Article Text

Download PDFPDF

Cognitive behavioural model of irritable bowel syndrome
  1. Francis Creed
  1. Correspondence to:
    Professor Francis Creed
    Department of Psychological Medicine, University of Manchester, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK; francis.creed{at}

Statistics from

“All or nothing” behaviour in irritable bowel syndrome

Recent epidemiological research into the syndromes of unknown aetiology such as irritable bowel syndrome (IBS), dyspepsia, chronic fatigue syndrome and fibromyalgia has moved away from cross-sectional studies since these can teach us little about causality. Prospective studies need to be focused on high-risk populations, and prospective cohort studies of people suffering infection have been performed to assess the biological and psychological risk factors for the later development of IBS and related conditions. In a study by Spence and Moss-Morris published in this issue of Gut (see page 1066), 447 patients with Campylobacter gastroenteritis completed a series of psychosocial questionnaires soon after the laboratory had processed their stool culture.1 They were followed up at 3 and 6 months after the initial infection and 49 (10%) met the criteria for IBS at both follow-up points. Compared with the remainder, those whose symptoms fulfilled the criteria for modified Rome I (requiring both pain-related symptoms and disturbed defaecation criteria) or Rome II IBS had, at the time of their infection, higher scores on a number of questionnaires. These were: perceived stress, anxiety, somatisation and more negative beliefs about the illness (unlikely to completely resolve, expectation of more persistent complaints). In addition, they had responded during their acute infection by failure to rest and remaining active in spite of their symptoms.


The authors claim that these findings confirm a theoretical model which suggests that people at risk of IBS have anxious or depressive tendencies together with high and often unrealistic personal expectations (high levels of perfectionism). When faced with an infection such as gastroenteritis, such individuals press on and remain active but this leads to worsened symptoms and they are forced to rest. Resting heightens feelings of anxiety and stress, so they try quickly to return to their former level of activity. This pattern of pushing too hard followed by enforced rest is known as “all or nothing behaviour”. If this behaviour is repetitive, the individual may start to believe that he/she has a chronic incurable condition and become increasingly distressed by the ongoing symptoms and illness. This model can explain persistent symptoms associated with distress and negative beliefs about their illness.

One of the most striking aspects of this study is the almost identical replication in a population-based sample of the results of Gwee et al in a hospitalised sample of patients with gastroenteritis.2 In that study the authors found that high scores of stressful life events, hypochondriasis, anxiety, neuroticism and somatisation predicted the development of post-infectious IBS. Spence and Moss-Morris1 have introduced for the first time the concept of “all or nothing behaviour”, a concept which will fit in with many clinicians’ observations of at least some of their patients with IBS.


Has the study by Spence and Moss-Morris truly confirmed the theoretical model? The main strengths of the study lie in its population-based sample derived from laboratories serving a community clinical diagnostic service in Auckland, New Zealand and the fact that standardised questionnaires were completed soon after the infection. The proportion of participants who developed IBS was within the range of previous studies.3–6 The authors used a high threshold for diagnosing IBS—meeting specified criteria at two points in time (3 and 6 months)—which is more stringent than most other studies. The main weakness of the study was the response rate to the first phase (approximately 50%), reflecting difficulties in distributing questionnaires through a third party. The follow-up rate (94% at 3 months and 88% at 6 months) was, however, very high. The other weakness was the inability of the researchers to measure the severity or duration of the Campylobacter infection, although the authors noted that the total symptom count during the initial illness and prescription of an antibiotic (proxy measures of severity) were not significantly associated with IBS. This is important as duration and/or severity of gastroenteritis has been found to be an important predictor in previous studies. It would have been preferable to examine whether the psychological predictors held, even when the severity of the Campylobacter infection was controlled.


This is now one of a series of studies which have shown that psychological characteristics measured at the time of infection is predictive of later IBS.2,6,7 The study by Spence and Moss-Morris is more detailed than any previous one in terms of the psychological assessment, and the authors delineated in multivariate analysis two aspects which were most closely associated with later IBS: a perfectionism, perceived stress and anxiety strand and a tendency to continue to be active in the face of acute infection. The study also found, along with all previous studies, that female sex was an important predictor; women were 2.6 times more likely to develop IBS than men.

One of the most important aspects of this type of study is the attempt to integrate—rather than separate—the psychological and “biological” aspects of the disorder. This study has emphasised the importance of psychological risk factors, but severity or duration of gastroenteritis suggests a strong biological component.4,8 Interestingly, one study showed that both increased enterochromaffin cell counts and depression scores were independent predictors of developing post-infectious IBS.6 This indicates that there may be different routes into post-infectious IBS, with psychological factors exerting their influence independent of the severity of the infection. Spence and Moss-Morris1 found that the female predominance was clear even after controlling for all psychological factors, which suggests that this is likely to be a constitutional, possibly genetic, component.


