A controlled evaluation of group cognitive therapy in the treatment of irritable bowel syndrome

https://doi.org/10.1016/j.brat.2006.07.003Get rights and content

Abstract

We randomized, at two sites, 210 patients with Rome II diagnosed irritable bowel syndrome (IBS), of at least moderate severity, to one of three conditions: group-based cognitive therapy (CT; n=120), psychoeducational support groups (n=46) as an active control, or intensive symptom and daily stress monitoring (n=44). One hundred eighty-eight participants completed the initial treatment. Those in symptom monitoring were then crossed over to CT. For an intent to treat analysis on a composite GI symptom measure derived from daily symptom diaries, both CT and the psychoeducational support groups were significantly more improved than those in the intensive symptom monitoring condition, but the CT and psychoeducational support group did not differ. Among treatment completers on the same composite measure of GI symptoms, again, both CT and psychoeducational support groups were statistically superior to symptom monitoring but did not differ on the symptom composite, or on any other measure. On individual IBS symptoms, both CT and psychoeducational support were statistically superior to symptom monitoring on reductions in abdominal pain and tenderness and for flatulence. Patient global ratings at the end of treatment showed the two active conditions statistically superior to symptom monitoring on change in Bowel Regularity, with CT superior to symptom monitoring on reduction in overall pain and in improvement in sense of well-being. Three-month follow-up data on 175 patients revealed maintenance of significant improvement or continued significant improvement on all IBS symptoms, including the McGill Pain Questionnaire. Group CT and psychoeducational support groups continued not to differ on any measure. We thus conclude that group CT is not superior to an attention placebo control condition.

Introduction

Irritable bowel syndrome (IBS) is a common, often disabling, functional disorder of the lower gastrointestinal (GI) tract characterized by abdominal pain or discomfort (e.g., tenderness, bloating) associated with abnormal bowel habit (diarrhea, constipation or both). With a world-wide prevalence of 10–20%, IBS ranks high among reasons for physician visits (Sandler, 1990) and is costly in terms of medical care (Levy et al., 2001) and time lost from work (Hahn, Yan, & Strassels, 1999) and non-monetary costs such as diminished quality of life and activity limitations (Whitehead, Burnett, Cook, & Taub, 1996).

At present there is no universally accepted medical treatment for the full range of IBS symptoms. For this reason there is a growing body of research concerning the efficacy of psychological treatments (for a recent comprehensive review see Blanchard, 2005; Blanchard & Scharff, 2002; or Lackner, Mesmer, Morley, Dowzer, & Hamilton, 2004). There are three broad categories of empirically validated psychological treatments: hypnotherapy (initially described by Whorwell, Prior, & Faragher, 1984); brief psychodynamic psychotherapy (initially described by Svedlund, Sjodin, Ottosson, & Dotevall, 1983) and various combinations of cognitive and/or behavioral therapies (collectively referred to as CBT).

Because the interpretability of the “first generation” of psychological trials was based on relatively small, randomized controlled trials (RCTs) with methodological shortcomings (Lackner et al., 2004), several larger-scale RCTs have been conducted with the goal of establishing more definitively the efficacy of these approaches. These trials have yielded less than overwhelming results. Creed et al. (2003) compared an interpersonally oriented version of brief psychodynamic psychotherapy, administered individually, to the selective serotonin reuptake inhibitor (SSRI), paroxetine and medical treatment as usual (TAU). Results at end of the acute treatment phase showed no differences among the three conditions on the primary outcome variable of pain, measured by visual analog scale, or the physical component score of the SF-36 (Ware, 1993). Paroxetine was superior to TAU on secondary measures of number of days with pain and overall psychological distress as measured by the global severity index of the SCL-90 (Derogatis, 1994). Both treatments were superior to TAU on the SF-36 mental component score. In no instances were there significant differences between psychotherapy and active drug. At a 12-month follow-up, the results showed no differences among the three conditions on any variable except the SF-36 physical component. On this quality of life variable, both active treatments showed greater improvement than TAU. The health care costs for IBS in the year follow-up were significantly less for the psychotherapy condition than for TAU with costs for paroxetine in between, and not different from either of the other two conditions.

Drossman et al. (2003), using only female patients, compared an individually administered version of CBT, adapted from a group CBT, initially described by Toner et al. (1998), to an individually administered attention-educational condition. Simultaneously, they compared the tricyclic antidepressant, desipramine, to a medication placebo. Using an idiosyncratic composite score of pain ratings, patient satisfaction with treatment, a quality of life measure and a rating of global well being, a significant advantage for CBT over attention-education condition was found at end of treatment. Further analyses revealed that CBT was superior to the attention-educational condition on patient satisfaction ratings and ratings of global well being, but not the other variables including daily pain ratings. There was no significant difference on the composite score between CBT and desipramine nor were there differences between active drug and placebo. No longer-term follow-up results were provided, nor were data on other individual GI symptoms reported.

