Association of participation in a mindfulness programme with bowel symptoms, gastrointestinal symptom-specific anxiety and quality of life


Dr D. J. Kearney, Seattle VAMC 111GI, 1660 S. Columbian Way, Seattle, WA 98108, USA.


Aliment Pharmacol Ther 2011; 34: 363–373


Background  Stress perception and GI-specific anxiety play key roles in irritable bowel syndrome (IBS). Mindfulness-based stress reduction (MBSR) is a widely available stress reduction course, which has not been evaluated for IBS.

Aim  To determine whether participation in MBSR is associated with improvement in bowel symptoms, GI-specific anxiety, and IBS-Quality of Life.

Methods  This is a prospective study of 93 participants in MBSR. We applied measures of Rome III IBS status, bowel symptoms (IBS-Severity Scoring System, IBS-SSS), IBS-Quality of Life (IBS-QOL), GI-specific anxiety (Visceral Sensitivity Index, VSI), mindfulness (Five Facet Mindfulness Questionnaire-FFMQ), and functional status (SF-8) at baseline and 2 and 6 months after enrolment.

Results  At 2 months, participation in MBSR was associated with small nonsignificant changes in IBS-SSS, IBS-QOL and VSI: d = −0.25, d = 0.08, d = −0.16, respectively. At 6 months, there was no significant change in IBS-SSS (d = −0.36); whereas for IBS-QOL and VSI there were significant improvements (IBS-QOL: d = 0.33, = 0.044; VSI: d = −0.40, = 0.014). For patients meeting Rome III IBS criteria (= 43), changes in IBS-SSS, IBS-QOL and VSI were not statistically significant, but there was a significant correlation between the change in VSI and the change in FFMQ across the three time periods (r = 0.33).

Conclusions  Participation in MBSR is associated with improvement IBS-related quality of life and GI-specific anxiety. Randomised controlled trials are warranted to further assess the role of MBSR for IBS symptomatology.


Irritable bowel syndrome (IBS) is a common, symptom-based syndrome estimated to affect approximately 10–15% of the adult population of North America.1 It has been estimated that IBS accounts for 12% of caseloads in primary care, and the total direct and indirect monetary costs of IBS are enormous.2–5 IBS, as defined by the Rome criteria, is characterised by chronic abdominal pain or discomfort associated with a change in bowel habits.6 The Rome criteria provide a standardised method of defining IBS for research purposes, but many patients who suffer from chronic gastrointestinal symptoms do not meet strict Rome criteria for IBS.7, 8 Symptoms of IBS are a source of significant distress for patients, which results in a reduction in quality of life across multiple domains of functioning when compared with the general US population.9–14 Severe IBS has a health utility similar to stage III congestive heart failure and rheumatoid arthritis, and when health status for patients with IBS is compared with patients with diabetes mellitus or end-stage renal disease, IBS patients score lower on several domains of health.15, 16

Recommended treatments for IBS are symptom-focused and include dietary, pharmacological and psychosocial interventions. Some patients with mild symptoms may respond to education, dietary and lifestyle modifications, whereas patients with moderate or severe symptoms are usually treated with pharmacological or psychological interventions.8, 17 For patients with constipation-predominant symptoms, a trial of increased dietary fibre or a stool softening regimen is often recommended, whereas for patients with diarrhoea-predominant symptoms, loperamide or other agents that slow intestinal motility are often prescribed. For patients with abdominal pain as a prominent symptom, antispasmodics or low-dose antidepressants are typically recommended.8, 17 Psychological treatments, such as cognitive behavioural therapy (CBT) or hypnotherapy, are sometimes recommended when symptoms are refractory to treatment. The effectiveness of cognitive behavioural therapy for IBS is diminished when there are high levels of baseline depression or anxiety.18–20 Overall, the treatment options for IBS are generally unsatisfactory; approximately one-third of patients do not achieve adequate response with existing treatments.8 The lack of patient satisfaction with existing approaches is reflected by findings that 16–33% of IBS out-patients seek alternative medicine consultation for their symptoms.21–24 Given the potential for poor health status relative to other chronic illness and the significant increases in health care utilisation attributable to IBS, we are in need of additional treatment modalities for IBS.2–5, 9–14