These studies raise important issues of specificity. Post-infectious IBS has been found following Shigella, Salmonella and Campylobacter infections, and one study found no difference according to type of infection.8 It therefore appears that post-infectious IBS is not a reaction to a particular type of infection. There is even one recent report that post-infectious IBS may occur after non-gastrointestinal infections, although the numbers in that study were too small for reliable results.9 Furthermore, Salmonella infection may be followed by functional dyspepsia as well as IBS, and only dyspepsia was associated with severity of infection indicating a possible causal relationship.10 This raises the question whether viral infections act as a “trigger” for a biological or psychological/behavioural response rather than as a specific effect on an end organ.

The counter argument is made in a companion paper by Spence and Moss-Morris in which they have shown that Campylobacter infection was followed by IBS whereas infectious mononucleosis was more likely to be followed by chronic fatigue syndrome.11 There was some overlap between IBS and chronic fatigue syndrome, but the model did seem to be more specific for IBS than for chronic fatigue. Further work is needed to test this argument more fully.


It is not clear also whether there is a specific psychological “profile” representing vulnerability to developing IBS after gastroenteritis. In spite of the similarity of the psychological profile found by Gwee and Spence and Moss-Morris, others have found slightly different patterns which suggests, for example, that depression may play a role as well as anxiety.6,12 Rather similar psychological attributes have been found to predict chronic fatigue syndrome and fibromyalgia.11,13–15 For example, two follow-up studies of infectious mononucleosis found that female sex, life events prior to the infection, illness perceptions and greater family support predicted failure to recover (ie, chronic fatigue) at 6 months.14,16 Fatigue is a common symptom of many viral infections and a well-recognised sequel and may, of course, accompany bowel dysfunction in IBS.


Studies of chronic fatigue syndrome mirror those in IBS. In the post-viral fatigue studies, severity of infectious mononucleosis—a proxy measure of the “biological” component of the subsequent disorder—is a predictor of failure to recover at 2–3 months following the infection, but its role is less clear subsequently.14,17 It is now established that the post-infective fatigue state is one of the pathophysiological pathways into chronic fatigue syndrome.18 The psychological pathway appears to be supported in some studies but not all.11,18 The difference between studies may relate to different symptom complexes.

A detailed study of symptoms has shown that two syndromes follow infectious mononucleosis: one (5% of the sample) represented a “pure” fatigue syndrome thought to be primarily “biological” in origin but, in the other (6% of the sample), fatigue symptoms were accompanied by numerous other bodily symptoms, anxiety and depressive symptoms.17 Thus, there seems to be evidence for both a specific fatigue syndrome following infectious mononucleosis that is independent of psychological factors19 and a chronic fatigue syndrome that has associated psychological features.17 Could the same be true in IBS?

In one study of bowel symptoms 6 months after gastroenteritis, 25% of participants had persistently altered bowel habits but only one-fifth of these had symptoms classified as Rome I IBS.8 The former may be independent of psychological factors and they may be accompanied by changes in the immune response and changes in central sensitivity and visceral hypersensitivity.20 It appears from the study by Spence and Moss-Morris that the latter group, whose symptoms fulfil Rome criteria for IBS, are the individuals who have the personality attributes of perfectionism, perceived stress and anxiety and a tendency to continue to be active in the face of acute infection, which means that their symptoms are perpetuated. Whether their symptoms are also accompanied by immune changes is a topic to be addressed in future studies of post-infectious IBS.


An “all or nothing” response has also been observed in chronic fatigue syndrome and fibromyalgia21 where it has been associated with both onset and persistence of the disorder. Importantly, it has been shown that this cognitive and behavioural “action-proneness” can be modified slightly by cognitive behaviour treatment.22 In addition, a brief rehabilitation intervention administered to people who have had infectious mononucleosis was helpful in the long term (6–12 months) but not in the short term (3 months), giving further evidence that it is the long-term sequelae of infections that have an important psychological component.23 This study did suggest, however, that early intervention for those who are at risk of developing subsequent IBS or a related syndrome may be possible when the risk factors have been firmly established.

Finally, the relationship between post-infectious IBS and IBS as a whole needs to be considered as post-infectious IBS only represents 6–17% of cases of IBS in the clinic.24 It will be helpful to study this pathophysiological pathway into IBS, but the results may not be generalisable to IBS as a whole. Strenuous efforts to study concurrently psychological and infective factors at the time of onset of IBS are likely to increase our knowledge significantly and to improve the treatments available to people with established IBS.

“All or nothing” behaviour in irritable bowel syndrome


View Abstract


  • Competing interests: None declared.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles