
2. IBS-SS Mean difference
3. Pain (0–10 scale) Mean difference
4. Days with hard/lumpy stools or no stools (%)
5. Bloating (0–10 scale) mean difference.

Combined n= 1171

Study A n=590

Study B n=581

Rome II diagnosis of IBS-C. Age 18+.

Compliance with daily diary completion ≥70% during the 4 week baseline period. Min 2 of the following

1. <3 SBMs / week
2. At least 25% SBMs accompanied by at least moderate straining
3. At least 25% SBMs associated with stool consistency rating.

Mean Age 47years, 91.6% female.

Lubiprostone 16µg (8µg twice daily) with breakfast and dinner and with 8oz water

Study A n=390

Study B

N=379

Placebo

1. IBS QOL, mean difference
2. Overall responders (degree of relief over time)
3. Spontaneous Bowel Movements (frequency) Mean difference.

195 participants of 18–80 years old, not pregnant, not lactating.

Rome II diagnostic criteria for IBS

Rome II modular questionnaire criteria for IBS-C

Sigmoidoscopy or colonoscopy within 5 years to rule out other causes/diseases.

In 4 week initiation period

• Avoidance of disallowed medications (not specified)
• Satisfactorily complete electronic diary

Min 2 of the following

1. <3 SBMs / week
2. At least 25% SBMs accompanied by at least moderate straining
3. At least 25% SBMs associated with stool consistency rating

Lubiprostone 16, 32, 48µg per day,

Split into 8µg twice daily (n=51), 16µg twice daily (n=49) or 24µg twice daily (n=45) with breakfast and dinner and 8oz H²0.

Placebo

1. IBS-QOL mean difference
2. Spontaneous bowel movements (weekly frequency) – mean difference
3. Constipation Severity (5 point scale) mean difference.

The important outcomes were prioritised by the topic-specific members (TSMs) through ranking methods and further confirmed by the standing Committee before the review was carried out. The relative value of different outcomes was discussed, and the prioritised, important outcomes were as follows:

Quality of life, symptoms, pain, patient preferences, deterioration, stool score/change in bowel habit and relapse, flatulence or bloating.

Across the included studies (n=7), more than 50 different outcome measures/metrics were reported that were relevant to the 7 agreed important outcomes. There were also differences in the way outcomes were reported between linaclotide and lubiprostone to decide which of these were the most clinically relevant and important to patients. The TSMs were again consulted, and a total of 21 outcomes were subsequently selected. These were:-

Linaclotide

Quality of Life

1.

Quality of Life (IBS QOL Scale)

2.

Quality of Life responder (>14 point change on IBS QOL Scale)

Symptoms

3.

Improvement of ≥ 30% from baseline in average daily worst abdominal pain score 50% of the time (calculated weekly) (FDA suggested)

4.

Rate increase in stool frequency - ≥1 complete spontaneous bowel movement (CSBM) per week from baseline (FDA suggested)

5.

Combined weekly FDA responder (improvements in both pain and stool frequency) (FDA suggested)

6.

Improvement of ≥ 30% from baseline in average daily worst abdo pain score 75% of the time (calculated wkly) (FDA suggested)

7.

Combined end point defined a responder (improvements in pain and stool frequency (≥3 CBSMs and increase of ≥1 CSBM from baseline) 75% of the time (FDA suggested)

8.

12-week abdominal pain/discomfort responders (≥30% reduction in mean abdominal pain and/or discomfort (11 point scale) with neither worsening from baseline for ≥6 weeks) (EMA suggested)

9.

12-week IBS degree of relief responders (symptoms ‘considerably’ or ‘completely’ relieved for ≥6/12wks) (EMA suggested)

Stool Score/Bowel habits

10.

Constipation Severity (% with decrease in ≥ 1 point on BSFS for ≥ 50% weeks)

11.

Mean change in constipation (5 point scale, 1 = none, 5=very severe)

Relapse or flatulence or bloating

12.

Abdominal bloating (% of patients with ≥30% decrease in discomfort for ≥50% of weeks)

Lubiprostone

Quality of Life

13.

Quality of Life (IBS QOL Scale)

14.

Life interference (0–10 scale)

Symptoms

15.

IBS Symptom severity (Score out of 500)

16.

Overall Relief Responder Status (based on reported 7-point relief scale)

Pain

17.

Pain (0–10 scale)

Stool Score/Bowel habits

18.

Spontaneous Bowel Movements (SBMs) (Frequency)

19.

Constipation severity (5 point scale) (0= absent, 4=very severe)

20.

Stool output (days with hard/lumpy stools or no stools %)

Relapse or flatulence or bloating

21.

Bloating (0–10 point scale)

The Committee was advised of the agreed sub outcomes prior to further analysis.

As the FDA and EMA suggested outcomes included individual and composite outcomes for pain and stool frequency, these were reported once under the outcome ‘symptoms’ and not separately under ‘pain’ or ‘stool frequency’.