There is evidence that the impairment in quality of life in IBS is not primarily caused by the severity or type of bowel symptoms a person experiences. Rather, reduced quality of life in IBS is largely related to extraintestinal factors, including symptoms of chronic stress (e.g. tiring easily, feeling nervous, feeling hopeless, difficulty sleeping), the belief that there is something seriously wrong with the body, and interference with sexual function.25 Patient-reported IBS illness severity, a construct related to but distinct from quality of life, has also been shown to be significantly related to the belief that the GI symptoms represent a serious underlying problem with the body.26 Persons with IBS demonstrate increased anxiety regarding gastrointestinal sensations (termed GI-specific anxiety); this has been shown to play a role in disease severity and is a key determinant of whether a person meets diagnostic criteria for IBS.27 GI-specific anxiety includes the thoughts, emotions and behaviours that stem from fear of GI sensations, symptoms, and the context in which these occur. GI-specific anxiety has been hypothesised to function as an endogenous stressor, which leads to exaggerated autonomic and neuroendocrine responses, changes in intestinal function and visceral pain perception and GI symptoms.27, 28 A situation that might produce GI-specific anxiety is being someplace where the location of bathroom facilities is not known, such as restaurants or social gatherings. Cognitions accompanying GI symptoms might include beliefs of poor ability to control symptoms, or beliefs that the GI sensations experienced represent a serious underlying illness. These contexts and cognitions can lead to fear, worry and avoidance behaviours, which can be associated with reduced quality of life. Persons with IBS also have heightened stress perception, which adversely affects symptoms29–32 and hypervigilance towards bodily sensations and symptoms.33, 34 The key roles of stress, GI-specific anxiety, hypervigilance towards bodily sensations, and extraintestinal factors in quality of life suggest that providers would improve treatment outcomes by teaching methods to reduce anxiety and stress, rather than solely focusing on managing specific bowel complaints (such as the frequency or consistency of bowel movements).25

Prior research on stress reduction techniques for IBS shows that improvement in IBS symptoms correlates with improvement in psychological symptoms in most trials.1 However, prior studies of stress reduction for IBS (most of which were performed approximately two decades ago) combined multiple behavioural therapies (e.g. relaxation therapy, stress management and biofeedback) as part of a multicomponent package, making it difficult to interpret the factor(s) responsible for any improvement in outcomes.35–42 One validated and increasingly utilised method to reduce stress is through teaching mindfulness. Mindfulness emphasises attentiveness to present moment experience, and has been defined as ‘the awareness that emerges by way of paying attention on purpose, in the present moment, and nonjudgmentally to the unfolding of experience moment by moment’.43 In mindfulness practice, thoughts, bodily sensations and emotions are regarded as objects of attention. In the context of IBS, enhanced mindfulness would be hypothesised to decrease GI-specific anxiety by teaching techniques that foster reappraisal of fears, thoughts, and beliefs related to GI sensations.

A validated clinical method of teaching mindfulness is an 8-week class series called mindfulness-based stress reduction (MBSR),44 which is widely available throughout the US and internationally. Prior studies indicate that mindfulness skills are increased through participation in MBSR and other similar training programmes.45 At a minimum, most mindfulness training programmes include instruction, scheduled home practice, as well as informal mindfulness practices.45, 46 Recent studies by a group in Sweden have evaluated a mindfulness-based programme, which also included exposure therapy, tailored to individuals with IBS and found that it was associated with clinically significant improvement in IBS symptoms and quality of life.47, 48 Changes in mindfulness were, however, not assessed. We sought to examine the association of a widely available 8-week MBSR programme with measures of bowel-related symptoms, GI-specific anxiety and quality of life. Given the high prevalence of post-traumatic stress disorder (PTSD) in our Veteran patient population, we also sought to explore whether outcomes differed according to whether a person had symptoms compatible with PTSD, which has clinical hallmarks of hypervigilance and avoidance, and might be expected to influence the impact of the intervention. We hypothesised that participation in MBSR would be associated with improvement in measures of bowel symptoms, IBS-related quality of life, gastrointestinal symptom-specific anxiety and health status.


We performed a before-and-after study of 93 Veterans who participated in MBSR classes at VA Puget Sound Health Care System. All recruitment and study procedures were approved by the institutional review board and Research and Development committees of VA Puget Sound Health Care System (PSHCS) in Seattle, WA. Subjects gave written informed consent prior to enrolment. No monetary compensation was provided to study participants and no fee was charged for participation in MBSR. This material is the result of work supported by resources from the VA Puget Sound Health Care System, Seattle, WA, USA. This trial was registered at and has the identifier: NCT00880256.


During a 17-month period, we enrolled 93 Veterans who were referred or self-referred themselves to the MBSR programme at VA Puget Sound. In our setting, MBSR is available to all patients as an adjunct to their usual clinical care for a variety of mental and physical health conditions. All patients who were eligible for clinical participation in MBSR were eligible for this study. Exclusion criteria for clinical participation in MBSR included: (i) any past or present psychotic disorder; (ii) mania, or poorly controlled bipolar disorder; (iii) borderline or antisocial personality disorder; (iv) current suicidal or homicidal ideation; (v) active substance abuse or dependence.


Self-report measures were obtained from study participants at three time points: at baseline, at 2 months (immediately post-MBSR) and at 6 months (4 months after completion of MBSR). We recorded demographic characteristics at baseline using a written questionnaire.

IBS symptoms.  The Irritable Bowel Severity Scoring System (IBSSS) provides a measure of the severity of IBS. The measure consists of five questions, which assess severity of abdominal pain, number of days with abdominal pain in past 10 days, severity of abdominal distension, satisfaction with bowel habits and impact of IBS on life in general. The score on each of the five questions ranges from 0 to 100, and the scores are summed with a range of total possible scores from 0 to 500. Higher scores reflect more severe IBS.49 The instrument has been shown to be reliable, valid and responsive to treatment effects.49

IBS-Quality of Life.  The IBS-QOL is a disease specific Quality-of-Life Measure for IBS. 50 IBS-QOL has been shown to have a high level of content validity and to be responsive to change and has been used in several outcome studies and clinical drug trials throughout the world. It consists of 34 questions that assess the influence of bowel habits on daily life. The response to each question is rated on a 5-point scale. A lower score indicates worse bowel-related quality of life.

Gastrointestinal symptom-specific anxiety.  The Visceral Sensitivity Index (VSI) is a 15-item self-report scale that measures GI-specific anxiety.51, 52 The VSI measures the fear, anxiety and hypervigilance that often accompany perceptions and misappraisals of bodily sensations. The VSI demonstrates high levels of convergent validity with other anxiety scales including the anxiety subscale of the Hospital Anxiety and Depression Scale (HAD) and the Anxiety Sensitivity Index (ASI).27 A higher score indicates greater GI-specific anxiety.

IBS diagnostic status.  Symptom-based diagnostic criteria for IBS were defined by having subjects complete a written version of the Rome III questionnaire. The Rome III criteria are considered the gold standard criteria for IBS diagnostic status.53

Health status.  The SF-8 measures health status and health-related quality of life. Both the Physical and Mental Component Summary Scores (PCS and MCS) were calculated.54 The Mental Component Summary (MCS) of the SF-8 integrates measures of functional status that span eight domains of health, with greater weight given to mental domains (social functioning, role limitations because of emotional problems and mental health) than physical domains (physical functioning, role limitations because of physical problems, bodily pain, general health and vitality), whereas the Physical Component Summary Score (PCS) gives greater weight to the physical domains than the mental domains. Higher PCS and MCS scores reflect greater quality of life.

PTSD symptoms.  The PTSD checklist (PCL) is a 17-item self-report measure that evaluates PTSD diagnostic status and symptom severity.55, 56 Each question is rated by the patient on a scale of 1–5. The PCL has good internal consistency, test-retest reliability and discriminant validity and convergent validity.57 The total score of the PCL is calculated as the sum of the all the items. Higher scores indicate more severe PTSD. Values of 38 for women and 44 for men were used as cut-off scores for identifying those who screened positive for a likely diagnosis of PTSD.55, 58

Mindfulness.  The Five Facet Mindfulness Questionnaire (FFMQ)45 measures mindfulness skills. The FFMQ is a 39-item measure comprised of subscales for nonreactivity to internal experience, observing internal experience, acting with awareness, describing internal experience and nonjudgment of experience. Psychometric analyses of this measure show that it has adequate convergent validity and incremental validity in the prediction of psychological symptoms.45 Higher scores on the FFMQ reflect greater mindfulness.


