Dr G. R. Locke III, Mayo Clinic, 200 First Street, SW, W-19A, Rochester, MN 55905, USA. E-mail: firstname.lastname@example.org
Background : The health-related quality of life is impaired in patients with functional gastrointestinal disorders seen in referral centres.
Aim : To determine whether the health-related quality of life is impaired in subjects with functional disorders in the community and whether any differences can be explained by psychological co-morbidity.
Methods : In a population-based, nested, case–control study, subjects reporting symptoms of either dyspepsia or irritable bowel syndrome and healthy controls were interviewed and completed a battery of psychological measures plus a validated, generic, health-related quality of life measure (Medical Outcomes Study 36-item short form health survey, SF-36). The association between irritable bowel syndrome and dyspepsia and the physical and mental composite scores of SF-36 were assessed with and without adjustment for psychological state.
Results : One hundred and twelve cases (30 dyspepsia, 39 irritable bowel syndrome, 32 dyspepsia and irritable bowel syndrome and 11 gastrointestinal symptoms but not dyspepsia or irritable bowel syndrome) and 110 controls were enrolled. In the unadjusted linear regression models, irritable bowel syndrome (but not dyspepsia) was negatively associated with the physical composite score (P < 0.05); in an adjusted model, the association between the physical health-related quality of life and irritable bowel syndrome was explained by the Symptom Checklist-90 somatization score alone. In unadjusted models, irritable bowel syndrome and dyspepsia were each negatively associated with the mental composite score (P < 0.05). The association between the mental health-related quality of life and dyspepsia remained after adjusting for psychological covariates, but the association between this and irritable bowel syndrome was not significant after adjustment.
Conclusions : In the community, health-related quality of life is impaired in subjects with irritable bowel syndrome and dyspepsia; however, much of this association can be explained by psychological factors.
Functional gastrointestinal disorders, including the irritable bowel syndrome and functional dyspepsia, are characterized by chronic symptoms referable to the gastrointestinal tract in the absence of a currently recognized biochemical or structural explanation.1,2 The cause of these disorders remains essentially unknown, but psychological factors have been implicated. For example, patients with dyspepsia report more life stress and psychological distress than healthy controls in clinic-based studies.3–8 Likewise, patients with irritable bowel syndrome who present to tertiary referral centres have a higher prevalence of psychological distress, major depression, somatization, hypochondriasis and anxiety than healthy controls.9–14 In addition to possibly being a risk factor for the development of functional gastrointestinal disorders, psychological well-being may be markedly affected as a consequence of severe gastrointestinal symptoms, although this assumption is based on limited data and the exact nature of the association remains to be ascertained.15,16
The health-related quality of life is a patient-focused concept; it usually refers to an impairment of functional status (physical or mental) and the sense of well-being, and is an important measure of the impact of chronic disease.17–21 Several studies have shown that the health-related quality of life of patients with irritable bowel syndrome seen in referral practices is decreased,22,23 but, in a systematic review of 13 trials, it was concluded that only one of 11 positive trials was of high quality, and there were two negative studies, one of which was of high quality.22 Data on patients with dyspepsia are particularly limited, but also report impairment of the quality of life in this disorder.24 Patients with irritable bowel syndrome and dyspepsia seen in referral practices, however, are known to have significant psychological co-morbidity.3–14 The role of psychological distress in causing impairment in the health-related quality of life is not known.