The FDA and EMA recommended that clinical relevance for continuous outcomes should be considered as ≥30% improvement. This figure was therefore used to assess clinical relevance for each continuous outcome.

The Committee considered the recent EMA recommendations not to use ‘overall relief’ as an outcome measure, and thus decided that this information weakened the suggested benefit of lubiprostone.

The Committee advised that changes to stool consistency should equate to a minimum of two points on the BSFS to be clinically important.

There was variation in how outcomes were reported for both drugs and several composite outcomes reported. For all included studies, the reviewer had to back-calculate statistics to obtain results of sufficient quality before evaluation could begin.

Linaclotide

Effects favouring linaclotide vs. placebo were both statistically and clinically significant in 7 out of 12 selected clinical outcomes. Six of these 7 outcomes included a meta-analysis of at least 2 studies. Quality ratings were moderate (3 pooled outcomes), low (2 pooled outcomes) and very low (1 pooled outcome). One outcome included 3 individual studies that could not be pooled (moderate to very low quality).

Moderate, low or very low quality evidence suggests that linaclotide may improve quality of life, stool frequency, combination of pain and stool frequency, degree of relief, constipation and bloating in people with IBS-C.

Potential confounders cannot be excluded. Use of rescue medication (other laxatives), use of concomitant laxatives (bulk forming and stool softeners), use of other medications e.g. anti-depressants, anti-spasmodics and analgesics, dietary fibre modification, fluid intake and exercise levels were not reported by study arm, leading to concerns about drug efficacy. As such the overall evidence quality was rated down due to risk of bias. The Committee acknowledged that we could not be sure whether it was the study drug, the rescue medication or concomitant medication, or the combinations of all of these that had a positive effect on the outcomes.

The efficacy of linaclotide was discussed in detail, taking into account the evidence quality and the risk of bias for the above reasons. It was acknowledged that many people who have tried multiple laxatives without adequate symptom relief over a period of time. The Committee discussed and agreed that people with IBS with constipation who have not achieved symptoms relieve after 12 months of optimnal or maximum doses of conventional laxativeswarrant a trial of linaclotide. The 12 months period was based on the trials’ baseline entry information (Chey 2012; Rao 2012; Johnston 2010). Therefore, the Committee decided that a weak recommendation would be appropriate for this drug, taking into account individual patient symptoms (severity and duration) and previous treatment options that may not have induced sufficient or long-lasting relief. The Committee also agreed that follow-up was very important for this subgroup of patients and they should be reviewed every 3 months to assess the efficacy of linaclotide.

Lubiprostone

Effects favouring lubiprostone vs. placebo were both statistically and clinically significant in only 2 out of 9 selected important outcomes (overall relief and spontaneous bowel movements) and the evidence rating for these outcomes was low. One of these outcomes was reported by 2 studies - 1 small study was both statistically and clinically significant, the other larger study was not.

As the evidence for efficacy of lubiprostone was low quality the Committee decided there was insufficient evidence to warrant a recommendation for the use of lubiprostone in this update.

Linaclotide

Benefits of linaclotide identified in the evidence review were improvements in quality of life, stool frequency, combination of pain and stool frequency,degree of relief, constipation and bloating. The outcome quality ratings were from moderate to very low.

Diarrhoea was the only adverse event that was both statistically and clinically worse in the linaclotide arm. The Committee acknowledged diarrhoea is an adverse event common to all laxatives. CG61 recommends dose titration and monitoring of laxatives according to clinical response. Taking multiple laxatives from different classes could contribute to polypharmacy which may be undesirable for some people and therefore affect adherence to prescribed doses.

Lubiprostone

Benefits of lubiprostone were more uncertain than those for linaclotide and were limited to an improvement in overall relief only (low quality evidence).

Nausea was more likely in the lubiprostone vs. placebo arms and this could be considered by some people as a minor harm, although nausea is also a potential consequence of constipation.

The Committee decided that the predominantly female population across all the included studies (86–92%) was reasonable as it reflected the epidemiology of the IBS population.

The Committee discussed the timing and setting for potential prescribing of linaclotide. It was agreed that prescribing recommendations should not be limited to secondary care, and that other conventional laxative classes recommended by the original guideline should be tried first, taking individual patient preferences into account.

The Committee further discussed whether a research recommendation was required for this topic. They agreed that further efficacy trials for linaclotide were not necessary. Regarding lubiprostone, as it is off-label for treating IBS and there is already a licensed and indicated alternative (linaclotide) for which a positive recommendation has been made, the Committee felt that a further research recommendation on this was not necessary.

The important outcomes were prioritised by the topic-specific members (TSMs) through ranking methods and further confirmed by the standing Committee before the review was carried out. The Committee discussed the outcomes data and agreed that patient’s quality of life is the most important outcome as IBS is a chronic condition. The Committee also agreed that the use of the IBS-QoL scale for this outcome in the evidence was appropriate as it has been validated and used widely in practice and research.