Subjects underwent an 8-week mindfulness course (mindfulness-based stress reduction – MBSR). All participants continued to receive their usual care from their primary care and other providers during the study. The MBSR courses were held at the Seattle VA Medical Center and taught by instructors who meet professional guidelines for teaching MBSR and have extensive prior experience in teaching MBSR. The intervention closely followed the standardised MBSR format developed by the University of Massachusetts Medical School.59 MBSR groups included Veterans with a wide variety of mental and physical health problems, including IBS, post-traumatic stress disorder (PTSD), general anxiety, chronic pain, depression, functional limitations because of physical disability and other health issues. There were approximately 5–10 study participants in each MBSR group of 20–30 male and female Veterans. There was not a specific focus on teaching mindfulness skills related to IBS symptoms or GI-specific anxiety as part of the MBSR course. Instead, mindfulness was taught as a way of being that encompasses all aspects of experience, including thoughts, emotions and bodily sensations (including GI sensations). Participants met once per week (2.5 h per session) for 8 weeks. During each meeting, participants practised mindfulness meditation and yoga, received instructions from the teacher, and discussed homework assignments. Between the sixth and seventh week, participants met for 7 h for a daylong session, to practice mindfulness exercises more intensively.

The mindfulness instructions in the class exercises, discussions and homework assignments emphasised intention, attention and attitude.60 Intention, as taught in MBSR, typically involved asking participants to bring to mind their personal motivation for participating in MBSR. For example, participants might bring to mind their intention for improved pain control, the ability to develop close relationships with others or an increased ability to deal with anger. The second area of instructions in all class sessions involved attention. In MBSR, participants are taught how to bring sustained attention to an aspect of their experience (e.g. a thought, emotion or bodily sensation), and they are also taught flexibility of attention. An attitude or quality of attention is also emphasised in MBSR. This quality or attitude is one of openness, kindness, curiosity and nonjudging of present-moment experience, including unpleasant or difficult experiences. These attentional abilities are taught through meditation exercises including the ‘body scan’ (a 45 min exercise in which attention is systematically directed throughout the body), sitting meditation (an exercise in which attention is brought to the sensation of the breath, with redirection of attention back to the breath when the mind wanders) and gentle yoga (taught as a practice of mindful movement, intended to develop present moment awareness of bodily sensations). The flexibility of attention taught in these exercises has been shown to foster the ability to ‘let go’ of ruminative cycles of thought, which is considered a key factor in relapse of depression.61 Homework was assigned as daily meditation or yoga for 45 min per day, 6 days per week, using CDs as a guide. Informal homework practices were also assigned, which involved bringing mindful attention to experiences in daily life (e.g. eating a meal or noticing a pleasant or unpleasant experience).

Statistical analyses

We compared scores at baseline, 2-month and 6-month time points, using independent sample t-tests, so as to not exclude the baseline information for patients who did not complete follow up. A two-sided P value of less than 0.05 was considered statistically significant. The standardised mean difference (Cohen’s d effect size) from baseline to 2-months, and baseline to 6-months was calculated for each variable. In addition, we used linear mixed-effects models to model change in outcomes over the three time periods. Mixed-effects models account for within subject correlation of responses over time, and allow use of all available data across all time points to increase statistical efficiency. To preliminarily evaluate whether changes in mindfulness play a role in mediating changes in key outcome variables, the overall postintervention correlation between change in mindfulness and change in bowel symptoms and other clinical outcome measures (IBS-SSS, VSI and IBS-QOL) was examined using linear mixed-effects models that incorporate all scores over time. Analyses were carried on an intent-to-treat basis, which included all participants regardless of their level of compliance with the MBSR intervention. In addition, we analysed patients meeting Rome III criteria for IBS at baseline stratified by baseline PTSD (as diagnosed by the PCL).