Boyce, Talley, Balaam, Koloski, and Gruman (2003) compared their version of CBT (previously evaluated only in an uncontrolled trial [Boyce, Gilchrist, Talley, & Rose, 2000]) to relaxation training and to somewhat enhanced routine medical care. Sixty-three percent of patients were evaluated at end of treatment and 49.5% at a 10-month follow-up. Although all 3 groups had improved significantly at the follow-up point, there were no significant differences among the cases assessed in the three conditions in terms of clinical improvement of bowel symptoms at end of treatment or at follow-up. Loss of half of the sample casts doubts on conclusions.

Thus, these three recent large-scale RCTs have failed to find differential effects of psychological treatments on GI symptoms.

Our own recent efforts on this problem have focused exclusively on a cognitive therapy (CT) approach to the treatment of IBS, rather than to multicomponent CBT “packages”. A series of small-scale RCTs of CT have shown good results: Greene and Blanchard (1994) found CT superior to GI symptom monitoring on a composite measure of GI symptoms derived from a daily symptom diary. Payne and Blanchard (1995) found the same form of CT (following the earlier treatment manual) significantly superior in GI symptom reduction to psychoeducational support groups that were equally credible to the CT and to symptom monitoring. Finally, Vollmer and Blanchard (1998) found the same CT, administered individually or in small groups, was significantly superior to GI symptom monitoring.

These results led to the present RCT in which we sought to address three issues: (1) Given the potential cost savings of group treatment over more labor-intensive individual treatment, we wanted to compare the group version of CT to (a) psychoeducational support groups as a commonly available treatment that controls for attention and group meetings, and (b) to an intensive symptom monitoring condition (participants simultaneously monitored GI symptoms and daily stressors for at least 10 weeks). (2) Our primary focus was on assessing the efficacy of the two psychological treatments for the broad spectrum of GI symptoms of IBS, measured primarily by a daily symptom diary and secondarily by global ratings of improvement from both patients and blinded physician to give values comparable to the other reports described earlier. Because of space limitations, secondary focus on other variables such as level of psychological distress, quality of life and life interference, are outside the scope of this report. (3) To evaluate the generality of the treatment, we added a second site with a strong background in CT approaches to pain but without either research experience with IBS or apparent allegiance to the treatment interventions developed at the University at Albany. In other words, we exported our current group CT and psychoeducational support manuals to determine whether their therapeutic value could be achieved by other trained health professionals at another site without allegiance to the two therapies.

Section snippets

Participants

Participants were recruited at each of the two academic tertiary care clinic sites, Albany, NY, and Buffalo, NY, primarily through referral of local physicians (e.g., gastroenterologists), local media coverage and community advertising. A total of 1298 consecutive individuals were screened by telephone; 520 reported a pattern (moderate to severe pain with abnormal bowel habits of at least twice weekly) of IBS-like symptoms that warranted assessment to confirm the IBS diagnosis and were given an

Results

As is obvious from the above description, this was a somewhat large study made complex by the variety of dependent variables. Change in the wide array of GI symptoms associated with IBS was our primary focus since the other recent RCTs found little in the way of differential somatic symptom improvement.

Discussion

This is a large and complex set of results, in large part because different methods of assessing outcome yield somewhat different patterns of results, and in part because of the multisymptomatic nature of IBS. Turning first to somatic symptoms, the reason patients seek treatment, on our most conservative and global measure, the CPSR from patient daily GI symptom diaries, the intent to treat analysis showed the group CT was superior to the intensive symptom monitoring but was not different from

Acknowledgements

This research was supported by a Grant from NIDDK, DK-54211. Requests for further information may be sent to Dr. Blanchard at 72 Old Route 66, Averill Park, NY 12018. The authors wish to acknowledge the assistance in Albany, of Drs. Janine Walsh, Elizabeth Mundy and Tara Galovski who served as post-doctoral fellows and treated patients in Albany, and of Dr. Howard Malamood and of Albany Gastroenterology Associates who provided pre-treatment examinations of patients. In Buffalo, we acknowledge

References (35)

  • A.T. Beck

    Cognitive therapy and the emotional disorders

    (1976)
  • E.B. Blanchard

    Irritable bowel syndrome: Psychosocial assessment and treatment

    (2001)
  • E.B. Blanchard

    A critical review of cognitive, behavioral, and cognitive–behavioral therapies for irritable bowel syndrome

    Journal of Cognitive Psychotherapy

    (2005)
  • E.B. Blanchard et al.

    Psychosocial aspects of assessment and treatment of irritable bowel syndrome in adults and recurrent abdominal pain in children

    Journal of Consulting and Clinical Psychology

    (2002)
  • E.B. Blanchard et al.

    Clinically significant changes in behavioral medicine

    Behavioral Assessment

    (1988)
  • T.D. Borkovec et al.

    Credibility of analogue therapy rationales

    Journal of Behavior Therapy and Experimental Psychiatry

    (1972)
  • P. Boyce et al.

    Cognitive–behaviour therapy as a treatment of irritable bowel syndrome: A pilot study

    Australian and New Zealand Journal of Psychiatry

    (2000)
  • Cited by (104)

    • Irritable Bowel Syndrome

      2020, Interventional Management of Chronic Visceral Pain Syndromes
    View all citing articles on Scopus
    View full text