The Committee discussed the importance of assessing the magnitude of improvement from baseline for the quality of life outcome (mean change from baseline scores) rather than just focussing on the difference between treatment groups (mean difference at endpoint).

The Committee noted that outcomes for IBS symptoms (as reported using the GSRS-IBS scale and the ISB-SS scale) were not useful in evaluating effectiveness of psychological interventions. This is because the aim of psychological interventions is to equip people with skills and techniques to manage their IBS symptoms better in the long-term to improve their quality of life overall. They are not aimed at reducing IBS symptoms.

The Committee agreed that the quality of evidence was mostly of low to very low quality due to a number of factors. All of the included studies have unclear baselines regarding any concomitant treatments for IBS; the study populations of all included studies, apart from 1 (Gaylord 2011), were self-referred, which was subject to selection bias; most included studies did not report reasons for withdrawal or lost to follow-up; finally the definition of the comparator in 2 included studies (Zernicke 2012, Treatment as usual) and (Hunt 2009, Waitlist control) was unclear.

The Committee also discussed the directness of the 6 included publications of 4 studies (3 of which were multiple publications of the same research) on the Computerised Cognitive Behavioural Therapy with Mindfulness and Exposure principles (CCBT-Mindfulness/Exposure) as these were conducted in Sweden. The Committee considered and agreed that the procedures of this particular intervention may not be applicable to UK setting for a number of reasons. The intervention package was in Swedish, and that translating the online materials into English may not be practical and there may be uncertainty around its effectiveness when delivered in different languages. Moreover, in this particular CCBT-Mindfulness/Exposure intervention, the participants have online access to a therapist or psychologist to gain detailed one-to-one advice, which was different to how the CCBT programme (for depression) was delivered in the UK.

The Committee moved on to discuss the 1 included study (Hunt 2009) on Computerised Cognitive Behavioural Therapy with Exposure principles (CCBT-Exposure) and another included study on Mindfulness-based stress reduction (MBSR) (Zernicke 2012). As the definition of the comparator in these 2 studies (waitlist control and treatment as usual, respectively) was unclear, the Committee agreed that the quality of the evidence was of very low quality and there was high uncertainty of the results reported in these 2 studies.

Finally, the Committee agreed that the evidence on Mindfulness group training was very limited (1 small study) and of very low quality due to the unclear baseline and reporting issues on reasons for withdrawal and lost to follow-up.

The Committee discussed the potential benefits of the psychological interventions where only limited evidence was identified.

CCBT-Mindfulness/Exposure:

The Committee acknowledged that the evidence suggested some benefits on quality of life and IBS symptoms at 10-week post treatment, but it failed to illustrate longer-term benefit (12-month follow-up). This could be due to potentially unsustainable benefits of the intervention or due to participants from the comparison group crossing over to the treatment arm after the 10-week treatment period.

Also, the limited evidence for this intervention was from the same study (with multiple publications across different time points) carried out in Sweden. As discussed above, the Committee agreed that currently there is still insufficient evidence to recommend such complex intervention in the UK.

A member of the Committee commented that, Ljotsson 2014 was a study of ‘mechanisms’ where a complex intervention that has found to be effective was “dismantled” to investigate the effectiveness of each component part. As such, it is not an efficacy or effectiveness study comparing intervention with a control.

CCBT-Exposure, MBSR and Mindfulness group training 1 very small study (n=31) suggested a small benefit at 6-weeks on quality of life and IBS symptoms without any further longer term data. The Committee agreed that currently there is still insufficient evidence to recommend CCBT-Exposure.

1 small study (n=75) on Mindfulness group training failed to illustrate benefits on quality of life for both the 10-week and 3-month follow-up time points. The Committee again agreed that currently there is still insufficient evidence to recommend Mindfulness group training.

Finally, 1 small study (n=75) on MBSR illustrated small benefit on the quality of life outcome at 8-week time-point, but the study failed to illustrate benefits on both quality of life and IBS symptoms at 6-month follow-up. The Committee again agreed that currently there is still insufficient evidence to recommend MBSR.

The Committee noted that due to the nature of psychological interventions, there was unlikely to be any treatment-related adverse effects.

No economic evaluations of mindfulness were identified. The Committee considered that mindfulness is usually delivered as part of cognitive behavioural therapy and therefore unlikely to involve any substantial impact on resource use.

No economic evaluations of computer-based cognitive behavioural therapy were included in the review of cost-effectiveness. The Committee considered that CCBT is likely to cost less than other psychological interventions.

The Committee acknowledged that although there is currently insufficient research evidence to recommend CCBT and Mindfulness therapy for the management of IBS, Mindfulness therapy has become increasingly popular in private practice, and widely available and free or commercial self-help websites. The Committee felt strongly that urgent UK-based good quality research on Mindfulness therapy is crucial to provide good quality accurate research data to inform both healthcare professionals and patients regarding the effectiveness of such interventions, so that appropriate standards and recommendations could be made for the NHS.

Therefore, the Committee agreed that a research recommendation should be made for investigating the effectiveness of Mindfulness therapy.