There are even fewer studies that have evaluated the impact of irritable bowel syndrome on the health-related quality of life in the general population, and the data have been conflicting.22 In a population-based study, Koloski et al. found that those with a functional gastrointestinal disorder according to the Rome I criteria had significantly more impaired mental and physical functioning than those not fulfilling the standard diagnostic criteria.25 Similarly, Badia et al. found that, compared with the general population, subjects from the community with irritable bowel syndrome reported significantly worse health-related quality of life scores on four dimensions of the Medical Outcomes Study 36-item short form health survey (SF-36) (‘bodily pain’, ‘vitality’, ‘social functioning’ and ‘role-emotional’).26 Furthermore, Koloski et al. observed that the quality of life was more significantly limited in consulters compared with non-consulters with a functional gastrointestinal diagnosis,25 but other studies focusing on irritable bowel syndrome have disagreed.22 Westbrook et al. evaluated uninvestigated subjects with dyspepsia and observed that dyspeptics reported poorer physical and mental well-being than did non-dyspeptics, with greater differences in mental well-being especially among women,27 but these results remain to be confirmed and psychological co-morbidity was not considered. Each of the previous population-based studies has been cross-sectional. Mailed survey or telephone interviews were used to collect data without a review of the medical record or interaction with a physician.
In summary, the health-related quality of life reported by subjects with irritable bowel syndrome and dyspepsia seen in referral centres is clearly diminished. However, these patients also have significant psychological co-morbidity, which also impacts on their quality of life. Studies have shown that patients with functional gastrointestinal disorders in referral centres are different from those with these disorders who do not seek care, and therefore population-based research is needed to better identify which issues are related to functional gastrointestinal disorders and which are related to health care seeking. In subjects with irritable bowel syndrome and dyspepsia in the community, the levels of impairment of the health-related quality of life and psychological co-morbidity are much less clear. Controversy exists as to whether non-consulters have any impairment in their health-related quality of life. If the health-related quality of life is impaired, it may simply be a manifestation of a subject's psychological state and have nothing whatsoever to do with the gastrointestinal disorder. This becomes important when determining how best to improve the health-related quality of life of these individuals. Should the emphasis be on their gastrointestinal disturbance or on their mental health? Because of the importance of these issues, we aimed to determine whether the health-related quality of life is reduced in unselected community residents with functional gastrointestinal diagnoses compared with controls. In addition, we aimed to evaluate whether any health-related quality of life impairment could be explained by psychological co-morbidity.
Materials and methods
The study was approved by the Institutional Review Board of the Mayo Clinic. The Olmsted County, Minnesota, population comprises approximately 100 000 subjects, of whom, currently, 90% are white; socio-demographically, this community is very similar to the US white population, and the Mayo Clinic is the major provider of medical care.28 Each year, over half of the county population is seen at one of the clinic facilities or by another local care provider. During any given 3-year period, over 95% of local residents will have had at least one local health care contact.28 An important feature of the local environment is that each of these providers employs a dossier or unit record system, whereby all medical information for each individual is accumulated in a single record. The pertinent clinical data are accessible because the Mayo Clinic has maintained, since 1910, extensive indices based on clinical and histological diagnoses and surgical procedures.29 The system was further developed by the Rochester Epidemiology Project, which created similar indices for the records of the other providers of medical care to Olmsted County residents. The Rochester Epidemiology Project records linkage system therefore provides what is essentially an enumeration of the population from which samples can be drawn.28 The data resources have been utilized in a series of investigations into the epidemiology of functional bowel disease.30–35
Using this system, an age- and gender-stratified random sample of 904 Olmsted County residents, aged 20–50 years, was mailed a Bowel Disease Questionnaire. The Bowel Disease Questionnaire has been shown to be a reliable and valid measure of gastrointestinal symptoms, and contains 42 questions that record symptoms of dyspepsia and irritable bowel syndrome.36,37 The instrument also contains items that assess past peptic ulcer disease, physician visits, impact of symptoms on daily activities, overall health status and demographic information. Reminder letters were sent at 2, 4 and 7 weeks unless the subjects indicated that they did not wish to participate, in which case they were not contacted further. A total of 598 (66%) of 904 eligible subjects responded.
Subjects for the present study were selected from participants in this population-based survey. Those who reported either dyspepsia (defined as any pain or ache above the navel) or irritable bowel syndrome (defined as a pain or ache above or below the navel plus at least two of the six standard Manning symptom criteria)37 were considered as possible cases. Those subjects who reported no abdominal pain and less than two other gastrointestinal symptoms were selected as possible controls. Subjects who reported more than two gastrointestinal symptoms but did not meet the definition of dyspepsia or irritable bowel syndrome were excluded (n = 211).
The complete in-patient and out-patient medical record of each potential case and control subject was reviewed. Subjects were excluded if they: (i) reported a history of peptic ulcer disease or gastric surgery on the questionnaire (n = 43); (ii) had a documented organic gastrointestinal or systemic disease (n = 6); (iii) had a previous history of psychosis or mental retardation (n = 4); or (iv) were not residents of Olmsted County or were lost to follow-up (n = 34), pregnant (n = 15) or deceased (n = 1). Because the number of non-Caucasians was small (n = 3), they were not included in the analysis. Potential controls were also excluded if they had a significant gastrointestinal symptom or disease documented in the medical record (n = 21).
Following additional approval by the Institutional Review Board, all 260 eligible subjects were sent a letter of invitation to participate in this second study. Those who did not refuse by mail were contacted by telephone. The subjects were informed that this was an extension of the previous questionnaire study, but were not advised of their status as a case or control. Those subjects who declined to come in for an out-patient visit were asked to participate by mail. All subjects received $25 for their participation in the study. There were 152 subjects who participated in person and 70 who participated by mail; thus, 222 (85%) of eligible subjects participated.
Conduct of the study
The 152 participants who came to the General Clinical Research Center at the Mayo Clinic underwent a structured interview and an abdominal examination by a physician (GRL) who was blind to their status as a case or control. The history included general medical information, cigarette, alcohol and coffee usage, a specific assessment of past and present gastrointestinal symptoms and enquiry as to physician visits. The abdominal examination was to ensure the absence of major pathology.
Following the interview, the subjects were asked to complete the following questionnaires.
(a) The Bowel Disease Questionnaire: similar to that which they had completed previously by mail in the initial survey.36
(b) The Medical Outcomes Study 36-item short form health survey (SF-36): the SF-36 is a widely used general health profile questionnaire with 36 question items comprising eight scales:38,39 physical functioning (e.g. walking, lifting); role functioning-physical (limitations in ability to perform usual activities); role functioning-emotional (impact of emotional problems on work or daily activities); social functioning (impact of health or emotional problems on social activities); bodily pain (level of bodily pain or discomfort); mental health (anxiety, depression, sense of psychological well-being); vitality (energy level or fatigue); and general health perceptions (global evaluations of health). Two summary measures, the physical composite score and the mental composite score, reflect any impact on physical and mental health, respectively, and are constructed from the eight sub-scales. The physical composite score is derived from the physical functioning, role functioning-physical, bodily pain and general health perceptions scales. The mental composite score is derived from the mental health, role functioning-emotional, social functioning and vitality scales.
(c) The Symptom Checklist-90 (SCL-90): a widely used instrument for the self-reporting of psychological distress and psychopathology. This instrument comprises three global indices and nine sub-scales.40
(d) The Life Experiences Survey: this collects information about life events which are scored negatively or positively depending on the nature of the event.7,41
(e) The Social Support Questionnaire: this asks subjects about the support they can rely on and their level of satisfaction with that support.42
(f) The Abuse Questionnaire: this asks subjects about any history of sexual, emotional, physical or verbal abuse either during childhood or adulthood.43
(g) The somatic symptom checklist: this is a measure of somatic complaints.36,44
Subjects who did not wish to come to the General Clinical Research Center were mailed the same questionnaires which they completed and returned by mail.
Immediately following the interview, the physician classified each subject using the following criteria.
(a) Functional gastrointestinal disorders: subjects who met any of the following four criteria were considered to have a functional gastrointestinal disorder: (i) dyspepsia (any persistent or recurrent abdominal pain or discomfort centred in the upper abdomen, including post-prandial fullness, early satiety, nausea or upper abdominal bloating);1 (ii) irritable bowel syndrome [any persistent or recurrent abdominal pain or discomfort with at least two of the following six features (Manning criteria) in the absence of any structural explanation: pain relief with bowel action, more frequent stools with pain, looser stools with pain, abdominal distension, sensation of incomplete evacuation and passage of mucus];30,37 (iii) both dyspepsia and irritable bowel syndrome (any patient who had an overlap of symptoms and thus met the definition of both dyspepsia and irritable bowel syndrome); (iv) other functional bowel disorder (any persistent or recurrent abdominal symptoms not meeting the criteria listed above); these subjects had symptoms consistent with the definitions of irritable bowel syndrome or dyspepsia on the cross-sectional survey, but did not report such symptoms at the time of the case–control study.
(b) Healthy control: no abdominal pain or frequent gastrointestinal symptoms.
All subjects were classified as irritable bowel syndrome, dyspepsia, dyspepsia/irritable bowel syndrome, functional bowel disorder or control in a blind fashion using the criteria outlined above and the symptoms reported on the Bowel Disease Questionnaire completed either at the time of the physician interview or returned by mail with the set of questionnaires. For the 152 subjects who came to the General Clinical Research Center, the case vs. control diagnoses made on the basis of the physician interview were compared with those from the Bowel Disease Questionnaire; they were found to be 94% concordant, confirming the validity of combining interview and mail-in data. When discordant, the physician diagnosis was used in the analysis.
All questionnaires were scored according to the standard instructions. The SF-36 domain and composite scores were age and gender adjusted; this adjustment results in an expected mean score of 50 for each domain and composite score.39
Potential selection bias was examined by assessing the association of factors such as age and gender in responders (vs. non-responders) and in those participating in person (vs. by mail) using logistic regression analyses.
The associations between functional gastrointestinal diagnoses and the health-related quality of life were examined using multiple linear regression models. Separate models were developed with each of the individual SF-36 domain scores and each of the composite scores as dependent variables. Functional gastrointestinal diagnosis was an independent variable in the model and coded as irritable bowel syndrome, dyspepsia, both or neither. Covariates included as predictor variables in the models included age, gender, somatic symptom checklist score, SCL-90 global severity index and somatization score, social support score, life event stress score (positive, negative and total) and abuse history (any adult or childhood abuse).
As mentioned above, the cases and controls were drawn on the basis of responses to a cross-sectional survey. The number of cases and controls available was constrained by the response rate to the survey and the proportions reporting symptoms on the survey. We aimed to recruit everyone who was eligible. If we had performed a standard case–control study in which cases and controls were randomly drawn from respective populations of cases and controls, there would have been approximately 80% power to detect differences in mean values between two samples of 110 controls and 112 cases of size 0.38 s.d. units of measurement (e.g. SF-36 composite physical score). Differences of size 0.43 s.d. units could have been detected with 80% power for n = 110 vs. n = 71 (i.e. irritable bowel syndrome) and of size 0.45 s.d. units for n = 110 vs. n = 62 (i.e. dyspepsia). These detectable size differences are based on the use of a two-sample t-test with a two-sided alpha level of 0.05. The (multiple linear) regression analyses used in the current study would have provided comparable power to detect corresponding associations between sub-group membership (e.g. cases vs. controls) and the SF-36 composite scores in models which did not include any covariates. Some of the variation in the composite scores that could be explained by the sub-groups was accounted for by the covariates, as these were associated with the sub-groups as well.
In comparison with non-responders, participants were more likely to be female (58% vs. 40%) and married (70% vs. 47%) (P < 0.05 for both). No demographic differences were detected between those who came to the General Clinical Research Center for an examination and those who participated by mail, except that college graduates were more likely to come to the General Clinical Research Center. Of the 222 subjects who participated, 112 were classified as having a functional gastrointestinal disorder (30 dyspepsia, 39 irritable bowel syndrome, 32 both dyspepsia and irritable bowel syndrome and 11 other functional bowel disorder) and 110 were classified as controls (Table 1). The mean age of the participants was 36 years and 58% were female. No statistically significant association was identified between case–control status and gender or educational level. However, age and physician visits were associated with case–control status; control subjects were significantly older and were less likely to have visited a physician for any reason. A higher proportion of controls participated by mail, but this was not statistically significant.
Table 1. The distribution of gastrointestinal diagnostic groups according to the method of classification
Neither IBS or dyspepsia
IBS, irritable bowel syndrome.
Age (mean ± s.d.) (years)
36.1 ± 9.1
37.6 ± 8.9
36.7 ± 9.7
33.8 ± 8.4
31.8 ± 7.8
40.6 ± 10.1
Patients given physician diagnosis (%)
Patients only mailed questionnaire (%)
The age- and gender-adjusted scores for the eight domains of the SF-36 are provided in Figure 1. The controls had slightly higher than average normal health-related quality of life scores compared with the population mean of 50. In the univariate linear regression analyses, functional gastrointestinal disorders were significantly associated with several domain scores (all P < 0.01). Specifically, subjects with any functional gastrointestinal disorder had significantly worse scores on six domains: role functioning-emotional, social functioning, bodily pain, mental health, vitality and general health perceptions (all P < 0.01). Subjects with irritable bowel syndrome and subjects with dyspepsia reported significantly lower scores on the same six domains as noted for all functional gastrointestinal disorders combined (each P < 0.01).
The physical composite scores from the SF-36 are summarized in Table 2. The controls had higher than population-mean scores for the physical composite score. The physical composite score of the cases was lower than that of the controls. In the unadjusted linear regression model predicting the physical composite score, irritable bowel syndrome was significantly associated with the physical composite score (P < 0.05), whereas dyspepsia was not. In the adjusted model, neither irritable bowel syndrome nor dyspepsia was significantly associated with the physical composite score. Only the SCL-90 somatization scale was associated with the physical composite score in the adjusted model.
Table 2. Quality of life: age- and gender-adjusted mental and physical composite scores from the Medical Outcomes Study 36-item short form health survey (SF-36)
Mental composite score
Physical composite score
Any functional gastrointestinal disorder
Irritable bowel syndrome
Neither irritable bowel syndrome or dyspepsia
The mental composite scores from the SF-36 are also shown in Table 2. Again, the healthy controls had higher scores than the population mean. The cases had lower mental composite score than controls. In the unadjusted linear regression model predicting the mental composite score, irritable bowel syndrome alone and dyspepsia alone were both significantly associated with the mental composite score (P < 0.05). In the adjusted model, irritable bowel syndrome was not significantly associated with the mental composite score adjusting for the global severity index of the SCL-90, life event stress and social support. However, the association with dyspepsia remained significant (P < 0.05), even after adjusting for the psychological covariates. Life event stress, social support and the global severity score of the SCL-90 were associated with the mental composite score.
Of the subjects with functional gastrointestinal disorders, 47 (46.1%) had reported visiting a physician for pain or discomfort in their stomach or belly. No significant associations of any of the quality of life scores were detected between those subjects with functional gastrointestinal disorders who had seen a physician in the past year and those who had not.
In this population-based, nested, case–control study, we examined the association between upper and lower functional gastrointestinal disorders and health-related quality of life. Subjects with either irritable bowel syndrome or dyspepsia reported poorer health-related quality of life on several domains: physical functioning, social functioning, bodily pain, mental health, vitality and general health perception. The two domains in which subjects with functional gastrointestinal disorders had comparative scores with healthy controls were in the role disability scales, namely role functioning-physical and role functioning-emotional domains. The physical and mental composite scores of the subjects with functional gastrointestinal symptoms were lower than those of healthy controls. There were, however, notable differences between irritable bowel syndrome and dyspepsia. In particular, psychological confounders explained the association of irritable bowel syndrome, but not dyspepsia, with impaired mental health quality of life scores.
Published data indicate that irritable bowel syndrome is associated with an impaired quality of life in those who consult. For example, a postal survey of irritable bowel syndrome patient support groups in the US and UK revealed that poor quality of life across all eight domains was universal amongst responders.45 However, no comparison was made with healthy controls, and so the findings can only be extrapolated to patients. Likewise, Whitehead et al. showed that the generic health-related quality of life scores in patients and non-patients with irritable bowel syndrome were significantly poorer than in healthy controls in the community.46 Gralnek et al. examined the health-related quality of life of irritable bowel syndrome patients and compared it with previously published SF-36 scores for subjects with other diseases, including gastro-oesophageal reflux disease, diabetes, depression and end-stage renal disease;47 the health-related quality of life was significantly impaired in irritable bowel syndrome patients on most domains compared with the other chronic disease groups, and impaired across all eight domains compared with the US healthy population. The patients surveyed were health care seeking by definition as they were either self-referred or referred by primary care providers or community gastroenterologists to an academic health centre, but the data for healthy controls were drawn from the Medical Outcomes Study. These historical data may not be directly comparable with the population from which the cases were drawn. Indeed, a rigorous meta-analysis of the irritable bowel syndrome and quality of life literature identified 13 studies that compared irritable bowel syndrome patients with healthy controls, of which 11 showed a significant reduction in health-related quality of life in patients with irritable bowel syndrome.22 Only one of these 11 was considered to be of high quality, however, demonstrating the need for further work.
It remains controversial whether the quality of life is impaired in non-consulters with irritable bowel syndrome, and data are lacking in dyspepsia. O'Keefe et al. in an elderly population-based cross-sectional study, showed that those with irritable bowel syndrome had impaired quality of life scores compared with asymptomatic subjects.48 In another cross-sectional study, Koloski et al. recently reported that, in those fulfilling the Rome I criteria for a diagnosis of a functional gastrointestinal disorder, compared with population controls, there was a significant association on the SF-36 in terms of impaired mental functioning (43.9 vs. 48.1, respectively) and physical functioning (47.7 vs. 51.6, respectively).25 Our results in irritable bowel syndrome and dyspepsia are consistent with these findings, and, notably, the mean differences were of a similar magnitude. On the other hand, Koloski et al. also reported that mental and physical functioning on the SF-36 were significantly more impaired in consulters vs. non-consulters with a functional gastrointestinal disorder, and, furthermore, non-patients with functional gastrointestinal complaints had more impaired mental and physical functioning than healthy controls.25 In contrast, we observed no differences in quality of life impairment between consulters and non-consulters in our population-based sample. It therefore seems doubtful that the quality of life alone is a key driver of health care seeking in this community, based on the quality of life scores observed.
The present study expands on previous research because of the level of detail of evaluation. Each of the previous population-based studies mentioned above performed a cross-sectional survey or interview, which collected symptom data and either SF-36 or SF-12. Such a survey was performed in the initial phase of this study. However, survey subjects were then asked to present for an interview. Their medical records were reviewed and 68% were interviewed by a physician. Each completed a battery of questionnaires. This is a critical validation step to confirm the survey data. Some investigators have interviewed volunteer samples to collect data on non-consulters. The present study is the only true population-based, nested, case–control study performed so far.
We were especially interested in examining whether psychological factors, and in particular personality, life event stress and abuse history, may confound the association between functional gastrointestinal complaints and health-related quality of life. It is well established that psychological distress is associated with irritable bowel syndrome and dyspepsia,3–14 but it is conceivable that psychological distress could also explain any decreased health-related quality of life. Whitehead et al. reported that neuroticism and psychological distress were correlated with impaired quality of life in irritable bowel syndrome.46 We therefore explored whether the association between functional gastrointestinal symptoms and health-related quality of life could have been confounded by psychological factors in multivariate models; we found that the association between impaired physical and mental health-related quality of life in irritable bowel syndrome was primarily explained by somatization. However, the association between dyspepsia and mental health-related quality of life was independent of the covariates measured, and dyspepsia was not associated with impaired physical health-related quality of life. An explanation for the observed differences by diagnosis is speculative, but the association between psychological distress and dyspepsia has been inconsistent in the literature.3–8 It is also conceivable that dyspepsia represents a more heterogeneous diagnostic grouping than irritable bowel syndrome; if many organic cases were included in the dyspepsia group, this might explain why the impaired mental composite score was independent of psychological status. Whatever the explanation, dyspepsia and irritable bowel syndrome appear to be distinct in terms of their impact on the quality of life, and somatization must be controlled for when considering this outcome in irritable bowel syndrome.
The present study evaluated well-characterized subjects with irritable bowel syndrome and dyspepsia, as well as controls, all from the same community, during the same time period, without regard to health care seeking. This should have removed any bias exerted by health care seeking behaviour on health-related quality of life scores. We did not use an irritable bowel syndrome or dyspepsia disease-specific instrument to measure the quality of life as no disease-specific measure can be applied to both with confidence, but instead we opted for a well-validated generic questionnaire. It may be argued that subjects who respond to postal surveys will have a vested interest in their health and may constitute the ‘worried well’ who subjectively feel their quality of life to be impaired. The 85% response rate should make any such bias of minimal or no importance. Of note was the fact that the composite scores were slightly higher than those in previously reported population norms. We suspect that this reflects the source of our study subjects. Subjects were selected on the basis of their responses to a survey sent by mail to a random sample of the community. Subjects with chronic health conditions other than functional gastrointestinal complaints were excluded. Thus, the study subjects were relatively healthy.
The definitions applied to categorize the subjects could conceivably have affected the findings. We took any upper abdominal pain or discomfort to fulfil a definition of dyspepsia. For irritable bowel syndrome, we used the Manning criteria, which are likely to be more sensitive but probably less specific than the Rome criteria.26,49–51 There was no minimal symptom time limitation for entry to the study. For example, if abdominal pain occurred less than six times a year, the subject was still included in the analysis. The broad definitions and lack of a minimum pain frequency, in addition to the sample size available, would probably make it harder to demonstrate differences between groups. In contrast, Gralnek et al. used more stringent criteria for the definition of irritable bowel syndrome, and hence a large proportion of patients were excluded from the original sample.47 Previous studies have also shown a significant fluctuation in the reporting of irritable bowel syndrome and dyspepsia symptoms.32 Some of our controls may have had symptoms of functional gastrointestinal disorders in the past, which could perhaps have diminished our chances of identifying differences. The fact that we still managed to show a significant difference between controls and all functional gastrointestinal sub-groups adds further weight to the conclusions.
In summary, the health-related quality of life was more impaired in subjects with functional gastrointestinal disorders, even though they remained in the community and were not actively seeking health care. Therapeutic interventions in irritable bowel syndrome and dyspepsia need to target not only changes in disease activity, but also the patient's overall well-being and burden of illness. Somatization should be considered when assessing the quality of life as an outcome in future irritable bowel syndrome trials, as our data suggest that psychological factors account in large part for the reduced quality of life scores in irritable bowel syndrome.
This study was sponsored in part by Novartis and supported by grants from the American College of Gastroenterology, the National Institutes of Health (AGO9440, Rochester Epidemiology Project grant AR30582, General Clinical Research Center grant M01RR00585) and the Mayo Foundation.
The authors wish to thank Carol Van Dyke for her assistance in co-ordinating the study and Karen Kruger and Mary Jo Philo for their assistance in the preparation of the manuscript.