Original Articles
The relationship between daily stress and symptoms of irritable bowel: a time-series approach

https://doi.org/10.1016/S0022-3999(97)00255-9Get rights and content

Abstract

Irritable bowel syndrome (IBS), a chronic disorder that includes symptoms such as abdominal pain and altered bowel habits, affects up to 22% of people in Western populations. The causes of IBS are not well understood, but are believed to be multifactorial. Although stress is widely believed to be implicated, empirical evidence in support of this is lacking, perhaps because a typical between-participants analysis ignores individual differences and therefore may obscure any link. The present study used a within-person, lagged time-series approach to investigate the links between everyday stress and symptomatology in 31 IBS sufferers. Both everyday stress and symptomatology exhibited serial dependence for a statistically significant proportion of sufferers. Multiple regression analysis carried out on same-day and lagged relationships up to and including 4 days found that, for over half the participants, everyday stress and symptoms were related. The best regression model was one in which symptoms were a function of hassles and symptoms on the previous 2 days, and hassles on the same day, fitting the data for 67% of participants. This prospective study confirms other studies that have suggested stress is a significant factor in IBS, and concludes that stress management programs may be both useful and cost-effective in the treatment of IBS.

Introduction

Irritable bowel syndrome is a chronic disorder affecting approximately up to 22% of people in Western populations [1]. Core symptoms include abdominal pain and altered bowel patterns (diarrhea, constipation, or an alternation of both, a feeling of urgency when needing a bowel movement; and a feeling of incomplete evacuation after a bowel movement). Other symptoms include bloating and flatulence. There are large individual differences in the severity and frequency of symptoms. Whereas symptoms may be hardly noticed by some sufferers, others suffer so severely they are unable to work: between 5% and 8% are nonemployed due to their IBS 2, 3, 4.

There is no consensus of opinion as to the cause or causes of IBS, although stress (everyday or sudden trauma) is widely believed to be implicated. Stress could be a trigger factor, or may contribute to the maintenance of symptoms, regardless of the initial cause or causes. The finding that hypnotherapy can help a number of people who continued to suffer while undergoing more traditional methods of treatment suggests that stress may contribute to IBS, at least for some people [5]. It is rare, however, for IBS sufferers to be offered psychological treatment under the NHS in the UK; most are treated with traditional medications and advice regarding diet [6]. Such treatments are, however, less than effective for a large number of IBS sufferers, who cost employers and the NHS vast sums of money in terms of loss of economic productivity, expensive hospital investigations, and repeated physician visits 7, 8.

Until recently, studies investigating the link between stress and symptoms of IBS used retrospective accounts given by sufferers. A large proportion of IBS sufferers believe stress plays a part in their condition [2]. Such studies are fraught with problems, however—people have a tendency to look back to find a “cause” that can explain their illness [9], and most people can recall stressful episodes, in retrospect. Whitehead [10]stated that stress explains only 10% of the variance in bowel symptoms and that, although this is statistically and clinically significant, it is less than expected, given the frequency with which patients attribute their symptoms to stress. Whitehead states “this finding suggests that people exaggerate the importance of stress in an effort after meaning (i.e., a tendency to dismiss inexplicable events and to focus on those that seem to have a recognizable cause)” (p. 28). Even if stress does lead to increased severity of symptoms, we do not know how long we should expect the timelag to be, although Dancey et al. [11]found that the relationship between stress and symptomatology was strongest on the same day.

Recent studies investigating the stress–symptom link have used a prospective format whereby participants fill out questionnaires measuring symptoms and stress, on a number of occasions. Even using this format, however, has problems if participants are considered as a homogeneous group. For instance, aggregating data and performing a typical between-participants analysis may obscure any link—it may be that stress is an important factor for some sufferers, but not others: For instance, whereas Dancey et al. [11]found that the relationship between symptoms and the next week's stress was stronger than the relationship between stress and the next week's symptoms (both findings were nonsignificant), they acknowledged that the between-participants design would have obscured any significant associations for individual participants. To overcome such problems, recent studies investigating the link between everyday stress and bowel symptoms have used time-series strategies, using within-person correlations as well as assessments between participants. In this study, correlations were computed for each individual separately, not only for stress and symptoms on the same day (which cannot suggest any causality), but for stress and symptoms on different days. Few empirical studies have been carried out using such methods with IBS sufferers. However, Suls et al. [12], studied 44 IBS patients over 21 days. They used several different time-series strategies, and concluded that prior and concurrent daily stress had no consistent effects.

The link between everyday stress and severity of symptoms in IBS sufferers has yet to be clarified. It is particularly important to study nonclinical samples of IBS sufferers, because these may differ from clinical samples in several respects [7]. The present study used a nonclinical sample (i.e., not drawn from doctors lists or out-patient clinics but with a previous diagnosis of IBS) to investigate the link between everyday stress and severity of irritable bowel symptoms. Based on the work of previous studies 11, 12, we expected that, if effects were found, then the relationships between symptoms and subsequent stress would be stronger than vice versa. Some participants were expected to show positively lagged relationships (stress precedes symptoms) and some negatively lagged (symptoms precede stress). It was not possible to hypothesize how many sufferers would be affected in this way.

Section snippets

Method

A within-person time-series analysis was conducted on questionnaire data provided by a nonclinical sample of male and female IBS sufferers. Furthermore, attempts were made at modeling the IBS symptoms by its lagged values and contemporaneous and lagged hassle measurements.

Results

Prior to performing a time-series analysis on the symptom and hassles variables we carried out a listwise deletion of one sufferer, due to the relatively large number of missing values. Also, a missing value for the hassles measurement of one participant (no. 10) was replaced by the mean value of the measurement for this sufferer. To examine daily variation of symptom and hassles measurements, the sample autocorrelation and partial autocorrelation functions of these measurements were obtained

Discussion

It is of interest that, for 17 participants (53%), hassles on one day is a function of hassles on the previous day, whereas for symptomatology (symptoms as a function of symptoms on the previous day) this is the case for only 7 participants (23%). Both of these proportions, however, are significantly different from zero.

Results show that for over 43% of the participants symptoms could be significantly predicted by hassles in the previous 4 days, and that for 37% of them hassles could be

References (14)

There are more references available in the full text version of this article.

Cited by (95)

  • Gastrointestinal symptoms in office workers are predicted by psychological distress and short sleep duration

    2020, Journal of Psychosomatic Research
    Citation Excerpt :

    First, we confirm the existence of a modest association between daily psychological distress and GI symptoms. The estimate (ρ = 0.19) was comparable to that reported by another study [5], despite the different study populations (patient sample vs office workers). It is likely that a large share of this association is causal, and stands under the influence of tertiary factors such as sleep duration, as suggested by the mediation analysis.

  • Individual variation in temporal relationships between stress and functional somatic symptoms

    2014, Journal of Psychosomatic Research
    Citation Excerpt :

    Diary studies using repeated measurements to evaluate associations between stress and FSSs within individual subjects over time can address these problems. A few of these studies have been performed, indicating that an increase in stress predicts an increase in FSSs within subjects [17,18]. These studies however are limited by their short duration and specific study population.

  • How IBD patients cope with IBD: A systematic review

    2013, Journal of Crohn's and Colitis
View all citing articles on Scopus
View full text