The patient population was predominantly male (75%) and most participants were Caucasian (85%). The mean age of participants was 51 ± 10.6 years. Of the 93 persons enrolled, 76 (82%) completed the 2-month measures and 65 (70%) completed 6-month measures. The mean number of class sessions attended was 5.7 ± 2.83, of nine possible classes, and 69 of 93 patients (74%) attended at least four class sessions, which we considered the minimum number of classes attended to be considered compliant.62 Veterans who did not attend at least four classes did not differ significantly from those who attended at least four classes on gender, age, ethnicity, PTSD severity and mindfulness (data not shown).

Forty three (46%) participants met Rome III IBS diagnostic criteria at baseline, 28 (30%) at 2 months and 20 (22%) at 6 months. When the 43 patients who met Rome III IBS criteria at baseline were followed over time, 24 (56%) met Rome III IBS criteria at 2 months and 15 (35%) met Rome III criteria at 6 months (< 0.001). As evidenced in Table 1, we found no significant differences in patient characteristics between patients with and without Rome III IBS criteria at baseline.

Table 1.   Patient baseline characteristics (= 92)*
CharacteristicNo IBS N (%) (= 49)Baseline IBS N (%) (= 43)
  1. There are no significant differences between no IBS and IBS groups.

  2. * One patient was missing Rome III baseline data.

Male35 (71)34 (79)
Age (mean ± s.d.)52 ± 10.649 ± 10.3
Ethnicity (= 91)
 White38 (81)38 (88)
 Black3 (6)1 (2)
 Hispanic4 (9)2 (5)
 Asian/PI/NA1 (2)1 (2)
 Other1 (2)1 (2)
 Christian28 (57)22 (51)
 Buddhist1 (2)1 (2)
 Other2 (4)4 (9)
 Unknown18 (37)16 (37)
Living situation (= 90)
 Own/rent43 (92)38 (91)
 Homeless3 (6)3 (7)
 Other1 (2)1 (2)
Classes attended (mean ± s.d.)5.8 ± 3.025.6 ± 2.66
 0 classes5 (10)2 (5)
 1–3 classes9 (18)8 (19)
 4–6 classes3 (6)5 (12)
 6–9 classes32 (65)28 (65)
PTSD at baseline35 (71)32 (74)

Table 2 shows the scores for outcome measures at each assessment point compared with baseline. At 2 months, participation in MBSR was associated with small nonsignificant changes in IBS-SSS, IBS-QOL and VSI (d = −0.25, d = 0.08, d = −0.16 respectively). At 6 months the magnitude of change was somewhat larger but continued to be nonsignificant for IBS-SSS (d = −0.36); while for IBS-QOL and VSI the magnitude of change was significant (IBS-QOL: d = 0.33, = 0.044; VSI: d = −0.40, = 0.014). The direction of change in values for IBS-SSS, IBS-QOL and VSI were all in the direction associated with clinical improvement. There was a large change in FFMQ at 2 months (d = 0.68, < 0.001) and at 6 months (d = 0.77, < 0.001). We found no significant improvement in physical functioning (PCS) over time (2 months d = 0.18, 6 months d = 0.26). We did, however, find a large improvement in mental functioning (MCS) (2 months d = 0.60, < 0.001; 6 months d = 0.72, < 0.001).

Table 2.   Mean summary scores and Cohen’s d effect sizes with 95% CI for all patients (= 93)
Summary scoresBaseline mean ± s.d.2 months mean ± s.d.6 months mean ± s.d.Standardised mean difference: baseline to 2 months (95% CI)Standardised mean difference: baseline to 6 months (95% CI)
  1. IBS-SSS, IBS-Severity Scoring System; IBS-QOL, IBS-Quality of Life; VSI, Visceral Sensitivity Index; FFMQ, Five Facet Mindfulness Questionnaire; PCS, physical component summary; MCS, mental component summary.

IBS-SSS256.9 ± 90.6233.5 ± 107.0224.1 ± 96.3−0.25 (−0.80 to 0.30)−0.36 (−1.1 to 0.36)
 P value N.S.N.S.  
IBS-QOL84.4 ± 20.686.0 ± 20.490.7 ± 16.90.08 (−0.23 to 0.39)0.33 (0.01 to 0.65)
 P value N.S.0.044  
VSI20.2 ± 18.617.1 ± 19.013.0 ± 16.8−0.16 (−0.47 to 0.15)−0.40 (−0.73 to −0.08)
 P value N.S.0.014  
FFMQ108.2 ± 25.2126.1 ± 27.5127.9 ± 26.10.68 (0.37 to 1.00)0.77 (0.44 to 1.10)
 P value <0.001<0.001  
PCS39.7 ± 10.741.7 ± 11.142.5 ± 10.40.18 (−0.13 to 0.48)0.26 (−0.06 to 0.58)
 P value N.S.N.S.  
MCS33.2 ± 10.640.0 ± 12.341.3 ± 12.20.60 (0.29 to 0.91)0.72 (0.39 to 1.04)
 P value <0.001<0.001  

Table 3 shows comparisons of mean scores and effect sizes limited to 43 patients meeting Rome III IBS criteria at baseline. While there was a range of effects sizes, none of the follow-up changes in IBS-SSS, IBS-QOL and VSI were significant (IBS-SSS, 2 months: d = −0.19, 6 months: d = −0.28; IBS-QOL, 2 months: d = 0.04, 6 months: d = 0.33; VSI 2 months: d = −0.62, 6 months: d = −0.40). The effect size for VSI at 6 months (d = −0.40) had a 95% confidence interval of −0.88 to 0.08, and a P value of 0.104. Change in FFMQ was large at 2 months (d = 0.75, = 0.001) and the effect was maintained through 6 months (d = 0.76, = 0.003). Improvements in physical functioning were not significant at 2 months (d = 0.34) but became so at 6 months (d = 0.49, = 0.046). We found the opposite pattern of results for mental functioning – the medium effect size of MCS was significant at 2 months (d = 0.53, = 0.021) and nonsignificant at 6 months (d = 0.44).

Table 3.   Mean summary scores and Cohen’s d effect sizes with 95% CI for patients meeting ROME III IBS criteria (= 43)
Summary scoresBaseline mean ± s.d.2 months mean ± s.d.6 months mean ± s.d.Standardised mean difference: baseline to 2 months (95% CI)Standardised mean difference: baseline to 6 months (95% CI)
  1. IBS-SSS, IBS-Severity Scoring System; IBS-QOL, IBS-Quality of Life; VSI, Visceral Sensitivity Index; FFMQ, Five Facet Mindfulness Questionnaire; PCS, physical component summary; MCS, mental component summary.

IBS-SSS261.2 ± 88.3243.2 ± 10.7236.3 ± 93.5−0.19 (−0.79 to 0.41)−0.28 (−1.04 to 0.48)
 P value N.S.N.S.  
IBS-QOL75.6 ± 2376.5 ± 24.383.0 ± 21.80.04 (−0.41 to 0.49)0.33 (−0.15 to 0.81)
 P value N.S.N.S.  
VSI29.4 ± 18.325.7 ± 20.927.8 ± 20.2−0.62 (−0.64 to 0.25)−0.40 (−0.88 to 0.08)
 P value N.S.N.S.  
FFMQ109.1 ± 23.3128.2 ± 28.2128.6 ± 29.40.75 (0.29 to 1.21)0.76 (0.26 to 1.25)
 P value 0.0010.003  
PCS36.3 ± 8.739.4 ± 9.740.7 ± 90.34 (−0.11 to 0.78)0.49 (0.01 to 0.98)
 P value N.S.0.046  
MCS34.2 ± 9.439.8 ± 11.539.4 ± 14.50.53 (0.08 to 0.98)0.44 (−0.04 to 0.92)
 P value 0.021N.S.  

Table 4 shows correlations of changes in FFMQ with changes in each of the other outcome measures among all patients and among patients with Rome III IBS criteria. Change scores were calculated both as the follow-up time minus baseline, as well as the change in the fitted value of a linear mixed model incorporating information from all three time periods. Among all patients, change in MCS was significantly correlated with change in FFMQ for each change score and across all three time periods (2 months = 0.44, 6 months = 0.61, over three time periods = 0.72). When looking at only the IBS patients, again, the change in MCS scores were significantly correlated with change in FFMQ (2 months = 0.54, 6 months = 0.73 and over the three time periods = 0.66). In addition, there was a significant correlation of the change in VSI and the change in FFMQ across the three time periods (= 0.33).

Table 4.   Correlations between change in FFMQ and change in other measures
 Correlations with FFMQ in all ptCorrelations with FFMQ in pt with IBS
2 month change score6 month change scoreOver 3 time periods2 month change score6 month change scoreOver 3 time periods
  1. FFMQ, Five Facet Mindfulness Questionnaire; IBS-SSS, IBS-Severity Scoring System; IBS-QOL, IBS-Quality of Life; VSI, Visceral Sensitivity Index; PCS, physical component summary; MCS, mental component summary.

  2. < 0.05.


For IBS patients, we also examined outcomes according to whether the patient met criteria for probable PTSD at baseline (Table 5). Thirty two of 43 (74%) patients meeting Rome III IBS criteria at baseline also met criteria for probable PTSD at baseline. The change in IBS-SSS was small and nonsignificant for patients with and without baseline PTSD. For IBS-QOL and VSI, there were significant changes at 6 months for patients without baseline PTSD (IBS-QOL, d = 1.13; VSI, d = −1.44), whereas for patients with baseline PTSD changes in IBS-QOL and VSI were not significantly different at 6 months.

Table 5.   Mean summary scores and Cohen’s d effect sizes with 95% CI for IBS patients meeting ROME III IBS criteria (= 43), according to PTSD status at baseline
Summary scoresBaseline mean ± s.d.2 months mean ± s.d.6 months mean ± s.d.Standardised mean difference: baseline to 2 months (95% CI)Standardised mean difference: baseline to 6 months (95% CI)
  1. IBS-SSS, IBS-Severity Scoring System; IBS-QOL, IBS-Quality of Life; VSI, Visceral Sensitivity Index.

  2. = 32 patients met criteria for probable PTSD at baseline.

  3. = 11 patients did not meet criteria for probable PTSD at baseline.

 PTSD262.9 ± 82.2260.9 ± 104.2239.3 ± 82.0−0.02 (−0.72 to 0.68)−0.29 (−1.15 to 0.58)
  P value N.S.N.S.  
 No PTSD255.7 ± 114.9197.3 ± 127.5226.0 ± 171.1−0.49 (−1.78 to 0.80)−0.24 (−1.84 to 1.37)
  P value N.S.N.S.  
 PTSD75.1 ± 23.970.8 ± 26.477.6 ± 24.6−0.17 (−0.70 to 0.35)0.10 (−0.47 to 0.67)
  P value N.S.N.S.  
 No PTSD77.1 ± 20.890.1 ± 7.594.5 ± 4.40.81 (−0.13 to 1.75)1.13 (0.15 to 2.10)
  P value 0.0960.025  
 PTSD29.7 ± 19.330.4 ± 22.427.5 ± 21.30.03 (−0.49 to 0.55)−0.11 (−0.68 to 0.46)
  P value N.S.N.S.  
 No PTSD28.4 ± 15.314.2 ± 11.39.6 ± 10.1−1.05 (−2.00 to −0.11)−1.44 (−2.46 to −0.41)
  P value 0.0300.006  


Our study found that patients who participated in an 8-week mindfulness intervention experienced significant improvements in IBS-related quality of life and GI-specific anxiety over a 6-month period. Participants also reported a significant increase in mindfulness skills over the course of the study, suggesting that the MBSR class series successfully taught mindfulness skills, which are postulated to account for the improvements in anxiety and quality of life measures. We found that changes in mindfulness skills were correlated with improvements in GI-specific anxiety among those meeting Rome IBS criteria. Moreover, there was a significant decrease in the percentage of subjects meeting Rome IBS criteria over 6-months follow-up. We also found that improvements in clinical outcome measures increased slightly from the time point when subjects completed the mindfulness training class series to the point of longest follow-up 4 months later. The continued improvement in outcome measures after completing the class suggests that subjects learned self-management skills that they continued to utilise over time. The magnitude of changes found for the measures of IBS-related symptoms, IBS-Quality of Life, and GI-specific anxiety at 6 months were of small standardised effect sizes and the magnitude of changes for measures of general mindfulness skills and mental health-related functional status were of medium to large effect size. Overall, our results suggest that patients with heightened GI-specific anxiety may benefit from participation in a mindfulness programme as an adjunct to their usual clinical care.

There have been two prior studies that assess the impact of mindfulness education on IBS symptoms or qualify of life.47,48 The two prior published studies each used an intervention specifically designed for IBS, whereas our study assessed the effect of a general mindfulness training programme intended for patients with a variety of problems. We studied this general mindfulness programme (mindfulness-based stress reduction – MBSR) in the format in which it is typically taught because MBSR is already widely available throughout the United States and in many other parts of the world. MBSR could thus represent a widely available additional resource for persons with IBS, if it is shown to be beneficial. In an open pilot study of a mindfulness intervention for IBS in 34 subjects, Ljotsson et al. studied a 10-week intervention specific to IBS that combined elements of mindfulness and exposure therapy.48 Improvements in IBS-QOL and GI-specific anxiety measures were found 6 months post-treatment, which are in agreement with our findings. In a follow-up study, Ljotsson et al. randomised 86 patients with IBS to an internet-delivered version of the above intervention, or a control condition (a waiting list, with access to an online discussion forum). They showed significant improvement in IBS-Quality of Life, but not symptoms of GI-specific anxiety, in the mindfulness and exposure arm as compared with the control condition at 3-month follow-up.47 In a report of longer term follow-up (mean 16.4 months) of 75 members of the same cohort, they reported that improvements in IBS-related outcome measures were maintained, with large effect sizes.63

The prior work by Ljotsson et al. found larger effect sizes from pre- to post-treatment (IBS-QOL, d = 1.30; GI-specific anxiety, d = 1.40) than were found in our study. The larger effect sizes found by these investigators in their open trial could be accounted for by differences in the intervention as well as the patient population studied. Ljotsson et al. included only female participants, had smaller group sizes (4–6 per group when compared with 20–30 participants per group in MBSR), provided specific teaching regarding psychological aspects of IBS and incorporated elements of exposure therapy, in which participants were intentionally exposed to foods or situations known to provoke symptoms of IBS. By contrast, all participants in our study were military Veterans, most were male, and the majority (72%) screened positive for probable post-traumatic stress disorder (PTSD). These differences in study populations are likely to have contributed to differences in the effect size of the mindfulness intervention. Indeed, when we examined IBS patients in our study according to whether they met criteria for probable PTSD, we found larger effect sizes for improvements in IBS-QOL and VSI (Table 5) for patients without PTSD, suggesting that improvements in these measures were attenuated by PTSD. Of note, a limitation of the studies by Ljottsson et al. is that no measure of mindfulness skills was applied to ascertain whether the intervention successfully taught mindfulness. Additional randomised controlled trials are needed and at least one is underway.64

Although our study subjects experienced improvement in clinical outcome measures after the MBSR intervention and these measures appeared to be sustained over the period of follow-up, it is possible that the findings are attributed to factors other than the mindfulness intervention. Given that our study lacked a control arm, one such explanation is regression to the mean. To address this possibility, we assessed the correlation between change in mindfulness and change in outcome variables among those meeting Rome III IBS criteria and found a significant correlation between a change in mindfulness and GI-specific anxiety, suggesting the possibility of a mediating influence of reduced GI-specific anxiety. Several factors may limit the ability to generalise these findings.

Our study was performed in a Veteran population that was predominantly Caucasian and male. The majority of our participants had sustained significant physical or psychological trauma, and met symptom criteria for PTSD, which could lead to a difference in the efficacy of a mindfulness intervention, likely due to avoidance behaviours, which are a hallmark of PTSD. Our study was also unique in that it also applied measures of IBS to persons who did and did not meet strict Rome III criteria for IBS. Our intention was to describe the impact of a stress-reduction intervention for patients across a wide clinical spectrum, given that many patients who report bowel complaints do not meet strict Rome III IBS criteria. These findings emphasise the need for randomised controlled trials of mindfulness interventions for IBS, with both moderation and mediation analyses as well as longer periods of follow-up, to further assess the relationship between mindfulness education and GI symptoms, GI-specific anxiety and quality of life.


Declaration of personal interests: This work was supported by resources from VA Puget Sound Health Care System. We wish to thank Kurt Hoelting, M.Div., who taught some of the MBSR courses for this study. We also wish to thank the Veterans who participated in this research. Declaration of funding interests: None.