Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Psychological illness is commonly associated with functional gastrointestinal disorders and is important to consider during patient consultation: a population-based study

Psychological illness is commonly associated with functional gastrointestinal disorders and is... Background: Some individuals with functional gastrointestinal disorders (FGID) suffer long-lasting symptoms without ever consulting their doctors. Our aim was to study co-morbidity and lifestyle differences among consulters and non-consulters with persistent FGID and controls in a defined adult population. Methods: A random sample of the general adult Swedish population was obtained by a postal questionnaire. The Abdominal Symptom Questionnaire (ASQ) was used to measure GI symptomatology and grade of GI symptom severity and the Complaint Score Questionnaire (CSQ) was used to measure general symptoms. Subjects were then grouped for study by their symptomatic profiles. Subjects with long-standing FGID (n = 141) and subjects strictly free from gastrointestinal (GI) symptoms (n = 97) were invited to attend their local health centers for further assessment. Results: Subjects with FGID have a higher risk of psychological illness [OR 8.4, CI (4.0–17.5)] than somatic illness [OR 2.8, CI (1.3–5.7)] or ache and fatigue symptoms [OR 4.3, CI (2.1–8.7)]. 95 95 Subjects with psychological illness have a higher risk of severe GI symptoms than controls; moreover they have a greater chance of being consulters. Patients with FGID have more severe GI symptoms than non-patients. Conclusion: There is a strong relation between extra-intestinal, mental and somatic complaints and FGID in both patients and non-patients. Psychological illness increases the chance of concomitantly having more severe GI symptoms, which also enhance consultation behaviour. Background in the Additional File.) The most common symptoms are Gastrointestinal (GI) symptoms are common and are gastroesophageal reflux symptoms/disease (GERS) and reported within three months by almost every other adult dyspepsia originating from the upper GI tract, and Irrita- in Western countries (UK, Sweden, USA, Australia) [1-5]. ble Bowel Syndrome (IBS) from the lower GI tract. Preva- (The abbreviations used in this manuscript are explained lence rates are reported to be 25% for both GERS and Page 1 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 dyspepsia and about 12% for IBS [6]. Although intermit- Methods tent, these disorders may linger in many sufferers [2,7]. Setting In January 1995 there were a total of 21,545 Swedish citi- The total cost of dyspepsia to society, with and without peptic ulcer disease, GERS and IBS, is considerable [8,9]. zens in the Östhammar community, 14,932 of whom Thus, these disorders constitute a major public health were born between 1909 and 1974. Data from three prior problem. studies[2], which included random samples of the Öst- hammar population, were used to provide symptom-free Dyspepsia without peptic ulcer disease (PUD) (or any controls in the current study. We called this new study the other organic GI disease) is called functional dyspepsia Gastrointestinal Consult Study (GiCon). A sample was [10]. In a Swedish population-based upper endoscope drawn from the National Swedish Population Registry study, the prevalence of dyspepsia was 38% and the prev- 1995, now involving all Swedish citizens born between alence of PUD 4%, with about a quarter of the PUD sub- 1909 to 1974 and thus 20 to 87 years of age, born on day jects having no GI symptoms [48]. Thus, the vast majority 3, 12 or 24 of each month (n = 1537), i.e. using the same of those reporting dyspepsia in the population can be date of birth criteria as in previous studies on the same expected to have functional dyspepsia. IBS is a functional population in 1988 and 1989. Thus, we increased the gastrointestinal disorder [11]. Although functional dys- upper age limit to 87 years in order to include all prior pepsia and IBS are considered separate disorders [10,12], participants. As a consequence of the sampling strategy, many subjects report concomitant symptoms thought to all subjects younger than 27 years of age were included for originate from both the upper and the lower the GI tract the first time (n = 86). Most the participants were now [1,13]. Also, sufferers may experience a change in pre- being approached for the third time. Of the 1537 subjects dominant symptom profile over time between dyspepsia in the study base (see Figure 1), 105 were excluded due to and IBS, but much less towards GERS [2]. This means that having declined further participation in the 1988 study dyspepsia and IBS sufferers probably have common and 4 were excluded due to unidentified ID. Thus, 1428 aspects of both pathophysiology and health care seeking subjects were still eligible. They were sent a validated behavior. Furthermore GERS, in a majority of cases, has a postal questionnaire, called "The Abdominal Symptom proven (non-functional) acid-related etiology [11]. Questionnaire" (ASQ) [21], described below. Two Accordingly, it seems reasonable to exclude subjects with reminders were sent out when necessary. Drop-out rea- only GERS symptoms from the functional definition and sons are shown in Figure 1, together with the mean age to investigate the burden on society of functional dyspep- and sex distributions of the 911 responders who consti- sia and IBS taken together. We thus label these conditions tuted the population sample of the current GiCon study. as functional gastrointestinal disorders or FGID. Formation of the study groups and symptom classification A minority of subjects with FGID (5–20%) are reported to All 911 responders in the population sample were classi- visit a doctor quarterly for their complaints [14,15]. On fied according to their response in the ASQ (see Figure 1) the other hand, up to half of those with FGID never see a as having either Functional Gastrointestinal Disorder doctor despite prolonged symptoms [14,16]. For IBS, the (FGID, n = 244), i.e. functional dyspepsia or irritable proportion is even higher, up to 80 % [14]. Thus, GI bowel syndrome, or being Strictly Symptom Free (SSF, n patient-based data will not cover all aspects of morbidity. = 219). Moreover, subjects reporting minor symptoms Moreover, it has been reported previously that most that did not fulfil the criteria for dyspepsia or irritable patients' sick leave is due to disorders other than their bowel syndrome, or reporting isolated symptoms of gas- abdominal complaints. Only 23% of those subjects with troesophageal reflux symptoms (GERS, n = 271), were abdominal pain who were on sick leave stated that excluded. Subjects who were free from abdominal symp- abdominal pain was the cause of their absenteeism [17]. toms in this 1995 survey but had reported symptoms in the previous studies (1988–1989) were labelled as Prior There are prior cross-sectional population-based point Symptomatic (n = 177) and were also excluded. prevalence studies on GI co-morbidity [18-20]. However, there is a lack of knowledge of the co-morbidity of sub- The FGID and SSF sample groups (n = 244 + 219 = 463) jects with proven chronic/long lasting/persistent FGID were then invited by post to participate in the current and its impact on health care seeking behaviour due to GI investigation and to visit one of the six local health cen- symptoms. The aim of this study was to compare the co- tres. Subjects who did not respond to the first letter were morbidity and its relation to health care seeking behav- sent a reminder, and non-responders to the second letter iour among non-patients with persistent FGID with those were finally contacted by phone. In total, 187 (77%) with who are persistently GI symptom free, in a randomly- FGID and 156 (71%) with SSF accepted the invitation. selected adult Swedish population. There were 117 (57 FGID, 60 SSF) subjects with incom- plete (n = 12) or no (n = 105) response. At the health Page 2 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Sampling Figure 1 procedure of the Gastrointestinal Consult Study (GiCon study) Sampling procedure of the Gastrointestinal Consult Study (GiCon study). Sampling procedure of the Gastrointesti- nal Consult Study (GiCon study) with relation to the prior study of 1995. Page 3 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 GI symptoms, pain mo Figure 2 dalities and location GI symptoms, pain modalities and location. The symptoms inquired are shown in Part 1, and the pain and discomfort modalities with the master sketch for indicating their abdominal location in Part 2. The asterisks are explained in the text. Swedish laymen terms were used in the