Original articles
The role of stress in symptom exacerbation among IBS patients

https://doi.org/10.1016/j.jpsychores.2007.10.010Get rights and content

Abstract

Over 200 treatment-seeking irritable bowel syndrome (IBS) patients completed 4 weeks of daily prospective measures of stress and gastrointestinal symptoms as well as retrospective measures of stress (life events over 12 months, hassles over 1 month). We also obtained the stress measures on 66 nonill controls. Irritable bowel syndrome patients report more frequent hassles than controls and a greater stress impact than controls. Using structural equation modeling, we found that the data were consistent with a model of robust autocorrelation effects of both week-to-week gastrointestinal (GI) symptom indices (r=.84) and stress indices (r=.73), as well as strong concurrent effects of stress on IBS symptoms (r=.90) and vice versa (r=.41). The data also were consistent with a model where there were effects of stress in Week t upon GI symptoms in Week t+1 and t+2, but they were mediated through the concurrent week effects and/or autocorrelation effects. There were no statistically significant independent pathways from stress in Week t to GI symptoms in Week t+1 or t+2. Thus, there is more support for a reciprocal relation between stress and symptoms than there is for a causal relation.

Introduction

Irritable bowel syndrome (IBS) is a widespread functional disorder of the lower gastrointestinal (GI) tract that is characterized primarily by abdominal pain and altered bowel habits (diarrhea and/or constipation). It has long been thought of as a stress-related disorder [1]. That is, psychological stress is posited to play either a causal role or exacerbating role in IBS and its symptoms. Certainly, if one asks treatment-seeking IBS patients about the role of stress in their IBS, a sizeable majority will answer affirmatively.

The role of stress, or stressful events, and IBS has been examined with three different methodologies for measuring stress: (a) the frequency and intensity of major life events or life changes over the past 6 or 12 months; (b) the frequency and intensity of minor annoyances or hassles [2] over the past month; (c) prospective monitoring studies measuring IBS symptoms and minor stressful events or self-ratings of stress experienced on a daily or weekly basis for several weeks to learn if stress on Day n predicts IBS symptoms on Day n+1. (For a summary of this literature, see Ref. [3], chapter 8.)

In these studies, IBS patients have been compared to patients with organic GI disease or to nonill controls. For the patient comparisons, two studies [4], [5] found more major stressful events among IBS patients than comparison patients, whereas two others [6], [7], using similar ill controls, did not find a difference. In the comparisons of IBS patients to nonill controls, four studies show greater history of major life events among IBS patients [4], [8], [9], [10], whereas two studies did not [7], [11]. Finally, Bennet et al. [12] found significant correlations between a number of different functional GI symptoms and a number of chronic life stressors over an extended period (M=3.9 years) in a mixed sample of patients with IBS and patients with functional dyspepsia.

For the three studies that report on hassles and IBS, none have compared IBS patients to others. The average weighted hassles score was 33.1 in one study [13] and 35.5 in the other [14].

The five prospective monitoring studies [10], [11], [14], [15], [16] yield very complex results, in part, because of the complex multivariate statistics used. The Whitehead et al. [10] study gathered data every 3 months for a year. The other four studies used daily monitoring of GI symptoms and stress (variously defined) for 3 to 8 weeks. Two consistent conclusions appear in these studies: For all studies, there is a significant concurrent association of GI symptoms and stress, that is, GI symptoms measured on Day n correlate with stress also measured on Day n. Second, there is no significant association across the sample between stress on Day n and GI symptoms on Day n+1. Exceptions to the latter conclusion are the findings by Levy et al. [11] that the combination of stress measured at Day n and n+1 predicts GI symptoms at Day n+1. Using time series analyses, Levy et al. found this relation was significant in 38% of IBS patients.

Dancy et al. [15] found that 43% of self-identified IBS patients showed relations between stress measured on the combination of Days n−4, n−3, n−2, and n−1 and GI symptoms on Day n. However, only 16% showed significant correlations between stress on Day n−1 and GI symptoms on Day n.

Suls et al. [16] criticized the methodology of Dancy for not correcting for the autocorrelation one finds in the two measurement domains. That is, GI symptoms on Day n predict GI symptoms on Day n+1, as do stress measures. When they controlled for these two sources of variance, they found only 1% to 2% of new variance in Day n GI symptoms could be accounted for Day n−1 stress. All of the daily monitoring studies used relatively small samples of IBS patients (from n=26 in Levy et al. to n=44 in Suls et al.), and most used exclusively female populations.

It thus seemed clear to us that a large-scale (n>100) study on both genders using appropriate statistics was warranted to try to make the relation (or lack thereof) clearer. We also gathered data on previous major life events and hassles to see what role they might play.

Section snippets

Irritable bowel syndrome patients

There were 254 IBS patients who had been recruited as possible participants in a two-site psychological treatment trial. All met Rome II [17] criteria for IBS as determined by a structured interview [3] by a psychologist and confirmed by history, physical examination, and appropriate laboratory tests by a board-certified gastroenterologist. Exclusion criteria were organic GI disease, other life-threatening disease (e.g., cancer), or a psychiatric diagnosis (see below) of a psychotic disorder

Clustering of GI symptoms

Inspection of the correlations among the separate GI symptoms revealed five clear clusters of symptoms that tended to be highly correlated within a cluster but only slightly to moderately correlated between clusters. The symptoms were classified into five factors defined as follows: (a) Factor 1, abdominal tenderness, abdominal pain, bloating; (b) Factor 2, number of times had diarrhea and severity of diarrhea; (c) Factor 3, belching and flatulence; (d) Factor 4, nausea; and (e) Factor 5,

Within group analyses

For the primary analysis, the prospective analyses of stress and GI symptoms, our findings mirror those of several previous smaller-scale studies: there are significant concurrent relations between reports of stress and GI symptoms across all GI symptom factors and across all weekly intervals. This finding agrees with the work of Levy et al. [11] and Dancy et al. [15].

Our results also highlight the strong associations between stress in Week t and stress in Week t+1 and between GI symptoms in

References (34)

  • AI Mendeloff et al.

    Illness experience and life stresses in patients with irritable colon and with ulcerative colitis

    N Engl J Med

    (1970)
  • MJ Ford et al.

    Life events, psychiatric illness in the irritable bowel syndrome

    Gut

    (1987)
  • EB Blanchard et al.

    Psychological comparisons of irritable bowel syndrome to chronic tension and migraine headache and non-patient controls

    Biofeedback Self Regul

    (1986)
  • WE Whitehead et al.

    Effects of stress or life events on bowel symptoms: subjects with irritable bowel syndrome compared with subjects without bowel dysfunction

    Gut

    (1992)
  • RL Levy et al.

    The relationship between daily life stress and gastrointestinal symptoms in women with irritable bowel syndrome

    J Behav Med

    (1997)
  • EJ Bennet et al.

    Functional gastrointestinal disorders: psychological, social, and somatic features

    Gut

    (1998)
  • A Payne et al.

    A controlled comparison of cognitive therapy and self-help support groups in the treatment of irritable bowel syndrome

    J Consult Clin Psychol

    (1995)
  • Cited by (125)

    • Norepinephrine induces growth of Desulfovibrio vulgaris in an iron dependent manner

      2022, Anaerobe
      Citation Excerpt :

      Stress has been reported to increase susceptibility to bacterial infections and exacerbate intestinal diseases such as inflammatory bowel disease (IBD), ulcer colitis (UC), irritable bowel syndrome (IBS), gastro-esophageal reflux disease (GERD), and peptic ulcer disease (PUD) [1–5].

    View all citing articles on Scopus
    View full text