questionnaire. The master sketch shows the eligible pain locations. centres, the participants filled in an ASQ and a Complaint Questionnaires Score Questionnaire (CSQ), as described below, and the The abdominal symptom questionnaire height and weight of each was measured. The time period The Abdominal Symptom Questionnaire (ASQ) has been between the original 1995 ASQ and the second ASQ sur- validated previously and found to be reliable and repro- vey in 1996 ranged from 7 to 15 months. Of the FGID ducible [2,13,22]. In the questionnaire, subjects were group of 187 subjects, 8 subjects who now reported no asked if they had been troubled (Yes/No) by any of the 29 symptoms and 38 with minor symptoms were excluded; listed general gastrointestinal symptoms over the previous and out of the 156 SSF, 13 who now reported FGID and three months. They were also asked if they had been trou- 46 with minor symptoms were also excluded. Thus, data bled by any of 11 listed descriptors of abdominal pain and from 141 FGID subjects (mean age 45.7 y, 34% men) and where the symptom was located (upper, centre or lower 97 SSF subjects (mean age 52.4 y, 48% men) remained in abdominal, right and left flank, as shown in Figure 2). A the analysis. The sampling procedure for retrospective similar ASQ was used for the postal survey and the surgery data – twice over one year for those with, and up to four visit, with the latter extended to include a symptom sever- times over eight years for those without, GI symptoms – ity Likert scale graded from zero to seven for each symp- assured persistent symptomatic status. tom asked for. In the analysis, the data were pooled into a three-grade scale (1, 1–4, 5–7) Page 4 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Definitions of symptoms from the ASQ The complaint score questionnaire (See Figure 2) [23]. The Complaint Score Questionnaire contains 30 ques- tions, as shown in Figure 3, indicating the presence or Dyspepsia was defined as one or more of the symptoms absence of 30 different symptoms [25]. The questions are (reflux episodes, heartburn, retrosternal pain, nausea, dichotomous and can be categorized into six domains: vomiting, early satiety, uncomfortable feeling of fullness abdominal/urinary, ache/fatigue, muscular-skeletal, after meals, abdominal distension), and one or more of nutritional, cardio-vascular and depressive. the pain modalities (burning sensation, aching, pain, ten- derness, sinking feeling, "butterflies", cramp, twinge, Other questions stitch, colic, gripes) with any abdominal location except The participants were asked to state their coffee, alcohol the lower part, but no concomitant IBS. and tobacco consumption, and whether they had ever had peptic ulcer disease or had ever consulted a doctor for GI Irritable Bowel Syndrome (IBS) was defined as one or more complaints. Also, the past three months of general pain of the symptoms (abdominal distension, abdominal dis- and all GI medication were indicated. Educational back- comfort or pain on defecation, abdominal discomfort or ground was registered at five levels (elementary, compre- pain relieved by defecation, feeling of incomplete defeca- hensive, secondary, upper secondary, university) and tion, mucous stools) and one or more of the symptoms medical knowledge was evaluated by means of a self- (diarrhea, constipation, alternating diarrhea and consti- explanatory questionnaire (see Additional File). The pation) together with one or more of the pain modalities answers were scored with a sum of 0–15. (burning sensation, aching, pain, tenderness, sinking feel- ing, "butterflies", cramp, twinge, stitch, colic, gripes), with Statistical power and analysis any location. One hundred and twenty three subjects were required (in each the FGID and SSF groups) in order to have a 90% When using these definitions of dyspepsia and IBS, power at the P < 0.05 level to detect a 20% absolute differ- responders with only reflux symptoms, i.e. heartburn ence in exposure variables. This assumed a 24% preva- and/or retrosternal pain but no abdominal pain or dis- lence of FGID in the population, equal numbers of comfort, were classified as having GERS and not dyspep- subjects in the FGID and SSF groups, 15 and 20% absolute sia, while those with such symptoms and abdominal pain difference in the exposure variables within the two steps or discomfort fell into the dyspepsia group. Also, IBS, as of the study, 75% response rate on the ASQ, and 25% defined below, was given priority over dyspepsia. Thus, exclusion from each group in the last step. The variables concomitant occurrence of both led to a diagnosis of IBS. analysed are presented in the Additional File. Univariate This definition is in accordance with published guidelines analyses were performed using Student's t-test, Wilcoxon's [24]. ranksum test (Mann Whitney) and Pearson's chi-2 test. Multivariate analysis with FGID as the dependent variable Functional Gastrointestinal Disorder, FGID was performed using a logistic regression, and with dys- FGID was defined as either dyspepsia or IBS. peptic severity as the dependent variable with the ordinal logistic regression technique. Age and BMI were linear to Minor symptoms refer to symptoms not fulfilling the crite- outcome and were thus handled as continuous variables. ria of dyspepsia or IBS. By definition, those with GERS To test the symptoms in the CSQ, we made an age and sex were included in this group adjusted logistic regression model for each symptom and accepted a P value less than 0.05/(number of symptoms) "Strictly Symptom Free" (SSF) is defined as having no = 0.05/30 ≈ 0.0016 as significant. The explanatory varia- reported symptoms at all in the Abdominal Symptom bles were evaluated by a logistic regression in a sex and age Questionnaire in the 1995 survey, and subjects stating adjusted model. that they had not even been troubled previously with abdominal symptoms. Those subjects who had partici- The association between each potential determinant pated in the two former surveys in 1988 and 1989 (n = obtained from the questionnaires and the presence of 265) should also have reported no symptoms in each of FGID was quantified by using odds ratios and 95% confi- those two investigations. dence intervals. All exposure variables with P < 0.25 were then entered together in a multivariate logistic regression "Prior symptomatics" were those with no symptoms in the model to evaluate which was independently associated 1995 survey but with GI symptoms reported in the prior with the presence of FGID. A factor analysis was per- studies of 1988 and 1989. formed using all 30 complaints and factors with eigenval- ues greater than 1.3. Four factors were obtained by a principal components analysis with varimax rotation. The Page 5 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Figure 3 Proportion of subjects with FGID and with SSF reporting complaints during the previous three months in the CSQ Proportion of subjects with FGID and with SSF reporting complaints during the previous three months in the CSQ. Proportions (0–1.0) of subjects with FGID and with SSF (n = 238) reporting complaints during the previous three months in the Complaint Score Questionnaire (CSQ). There were significance differences *(p < 0.0016) on an age and sex adjusted logistic regression for all variables except coughing (ns) and excessive weight (ns). 30 variables were dichotomized with the highest 1/3 Results given the value 1 and the rest the value 0. A logistic regres- Representativeness of study sample sion was performed with the four factors age and sex In order to investigate the effect of the drop-outs during adjusted, and factors with p > 0.05 were excluded. To test the sampling procedure, the mean ages, sex ratios and whether the model fitted the data, a Pearson goodness-of- education levels in the corresponding FGID and SSF fit test with p values greater than 0.05 was performed. groups were analysed as shown in Table 1, for the samples When the number of covariates approximated the number from 1988, 1995 and 1996 (see Table 1) [13]. There were of observations, the Hosmer-Lemeshow test was per- no significant differences in any of these aspects (see Table formed to determine whether the model fitted the data. 1). For the ordinal logistic model, comparison with a multi- nomial model made an approximate fit test. No interac- Study group characteristics tions were found between the variables in the main As shown in Table 2, there were more females among those model. Ninety five per cent confidence intervals (CI) were with FGID. However, there were no intergroup differences in computed using parametric methods. A p-value of 0.05 or education, medical knowledge, BMI, intake of coffee, alcohol less was generally regarded as statistically significant. All and smoking. The age difference was irrelevant, as those with tests were two-tailed and the statistical package Stata 8 was SSF had been largely excluded due to prior study results. Dis- used for analysis [27]. ease-related variables were significantly different between the study groups. These variables were introduced to further model- Ethics ling, as shown below. The variables 'intake of GI medicine' and This study was approved by the Ethics Committee of the 'previous PUD' were not included due to sparse data. Medical Faculty, Uppsala University, June 5th 1996. Page 6 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Table 1: Comparison between previous population studies in Östhammar, Sweden. Age, sex and education level at different stages of the sampling process. From the first population sample in 1988 to the present study 1996. ns = p > 0.05. Group (G) Sample Year n Age years mean (SD) Sex % males Education median level (range) 1 First population sample 1988 1156 48.9 (16.0) 50.4 3 (1–4)* 2 Eligible sample 1995 1428 49.9 (17.2) 49.9 3 (1–4)** 3 Population sample 1995 911 49.2 (16.46) 47.0 3 (1–4) 4 Sample group FGID 1995 244 45.5 (15.3) 36.1 3 (2–4) 5 Sample Group SSF 1995 219 51.7 (17.6) 51.6 3 (1–4) 6 Study Group FGID 1996 141 45.7 (14.3) 34.0 3 (2–4) 7 Study Group SSF 1996 97 52.4 (15.3) 48.0 3 (1–4) G2 vs G3 ns ns G2 vs G3 ns ns ns G4 vs G6 ns ns ns G5 vs G7 ns ns ns *responders = 1156 **responders n = 1384 Table 2: Comparison between explanatory variables for subjects with FGID and SSF. Comparison between explanatory variables for subjects with FGID and SSF adjusted for sex and age. Ordinal variables are presented as median (range), dichotomous variables as proportion %, and continuous variables as mean (SD). Variable FGID SSF P n = 141 n = 97 AGE 45.7 (14) 52.4 (15) not relevant SEX (female %) 66 53 0.026 *** BMI 26.3 (4.7) 26.2 (4.2) Ns * Education 3 (2–4) 3 (1–4) Ns ** Medical knowledge 10 (4–15) 11 (5–15) Ns ** GI sympt severity 4 (3–5) 0 (0–0) <0.0005 ** GI Consultation (ever) 72% 8% <0.0005 *** Pain medicine (3 month) 77% 34% <0.0005 *** GI medicine (3 month) 32% 1% <0.0005 *** Previous PUD (ever) 12% 1% 0.006 *** Coffee 2 (2–3) 2 (2–3) Ns ** Alcohol 4 (3–4) 3 (2–4) Ns ** Smoking 1 (1–2) 1 (1–2) Ns ** * = Student's t-test ** = Mann-Whitney test *** = Pearson Chi-2 test CSQ and FGID vs. SSF cant for all except four complaints. After adjusting for The results from the 30 CSQ complaints for the FGID and alcohol and pain and GI drug intake (Table 3), 20 SSF study groups are presented in Figure 3. Those with complaints remained significant. A factor analysis was persistent FGID scored statistically higher on all variables performed including the 26 non-GI complaints. After a except difficulties in passing urine, excessive weight, varimax rotation of the four factors with eigenvalues > 1.3, coughing and impaired hearing. we found four factors representing psychological illness, somatic illness, ache/fatigue and one "miscellaneous" Risk modelling (Table 3). Each factor was then introduced in a logistic Risks of reported CSQ complaints for FGID vs. SSF, regression model adjusted for sex and age (Table 3), and expressed as age and sex adjusted OR, are presented in the "miscellaneous" factor was shown to be non-signifi- Table 3. GI complaints (abdominal pain, nausea, diarrhea cant. In the last sequential analysis, the three factors that and constipation) were excluded as we aimed to analyze remained significant in the prior analysis were introduced the co-morbidity with GI symptoms. The OR was signifi- Page 7 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Table 3: Odds ratios of FGID/SSF for complaints in the Complaints Score Questionnaire (CSQ). Odds ratios (OR, with 95% confidence intervals (CI)) of FGID/SSF (n = 238) for complaints elicited by the CSQ. Logistic regression is presented in different models. A factor analysis extracted four factors: A = psychological illness factor, B = somatic illness factor, C = miscellaneous factor, D = ache/fatigue factor. These were used in the modelling in the right two columns. I II III IV V Symptom OR (CI) by Models OR (CI) by Models FACTOR OR (CI) by Models adjusted OR (CI) by a main effect model adjusted for sex adjusted for sex, for sex and age adjusted for sex and age and age age, alcohol, pain tablets, GI-tablets SSF (all variables) 1 (Reference) 1 (Reference) FGID 1) Psychological illness 1) Psychological illness Cries easily 6.7 (2.3–19.9) 9.8 (2.0–47) A 8.0 (4.1–15.8) 8.4 (4.0–17.5) Sleeping disturbance 6.2 (2.7–14.0) 3.2 (1.3–8) A General fatigue 14.5 (7.4–28.7) 12.6 (5.3–30) A, D Irritability 8.8 (4.1–17.8) 5.6 (2.3 – 13.7) A, D Nervousness 18.4 (4.2–80.3) 14.3 (2.8 – 72) A Impaired concentration 19.0 (5.7–63.8) 15.3 (4.0 – 58) A Difficulty to relax 15.7 (6.0–41.5) 10.9 (3.7–32) A Restlessness 40.0 (9.4–170) 32.2 (6.7–154) A Depression 8.6 (4.1–18) 4.7 (2.0 – 11) A Exhaustion 12.7 (4.4–37) 9.1 (2.7–30) A 1) Somaticillness 1) Somaticillness Chest pain 40.0 (5.3–300) B 3.7 (2.0–7.1) 2.8 (1.3–5.7) Pain in the joints 6.2 (2.8–13.6) 7.5 (2.6–21) B Pain in the legs 4.4 (2.2–8.9) 3.8 (1.6–9.3) B Overweight 2.0 (1.0–4.0) 1.4 (0.6–3.6) B Breathlessness 8.9 (3.2–25) 12.1 (3.3–45) B Dizziness 10.1 (4.2–24) 11.4 (3.8–34) B Impaired hearing 3.0 (1.3–6.8) 3.1 (1.0–9.5) B Eye problem 4.2 (1.9–9.1) 3.4 (1.3–9.0) B 1) Miscellaneous Loss of weight - - C 1.7 (0.9–3.0) Bad appetite - - C Feeling cold 7.3 (3.1–17) 7.0 (2.6–19) C Difficulty in passing urine 9.6 (2.0–47) 9.1 (1.4–59) C 1) Ache/fatigue 1) Ache/fatigue Back ache 4.4 (2.4–8.2) 2.0 (0.9–4.3) D 2.9 (1.6–5.2) 4.3 (2.1–8.7) Headache 6.3 (3.4–12) 4.1 (1.9 – 9.1) D Sweating 3.6 (1.7 – 7.4) 3.3 (1.3 – 8.5) Coughing 2.0 (0.98–4.2) 1.7 (0.6–4.4) 1) Reference group (OR = 1) is those coded 0 in each factor together into a main effect model, adjusted for age and ASQ symptom severity and consulting behaviour vs CSQ factors sex. The OR for these factors remained significant, as shown in Table 3. From the ASQ, the mean grades of GI symptom severity for affirmative symptoms were analyzed against the three Consulters versus non consulters final CSQ factors from Table 3 (psychological illness, Consulters and non-consulters among those with persist- somatic illness, ache/fatigue) and for age, sex and consult- ent FGID were compared regarding their complaints, as ing behaviour, as shown in Table 4. The analysis showed reported in the CSQ. The proportion (0–1.0) per an obviously higher risk of increased GI symptom severity complaint is shown in Figure 4. There were no statistically for consulters (OR 12.3) and for psychological illness (OR significant differences between the consulters and non- 4.5), while somatic illness and ache/fatigue had a low risk, consulters for any of the complaints (adjusted for age and with the 95% CI close to 1.0. From the ASQ, consulting sex, p > 0.0016). behaviour was analyzed for the CSQ factors psychological illness, abdominal illness, age and sex (see Table 5). The analysis showed a greater chance of being a consulter for abdominal illness (OR 2.0) and psychological illness (OR 2.2). Page 8 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Consulters Figure 4 Proportion of complaints from the Complaint Score Questionnaire among those with FGID, divided into Consulters and Non- Proportion of complaints from the Complaint Score Questionnaire among those with FGID, divided into Con- sulters and Non-Consulters. Proportion (0–1.0) of complaints from the Complaint Score Questionnaire (CSQ) among those with FGID, divided into Consulters and Non-Consulters (n = 141). None of the variables showed a significant difference between Consulters and Non-Consulters for P values less than 0.0016, tested by a sex and age adjusted logistic regression. Table 4: Odds ratios of graded GI symptom severities in the Discussion ASQ for consulting, psychological illness, somatic illness and This study shows that among subjects with longstanding ache/fatigue factors. Odds ratios (OR, with 95% confidence FGID, there is a remarkably high prevalence of psycholog- intervals (CI)) of graded (0,1,2) GI symptom severity in the ASQ ical illness and also of non-GI somatic complaints. These for consulting, psychological illness, somatic illness and ache/ fatigue factors, age and sex, for both FGID and SSF (n = 232). are present regardless of whether the subjects have con- Ordinal logistic regression. sulted their doctor about their GI problems, and are more severe in subjects with persistent FGID. Although FGID Variable OR (CI) was more common in women, the consultation rates in sufferers were similar for males and females and were not Psychological illness low 1 age-related. Only about a quarter of the sufferers had high 4.5(2.4–8.4) Somatic illness low 1 never consulted their doctor. high 2.0(1.1–3.8) Ache/fatigue low 1 We consider that our findings can be generalized to the high 2.1(1.1–4.2) whole population, as the study groups were sampled from Consulters no 1 a well-defined and thoroughly investigated population, yes 12.3(6.3–23.9) most of whom had participated in prior studies [28,29]. Age (continuous) 0.96 (0.94–0.99) The original study base was formed in 1988 from the Sex female 1 male 0.9 (0.5–1.7) Swedish National Population register, which guarantees complete coverage of all citizens. There were no differ- ences in age, gender or education level between the study Page 9 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Table 5: Odds ratio of consulting for abdominal complaints in the Subjects with FGID were on average younger than con- ASQ, 1995, for psychological illness, abdominal illness, age and trols without FGID, which may be expected as the preva- sex, for both FGID and SSF. Odds ratio (OR, with 95% lence of dyspepsia and IBS is generally higher in younger confidence intervals (CI) of consulting for abdominal complaints age groups [32]. This study was not a case control study, in the ASQ for 1995 for psychological illness, abdominal illness, age and sex for both FGID and SSF (n = 232). Logistic regression. however, but a study of all subjects with FGID and SSF Somatic illness and ache/fatigue were excluded in the final model within the cohort. Any differences caused by this grouping because they showed no significance in the prior step. strategy were controlled for in the analysis by gender- and age-adjusted logistic regression. There was a particular Variable OR (CI) association between FGID and psychological illness, Psychological illness low 1 although "fibromyalgia-like" symptoms (ache and High 2.2 (1.2–4.0) fatigue) [49] and other somatic complaints were also Abdominal illness low 1 common. Previous outpatient studies have shown that High 2.0 (1.1–3.8) IBS is associated with psychiatric illnesses such as depres- Age (continuous) 1.0 (0.99–1.0) sion [34], dysthymia [35] and anxiety [36], and similar Sex female 1 findings are reported for dyspepsia [50,51]. Greater men- Male 0.7 (0.40–1.3) tal pathology has been reported for consulters, i.e. patients, than non-consulters with IBS [44,45,52], but this has not been demonstrated convincingly in dyspepsia subjects [53]. and the sample groups or between the population sam- Healthcare seeking behaviour is complex and until now it ples from 1988 and 1995. Also, the proportions of those has been studied largely in patient samples. Nyrén et al. reporting symptoms explicable in terms of an organic dis- found that patients with non-ulcer dyspepsia had an ease have been shown to be insignificant [13]. Thus, the excessive need for sick leave compared with ulcer patients, FGID subjects and the symptom-free subjects would seem but that the predominant reason for leave was related to to be representative. musculoskeletal rather than abdominal symptoms [41]. Kettell et al. [42] found that severity of abdominal pain The validity of the research tool is a potential source of and anxiety about the seriousness of their condition were bias. However, for symptom reporting, only well-vali- important factors in patients consulting for IBS [42]. dated questionnaires were used. Both the ASQ and the There seem to be no differences in the use of healthcare CSQ have been adequately validated [28-30]. The psycho- services or co-morbidity status between the year before logical illness factor identified in this study embraces and the year after diagnosis of IBS [43,46]. symptoms of both "neurotic" and "personality" dimen- sions, listed in Table 3, and the outcome seems plausible. Although this study was focused on total co-morbidity, in The questionnaire used to assess medical knowledge was terms of consultation, other factors associated with FGID simple and straightforward, with kappa values per ques- consultation were also considered. The total consultation tion 1 = 0.70, 2 = 0.89, 3 = 0.47, 4 = 0.78, 5 = 0.80, 6 = pattern for the subjects will be published elsewhere. In 0.70 when repeated within a week, considered acceptable essence, effective treatment of patients with FGID involves for all with some reservation in the 0.47 case [53]. not only addressing GI discomfort, but also considering mental and somatic complaints such as depression and The definitions of dyspepsia and IBS used in this study exhaustion. were those used in the original study from 1988 [31], when the Rome II criteria [11] were unavailable. We opted Conclusion to retain our original study definitions despite ongoing In the present study, psychological illness proved to be an changes. important co-morbidity factor among subjects with FGID, and the severities of the two were linked. We cannot con- Our definition of dyspepsia was more restricted in terms clude anything about the cause of the relationship. The of symptoms than the Rome II definition, but wider in presence of psychological illness was also associated with terms of abdominal location, as not only epigastric but a greater need for medical consultation. Fear of life-threat- also mid abdominal symptoms were included. The IBS ening illness has been reported to be an important reason definition used was in accordance with the Rome II crite- for consultation in FGID [1], and this worry and anxiety ria [22]. Consequently, we consider the overall prevalence needs to be taken into account when attempting to man- of FGID in this study and the concordance on an individ- age FGID successfully. Somatic co-morbidity was found to ual level to be applicable within today's definitions [11]. be a less important reason for consultation, although a high proportion of subjects with FGID (77%) in our study Page 10 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 14. Agreus L: Socio-economic factors, health care consumption were taking analgesics while only 32% used specific gas- and rating of abdominal symptom severity. A report from trointestinal medication. the abdominal symptom study. Fam Pract 1993, 10:152-63. 15. Haycox A, Einarson T, Eggleston A: The health economic impact of upper gastrointestinal symptoms in the general popula- Competing interests tion: results from the Domestic/International Gastroenter- The author(s) declare that they have no competing ology Surveillance Study (DIGEST). Scand J Gastroenterol Suppl 1999, 231:38-47. interests. 16. Westbrook JI, McIntosh J, Talley NJ: Factors associated with con- sulting medical or non-medical practitioners for dyspepsia: Authors' contributions an Australian population-based study. Aliment Pharmacol Ther 2000, 14:1581-8. TÅ participated in the study design and checking the cod- 17. Nyren O, Lindberg G, Lindstrom E, Marke LA, Seensalu R: Eco- ing of the data, carried out the construction of the "medi- nomic costs of functional dyspepsia. Pharmacoeconomics 1992, cal knowledge questionnaire" including a test-retest 1:312-24. 18. Stanghellini V: Relationship between upper gastrointestinal survey, distributed and collected the questionnaires, symptoms and lifestyle, psychosocial factors and comorbid- called the participants to his health centre, supported the ity in the general population: results from the Domestic/ International Gastroenterology Surveillance Study other health centers personnel, performed all the statisti- (DIGEST). Scand J Gastroenterol Suppl 1999, 231:29-37. cal analysis, participated in writing the manuscript, and 19. Haug TT, Mykletun A, Dahl AA: Are anxiety and depression submitted the final manuscript. LA carried out the study related to gastrointestinal symptoms in the general population? Scand J Gastroenterol 2002, 37:294-8. design and contributed to the statistical analysis and the 20. Koloski NA, Talley NJ, Boyce PM: Epidemiology and health care writing of the manuscript. KS contributed to checking the seeking in the functional GI disorders: a population-based study. Am J Gastroenterol 2002, 97:2290-9. coding of the data and participated in the study design. SJ 21. Agreus L, Svardsudd K, Nyren O, Tibblin G: Reproducibility and contributed help with the statistical analysis. validity of a postal questionnaire. The abdominal symptom study. Scand J Prim Health Care 1993, 11:252-62. 22. Agreus L, Talley NJ, Svardsudd K, Tibblin G, Jones MP: Identifying Acknowledgements dyspepsia and irritable bowel syndrome: the value of pain or This work was supported by the personnel of the Primary Health Care discomfort, and bowel habit descriptors [In Process units of Östhammar, the Uppsala County Council and Anna Hedin Ph D, Citation]. Scand J Gastroenterol 2000, 35:142-51. National Institute of Public Health, Sweden. We also thank Madeline Frame 23. Agréus L, Svärdsudd K, Tibblin G, Nyrén O: The epidemiology of dyspepsia: Symptom clusters and demographic for assistance with the manuscript preparation. characteristics. Gastroenterology 1991, 100:A1. 24. Talley NJ: Working Team Report. Functional Dyspepsia: A References Classification with Guidelines for Diagnosis and 1. Jones R, Lydeard SE, Hobbs FD, Kenkre JE, Williams EI, Jones SJ, Jones Management. Gastroenterology International 1991, 4:145-160. , Repper JA, Caldow JL, Dunwoodie WM, Bottomley JM: Dyspepsia 25. Tibblin G, Tibblin B, Peciva S, Kullman S, Svardsudd K: "The Gote- in England and Scotland. Gut 1990, 31:401-5. borg quality of life instrument" – an assessment of well-being 2. Agreus L, Svardsudd K, Talley NJ, Jones MP, Tibblin G: Natural his- and symptoms among men born 1913 and 1923. Methods tory of gastroesophageal reflux disease and functional and validity. Scand J Prim Health Care Suppl 1990, 1:33-8. abdominal disorders: a population-based study. Am J 26. World Health Organization: WHO/46 Press Release: World Health Gastroenterol 2001, 96:2905-14. Organization 1997. 3. Agreus L: Natural history of dyspepsia. Gut 2002, 50:iv2-9. 27. StataCorp: Stata Statistical Software: Release 8.0 College Station, TX: 4. Talley NJ, Zinsmeister AR, Schleck CD, Melton LJ III: Dyspepsia and Stata Corporation; 2003. dyspepsia subgroups: a population-based study. Gastroenterol- 28. Agréus L, Svärdsudd K, Nyrén O, Tibblin G: Reproducibility and ogy 1992, 102:1259-68. validity of a postal questionnaire. The abdominal symptom 5. Talley N, Boyce P, Jones M: Identification of upper and lower study. Scand J Prim Health Care 1993, 11:252-62. gastrointestinal symptom groupings in an urban population. 29. Agréus L: The Abdominal Symptom Study. An Epidemiologi- Gut 1998, 42:690-95. cal Survey of Gastrointestinal and Other Abdominal Symp- 6. Agréus L, Talley NJ: Dyspepsia: current understanding and toms in the Adult Population of Östhammar, Sweden. In PhD management. Annu Rev Med 1998, 49:475-93. thesis Uppsala University, Department of Family Medicine; 1993. 7. McDougall NI, Johnston BT, Kee F, Collins JS, McFarland RJ, Love AH: 30. Tibblin G, Cato K, Svärdsudd K: Goteborg quality of life study of Natural history of reflux oesophagitis: a 10 year follow up of men born in 1913 and 1923 – age, sex, job satisfaction and its effect on patient symptomatology and quality of life. Gut cardiovascular diseases. Scand J Prim Health Care 1990, 1996, 38:481-6. 1(Suppl):39-45. 8. Agreus L, Borgquist L: The cost of gastro-oesophageal reflux 31. Agréus L, Nyrén O, Svärdsudd K, Tibblin G: Ont i magen-En epi- disease, dyspepsia and peptic ulcer disease in Sweden. Phar- demiologisk studie om bukbesvär i Östhammars kommun. macoeconomics 2002, 20:347-55. Svenska Läkarsällskapets Hygea 1990, 99(1):90. 9. Leong SA, Barghout V, Birnbaum HG, Thibeault CE, Ben-Hamadi R, 32. Agréus L, Svärdsudd K, Talley NJ, Jones MP, Tibblin G: Natural his- Frech F, Ofman JJ: The economic consequences of irritable tory of gastroesophageal reflux disease and functional bowel syndrome: a US employer perspective. Arch Intern Med abdominal disorders: a population-based study. Am J 2003, 163:929-35. Gastroenterol 2001, 96:2905-14. 10. Drossman DA, Corazziari E, Thompson WG, Talley NJ, Whitehead 33. Gupta S, Masand PS, Kaplan D, Bhandary A, Hendricks S: The rela- W: The Functional Gastrointestinal Disorders. Second edition. tionship between schizophrenia and irritable bowel syn- Degnon Associates, McLean, Va, USA; 2000. drome (IBS). Schizophr Res 1997, 23:265-8. 11. Talley N, Stanghellini V, Heading R, Koch K, Malagelada J, Tytgat G: 34. Masand PS, Kaplan DS, Gupta S, Bhandary AN, Nasra GS, Kline MD, Functional Gastroduodenal Disorders. Gut 1999, 45(Suppl Margo KL: Major depression and irritable bowel syndrome: is 11):1137-42. there a relationship? J Clin Psychiatry 1995, 56:363-7. 12. Whitehead WE, Gibbs NA, Li Z, Drossman DA: Is functional dys- 35. Masand PS, Kaplan DS, Gupta S, Bhandary AN: Irritable bowel syn- pepsia just a subset of the irritable bowel syndrome? Baillieres drome and dysthymia. Is there a relationship? Psychosomatics Clin Gastroenterol 1998, 12:443-61. 1997, 38:63-9. 13. Agreus L, Svardsudd K, Nyren O, Tibblin G: Irritable bowel syn- drome and dyspepsia in the general population: overlap and lack of stability over time. Gastroenterology 1995, 109:671-80. Page 11 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 36. Lydiard RB, Fossey MD, Marsh W, Ballenger JC: Prevalence of psy- chiatric disorders in patients with irritable bowel syndrome. Psychosomatics 1993, 34:229-34. 37. Sullivan G, Blewett AE, Jenkins PL, Allison MC: Eating attitudes and the irritable bowel syndrome. Gen Hosp Psychiatry 1997, 19:62-4. 38. Walker EA, Gelfand AN, Gelfand MD, Green C, Katon WJ: Chronic pelvic pain and gynecological symptoms in women with irri- table bowel syndrome. J Psychosom Obstet Gynaecol 1996, 17:39-46. 39. Woodman CL, Breen K, Noyes R Jr, Moss C, Fagerholm R, Yagla SJ, Summers R: The relationship between irritable bowel syn- drome and psychiatric illness. A family study. Psychosomatics 1998, 39:45-54. 40. Sperber AD, Atzmon Y, Neumann L, Weisberg I, Shalit Y, M Abu- Shakrah, Fich A, Buskila D: Fibromyalgia in the irritable bowel syndrome: studies of prevalence and clinical implications. Am J Gastroenterol 1999, 94:3541-6. 41. Nyrén O, Adami HO, Gustavsson S, Lööf L: Excess sick-listing in nonulcer dyspepsia. J Clin Gastroenterol 1986, 8:339-45. 42. Kettell J, Jones R, Lydeard S: Reasons for consultation in irritable bowel syndrome: symptoms and patient characteristics. Br J Gen Pract 1992, 42:459-61. 43. Ruigomez A, Wallander MA, Johansson S, Garcia Rodriguez LA: One- year follow-up of newly diagnosed irritable bowel syndrome patients. Aliment Pharmacol Ther 1999, 13:1097-102. 44. Ballenger JC, Davidson JR, Lecrubier Y, Nutt DJ, Lydiard RB, Mayer EA: Consensus statement on depression, anxiety, and func- tional gastrointestinal disorders Irritable bowel syndrome, anxiety, and depression: what are the links? J Clin Psychiatry 2001, 62(Suppl 8):48-51. 45. Lydiard RB: Irritable bowel syndrome, anxiety, and depres- sion: what are the links? J Clin Psychiatry 2001, 62(Suppl 8):38-47. 46. Whitehead WE, Palsson O, Jones KR: Systematic review of the comorbidity of irritable bowel syndrome with other disor- ders: what are the causes and implications? Gastroenterology 2002, 122:1140-56. 47. Jarrett ME, Burr RL, Cain KC, Hertig V, Weisman P, Heitkemper MM: Anxiety and depression are related to autonomic nervous system function in women with irritable bowel syndrome. Dig Dis Sci 2003, 48:386-94. 48. Aro P, Ronkainen J, Storskrubb T, Bolling-Sternevald E, Talley NJ, Agréus L: Prevalence of symptoms and upper endoscopic find- ings in a random adult population. A report from Kalixanda study. Gastroenterology 2001, 120(Suppl 1):A-231. 49. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Gold- enberg DL, Tugwell P, Campbell SM, Abeles M, Clark P: The Amer- ican College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Cri- teria Committee. Arthritis Rheum 1990, 33(2):160-72. 50. Talley NJ, Fung LH, Gilligan IJ, McNeil D, Piper DW: Association of anxiety, neuroticism, and depression with dyspepsia of unknown cause. A case-control study. Gastroenterology 1986, 90(4):886-92. 51. Kane F Jr, Strohlein J, Harper RG: Nonulcer dyspepsia associated with psychiatric disorder. South Med J 1993, 86(6):641-6. 52. Drossman DA, McKee DC, Sandler RS, Mitchell CM, Cramer EM, Lowman BC, Burger AL: Psychosocial factors in the irritable bowel syndrome. A multivariate study of patients and non- patients with irritable bowel syndrome. Gastroenterology 1988, 95(3):701-8. 53. Sackett DL: A primer on the precision and accuracy of the clin- Publish with Bio Med Central and every ical examination. JAMA 1992, 267:2638-44. scientist can read your work free of charge "BioMed Central will be the most significant development for Pre-publication history disseminating the results of biomedical researc h in our lifetime." The pre-publication history for this paper can be accessed Sir Paul Nurse, Cancer Research UK here: Your research papers will be: available free of charge to the entire biomedical community http://www.biomedcentral.com/1741-7015/3/8/prepub peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 12 of 12 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Medicine Springer Journals

Psychological illness is commonly associated with functional gastrointestinal disorders and is important to consider during patient consultation: a population-based study

Loading next page...
 
/lp/springer-journals/psychological-illness-is-commonly-associated-with-functional-oQondPPDiF

References (67)

Publisher
Springer Journals
Copyright
Copyright © 2005 by Ålander et al; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; Medicine/Public Health, general; Biomedicine, general
eISSN
1741-7015
DOI
10.1186/1741-7015-3-8
pmid
15892883
Publisher site
See Article on Publisher Site

Abstract

Background: Some individuals with functional gastrointestinal disorders (FGID) suffer long-lasting symptoms without ever consulting their doctors. Our aim was to study co-morbidity and lifestyle differences among consulters and non-consulters with persistent FGID and controls in a defined adult population. Methods: A random sample of the general adult Swedish population was obtained by a postal questionnaire. The Abdominal Symptom Questionnaire (ASQ) was used to measure GI symptomatology and grade of GI symptom severity and the Complaint Score Questionnaire (CSQ) was used to measure general symptoms. Subjects were then grouped for study by their symptomatic profiles. Subjects with long-standing FGID (n = 141) and subjects strictly free from gastrointestinal (GI) symptoms (n = 97) were invited to attend their local health centers for further assessment. Results: Subjects with FGID have a higher risk of psychological illness [OR 8.4, CI (4.0–17.5)] than somatic illness [OR 2.8, CI (1.3–5.7)] or ache and fatigue symptoms [OR 4.3, CI (2.1–8.7)]. 95 95 Subjects with psychological illness have a higher risk of severe GI symptoms than controls; moreover they have a greater chance of being consulters. Patients with FGID have more severe GI symptoms than non-patients. Conclusion: There is a strong relation between extra-intestinal, mental and somatic complaints and FGID in both patients and non-patients. Psychological illness increases the chance of concomitantly having more severe GI symptoms, which also enhance consultation behaviour. Background in the Additional File.) The most common symptoms are Gastrointestinal (GI) symptoms are common and are gastroesophageal reflux symptoms/disease (GERS) and reported within three months by almost every other adult dyspepsia originating from the upper GI tract, and Irrita- in Western countries (UK, Sweden, USA, Australia) [1-5]. ble Bowel Syndrome (IBS) from the lower GI tract. Preva- (The abbreviations used in this manuscript are explained lence rates are reported to be 25% for both GERS and Page 1 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 dyspepsia and about 12% for IBS [6]. Although intermit- Methods tent, these disorders may linger in many sufferers [2,7]. Setting In January 1995 there were a total of 21,545 Swedish citi- The total cost of dyspepsia to society, with and without peptic ulcer disease, GERS and IBS, is considerable [8,9]. zens in the Östhammar community, 14,932 of whom Thus, these disorders constitute a major public health were born between 1909 and 1974. Data from three prior problem. studies[2], which included random samples of the Öst- hammar population, were used to provide symptom-free Dyspepsia without peptic ulcer disease (PUD) (or any controls in the current study. We called this new study the other organic GI disease) is called functional dyspepsia Gastrointestinal Consult Study (GiCon). A sample was [10]. In a Swedish population-based upper endoscope drawn from the National Swedish Population Registry study, the prevalence of dyspepsia was 38% and the prev- 1995, now involving all Swedish citizens born between alence of PUD 4%, with about a quarter of the PUD sub- 1909 to 1974 and thus 20 to 87 years of age, born on day jects having no GI symptoms [48]. Thus, the vast majority 3, 12 or 24 of each month (n = 1537), i.e. using the same of those reporting dyspepsia in the population can be date of birth criteria as in previous studies on the same expected to have functional dyspepsia. IBS is a functional population in 1988 and 1989. Thus, we increased the gastrointestinal disorder [11]. Although functional dys- upper age limit to 87 years in order to include all prior pepsia and IBS are considered separate disorders [10,12], participants. As a consequence of the sampling strategy, many subjects report concomitant symptoms thought to all subjects younger than 27 years of age were included for originate from both the upper and the lower the GI tract the first time (n = 86). Most the participants were now [1,13]. Also, sufferers may experience a change in pre- being approached for the third time. Of the 1537 subjects dominant symptom profile over time between dyspepsia in the study base (see Figure 1), 105 were excluded due to and IBS, but much less towards GERS [2]. This means that having declined further participation in the 1988 study dyspepsia and IBS sufferers probably have common and 4 were excluded due to unidentified ID. Thus, 1428 aspects of both pathophysiology and health care seeking subjects were still eligible. They were sent a validated behavior. Furthermore GERS, in a majority of cases, has a postal questionnaire, called "The Abdominal Symptom proven (non-functional) acid-related etiology [11]. Questionnaire" (ASQ) [21], described below. Two Accordingly, it seems reasonable to exclude subjects with reminders were sent out when necessary. Drop-out rea- only GERS symptoms from the functional definition and sons are shown in Figure 1, together with the mean age to investigate the burden on society of functional dyspep- and sex distributions of the 911 responders who consti- sia and IBS taken together. We thus label these conditions tuted the population sample of the current GiCon study. as functional gastrointestinal disorders or FGID. Formation of the study groups and symptom classification A minority of subjects with FGID (5–20%) are reported to All 911 responders in the population sample were classi- visit a doctor quarterly for their complaints [14,15]. On fied according to their response in the ASQ (see Figure 1) the other hand, up to half of those with FGID never see a as having either Functional Gastrointestinal Disorder doctor despite prolonged symptoms [14,16]. For IBS, the (FGID, n = 244), i.e. functional dyspepsia or irritable proportion is even higher, up to 80 % [14]. Thus, GI bowel syndrome, or being Strictly Symptom Free (SSF, n patient-based data will not cover all aspects of morbidity. = 219). Moreover, subjects reporting minor symptoms Moreover, it has been reported previously that most that did not fulfil the criteria for dyspepsia or irritable patients' sick leave is due to disorders other than their bowel syndrome, or reporting isolated symptoms of gas- abdominal complaints. Only 23% of those subjects with troesophageal reflux symptoms (GERS, n = 271), were abdominal pain who were on sick leave stated that excluded. Subjects who were free from abdominal symp- abdominal pain was the cause of their absenteeism [17]. toms in this 1995 survey but had reported symptoms in the previous studies (1988–1989) were labelled as Prior There are prior cross-sectional population-based point Symptomatic (n = 177) and were also excluded. prevalence studies on GI co-morbidity [18-20]. However, there is a lack of knowledge of the co-morbidity of sub- The FGID and SSF sample groups (n = 244 + 219 = 463) jects with proven chronic/long lasting/persistent FGID were then invited by post to participate in the current and its impact on health care seeking behaviour due to GI investigation and to visit one of the six local health cen- symptoms. The aim of this study was to compare the co- tres. Subjects who did not respond to the first letter were morbidity and its relation to health care seeking behav- sent a reminder, and non-responders to the second letter iour among non-patients with persistent FGID with those were finally contacted by phone. In total, 187 (77%) with who are persistently GI symptom free, in a randomly- FGID and 156 (71%) with SSF accepted the invitation. selected adult Swedish population. There were 117 (57 FGID, 60 SSF) subjects with incom- plete (n = 12) or no (n = 105) response. At the health Page 2 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Sampling Figure 1 procedure of the Gastrointestinal Consult Study (GiCon study) Sampling procedure of the Gastrointestinal Consult Study (GiCon study). Sampling procedure of the Gastrointesti- nal Consult Study (GiCon study) with relation to the prior study of 1995. Page 3 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 GI symptoms, pain mo Figure 2 dalities and location GI symptoms, pain modalities and location. The symptoms inquired are shown in Part 1, and the pain and discomfort modalities with the master sketch for indicating their abdominal location in Part 2. The asterisks are explained in the text. Swedish laymen terms were used in the questionnaire. The master sketch shows the eligible pain locations. centres, the participants filled in an ASQ and a Complaint Questionnaires Score Questionnaire (CSQ), as described below, and the The abdominal symptom questionnaire height and weight of each was measured. The time period The Abdominal Symptom Questionnaire (ASQ) has been between the original 1995 ASQ and the second ASQ sur- validated previously and found to be reliable and repro- vey in 1996 ranged from 7 to 15 months. Of the FGID ducible [2,13,22]. In the questionnaire, subjects were group of 187 subjects, 8 subjects who now reported no asked if they had been troubled (Yes/No) by any of the 29 symptoms and 38 with minor symptoms were excluded; listed general gastrointestinal symptoms over the previous and out of the 156 SSF, 13 who now reported FGID and three months. They were also asked if they had been trou- 46 with minor symptoms were also excluded. Thus, data bled by any of 11 listed descriptors of abdominal pain and from 141 FGID subjects (mean age 45.7 y, 34% men) and where the symptom was located (upper, centre or lower 97 SSF subjects (mean age 52.4 y, 48% men) remained in abdominal, right and left flank, as shown in Figure 2). A the analysis. The sampling procedure for retrospective similar ASQ was used for the postal survey and the surgery data – twice over one year for those with, and up to four visit, with the latter extended to include a symptom sever- times over eight years for those without, GI symptoms – ity Likert scale graded from zero to seven for each symp- assured persistent symptomatic status. tom asked for. In the analysis, the data were pooled into a three-grade scale (1, 1–4, 5–7) Page 4 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Definitions of symptoms from the ASQ The complaint score questionnaire (See Figure 2) [23]. The Complaint Score Questionnaire contains 30 ques- tions, as shown in Figure 3, indicating the presence or Dyspepsia was defined as one or more of the symptoms absence of 30 different symptoms [25]. The questions are (reflux episodes, heartburn, retrosternal pain, nausea, dichotomous and can be categorized into six domains: vomiting, early satiety, uncomfortable feeling of fullness abdominal/urinary, ache/fatigue, muscular-skeletal, after meals, abdominal distension), and one or more of nutritional, cardio-vascular and depressive. the pain modalities (burning sensation, aching, pain, ten- derness, sinking feeling, "butterflies", cramp, twinge, Other questions stitch, colic, gripes) with any abdominal location except The participants were asked to state their coffee, alcohol the lower part, but no concomitant IBS. and tobacco consumption, and whether they had ever had peptic ulcer disease or had ever consulted a doctor for GI Irritable Bowel Syndrome (IBS) was defined as one or more complaints. Also, the past three months of general pain of the symptoms (abdominal distension, abdominal dis- and all GI medication were indicated. Educational back- comfort or pain on defecation, abdominal discomfort or ground was registered at five levels (elementary, compre- pain relieved by defecation, feeling of incomplete defeca- hensive, secondary, upper secondary, university) and tion, mucous stools) and one or more of the symptoms medical knowledge was evaluated by means of a self- (diarrhea, constipation, alternating diarrhea and consti- explanatory questionnaire (see Additional File). The pation) together with one or more of the pain modalities answers were scored with a sum of 0–15. (burning sensation, aching, pain, tenderness, sinking feel- ing, "butterflies", cramp, twinge, stitch, colic, gripes), with Statistical power and analysis any location. One hundred and twenty three subjects were required (in each the FGID and SSF groups) in order to have a 90% When using these definitions of dyspepsia and IBS, power at the P < 0.05 level to detect a 20% absolute differ- responders with only reflux symptoms, i.e. heartburn ence in exposure variables. This assumed a 24% preva- and/or retrosternal pain but no abdominal pain or dis- lence of FGID in the population, equal numbers of comfort, were classified as having GERS and not dyspep- subjects in the FGID and SSF groups, 15 and 20% absolute sia, while those with such symptoms and abdominal pain difference in the exposure variables within the two steps or discomfort fell into the dyspepsia group. Also, IBS, as of the study, 75% response rate on the ASQ, and 25% defined below, was given priority over dyspepsia. Thus, exclusion from each group in the last step. The variables concomitant occurrence of both led to a diagnosis of IBS. analysed are presented in the Additional File. Univariate This definition is in accordance with published guidelines analyses were performed using Student's t-test, Wilcoxon's [24]. ranksum test (Mann Whitney) and Pearson's chi-2 test. Multivariate analysis with FGID as the dependent variable Functional Gastrointestinal Disorder, FGID was performed using a logistic regression, and with dys- FGID was defined as either dyspepsia or IBS. peptic severity as the dependent variable with the ordinal logistic regression technique. Age and BMI were linear to Minor symptoms refer to symptoms not fulfilling the crite- outcome and were thus handled as continuous variables. ria of dyspepsia or IBS. By definition, those with GERS To test the symptoms in the CSQ, we made an age and sex were included in this group adjusted logistic regression model for each symptom and accepted a P value less than 0.05/(number of symptoms) "Strictly Symptom Free" (SSF) is defined as having no = 0.05/30 ≈ 0.0016 as significant. The explanatory varia- reported symptoms at all in the Abdominal Symptom bles were evaluated by a logistic regression in a sex and age Questionnaire in the 1995 survey, and subjects stating adjusted model. that they had not even been troubled previously with abdominal symptoms. Those subjects who had partici- The association between each potential determinant pated in the two former surveys in 1988 and 1989 (n = obtained from the questionnaires and the presence of 265) should also have reported no symptoms in each of FGID was quantified by using odds ratios and 95% confi- those two investigations. dence intervals. All exposure variables with P < 0.25 were then entered together in a multivariate logistic regression "Prior symptomatics" were those with no symptoms in the model to evaluate which was independently associated 1995 survey but with GI symptoms reported in the prior with the presence of FGID. A factor analysis was per- studies of 1988 and 1989. formed using all 30 complaints and factors with eigenval- ues greater than 1.3. Four factors were obtained by a principal components analysis with varimax rotation. The Page 5 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Figure 3 Proportion of subjects with FGID and with SSF reporting complaints during the previous three months in the CSQ Proportion of subjects with FGID and with SSF reporting complaints during the previous three months in the CSQ. Proportions (0–1.0) of subjects with FGID and with SSF (n = 238) reporting complaints during the previous three months in the Complaint Score Questionnaire (CSQ). There were significance differences *(p < 0.0016) on an age and sex adjusted logistic regression for all variables except coughing (ns) and excessive weight (ns). 30 variables were dichotomized with the highest 1/3 Results given the value 1 and the rest the value 0. A logistic regres- Representativeness of study sample sion was performed with the four factors age and sex In order to investigate the effect of the drop-outs during adjusted, and factors with p > 0.05 were excluded. To test the sampling procedure, the mean ages, sex ratios and whether the model fitted the data, a Pearson goodness-of- education levels in the corresponding FGID and SSF fit test with p values greater than 0.05 was performed. groups were analysed as shown in Table 1, for the samples When the number of covariates approximated the number from 1988, 1995 and 1996 (see Table 1) [13]. There were of observations, the Hosmer-Lemeshow test was per- no significant differences in any of these aspects (see Table formed to determine whether the model fitted the data. 1). For the ordinal logistic model, comparison with a multi- nomial model made an approximate fit test. No interac- Study group characteristics tions were found between the variables in the main As shown in Table 2, there were more females among those model. Ninety five per cent confidence intervals (CI) were with FGID. However, there were no intergroup differences in computed using parametric methods. A p-value of 0.05 or education, medical knowledge, BMI, intake of coffee, alcohol less was generally regarded as statistically significant. All and smoking. The age difference was irrelevant, as those with tests were two-tailed and the statistical package Stata 8 was SSF had been largely excluded due to prior study results. Dis- used for analysis [27]. ease-related variables were significantly different between the study groups. These variables were introduced to further model- Ethics ling, as shown below. The variables 'intake of GI medicine' and This study was approved by the Ethics Committee of the 'previous PUD' were not included due to sparse data. Medical Faculty, Uppsala University, June 5th 1996. Page 6 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Table 1: Comparison between previous population studies in Östhammar, Sweden. Age, sex and education level at different stages of the sampling process. From the first population sample in 1988 to the present study 1996. ns = p > 0.05. Group (G) Sample Year n Age years mean (SD) Sex % males Education median level (range) 1 First population sample 1988 1156 48.9 (16.0) 50.4 3 (1–4)* 2 Eligible sample 1995 1428 49.9 (17.2) 49.9 3 (1–4)** 3 Population sample 1995 911 49.2 (16.46) 47.0 3 (1–4) 4 Sample group FGID 1995 244 45.5 (15.3) 36.1 3 (2–4) 5 Sample Group SSF 1995 219 51.7 (17.6) 51.6 3 (1–4) 6 Study Group FGID 1996 141 45.7 (14.3) 34.0 3 (2–4) 7 Study Group SSF 1996 97 52.4 (15.3) 48.0 3 (1–4) G2 vs G3 ns ns G2 vs G3 ns ns ns G4 vs G6 ns ns ns G5 vs G7 ns ns ns *responders = 1156 **responders n = 1384 Table 2: Comparison between explanatory variables for subjects with FGID and SSF. Comparison between explanatory variables for subjects with FGID and SSF adjusted for sex and age. Ordinal variables are presented as median (range), dichotomous variables as proportion %, and continuous variables as mean (SD). Variable FGID SSF P n = 141 n = 97 AGE 45.7 (14) 52.4 (15) not relevant SEX (female %) 66 53 0.026 *** BMI 26.3 (4.7) 26.2 (4.2) Ns * Education 3 (2–4) 3 (1–4) Ns ** Medical knowledge 10 (4–15) 11 (5–15) Ns ** GI sympt severity 4 (3–5) 0 (0–0) <0.0005 ** GI Consultation (ever) 72% 8% <0.0005 *** Pain medicine (3 month) 77% 34% <0.0005 *** GI medicine (3 month) 32% 1% <0.0005 *** Previous PUD (ever) 12% 1% 0.006 *** Coffee 2 (2–3) 2 (2–3) Ns ** Alcohol 4 (3–4) 3 (2–4) Ns ** Smoking 1 (1–2) 1 (1–2) Ns ** * = Student's t-test ** = Mann-Whitney test *** = Pearson Chi-2 test CSQ and FGID vs. SSF cant for all except four complaints. After adjusting for The results from the 30 CSQ complaints for the FGID and alcohol and pain and GI drug intake (Table 3), 20 SSF study groups are presented in Figure 3. Those with complaints remained significant. A factor analysis was persistent FGID scored statistically higher on all variables performed including the 26 non-GI complaints. After a except difficulties in passing urine, excessive weight, varimax rotation of the four factors with eigenvalues > 1.3, coughing and impaired hearing. we found four factors representing psychological illness, somatic illness, ache/fatigue and one "miscellaneous" Risk modelling (Table 3). Each factor was then introduced in a logistic Risks of reported CSQ complaints for FGID vs. SSF, regression model adjusted for sex and age (Table 3), and expressed as age and sex adjusted OR, are presented in the "miscellaneous" factor was shown to be non-signifi- Table 3. GI complaints (abdominal pain, nausea, diarrhea cant. In the last sequential analysis, the three factors that and constipation) were excluded as we aimed to analyze remained significant in the prior analysis were introduced the co-morbidity with GI symptoms. The OR was signifi- Page 7 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Table 3: Odds ratios of FGID/SSF for complaints in the Complaints Score Questionnaire (CSQ). Odds ratios (OR, with 95% confidence intervals (CI)) of FGID/SSF (n = 238) for complaints elicited by the CSQ. Logistic regression is presented in different models. A factor analysis extracted four factors: A = psychological illness factor, B = somatic illness factor, C = miscellaneous factor, D = ache/fatigue factor. These were used in the modelling in the right two columns. I II III IV V Symptom OR (CI) by Models OR (CI) by Models FACTOR OR (CI) by Models adjusted OR (CI) by a main effect model adjusted for sex adjusted for sex, for sex and age adjusted for sex and age and age age, alcohol, pain tablets, GI-tablets SSF (all variables) 1 (Reference) 1 (Reference) FGID 1) Psychological illness 1) Psychological illness Cries easily 6.7 (2.3–19.9) 9.8 (2.0–47) A 8.0 (4.1–15.8) 8.4 (4.0–17.5) Sleeping disturbance 6.2 (2.7–14.0) 3.2 (1.3–8) A General fatigue 14.5 (7.4–28.7) 12.6 (5.3–30) A, D Irritability 8.8 (4.1–17.8) 5.6 (2.3 – 13.7) A, D Nervousness 18.4 (4.2–80.3) 14.3 (2.8 – 72) A Impaired concentration 19.0 (5.7–63.8) 15.3 (4.0 – 58) A Difficulty to relax 15.7 (6.0–41.5) 10.9 (3.7–32) A Restlessness 40.0 (9.4–170) 32.2 (6.7–154) A Depression 8.6 (4.1–18) 4.7 (2.0 – 11) A Exhaustion 12.7 (4.4–37) 9.1 (2.7–30) A 1) Somaticillness 1) Somaticillness Chest pain 40.0 (5.3–300) B 3.7 (2.0–7.1) 2.8 (1.3–5.7) Pain in the joints 6.2 (2.8–13.6) 7.5 (2.6–21) B Pain in the legs 4.4 (2.2–8.9) 3.8 (1.6–9.3) B Overweight 2.0 (1.0–4.0) 1.4 (0.6–3.6) B Breathlessness 8.9 (3.2–25) 12.1 (3.3–45) B Dizziness 10.1 (4.2–24) 11.4 (3.8–34) B Impaired hearing 3.0 (1.3–6.8) 3.1 (1.0–9.5) B Eye problem 4.2 (1.9–9.1) 3.4 (1.3–9.0) B 1) Miscellaneous Loss of weight - - C 1.7 (0.9–3.0) Bad appetite - - C Feeling cold 7.3 (3.1–17) 7.0 (2.6–19) C Difficulty in passing urine 9.6 (2.0–47) 9.1 (1.4–59) C 1) Ache/fatigue 1) Ache/fatigue Back ache 4.4 (2.4–8.2) 2.0 (0.9–4.3) D 2.9 (1.6–5.2) 4.3 (2.1–8.7) Headache 6.3 (3.4–12) 4.1 (1.9 – 9.1) D Sweating 3.6 (1.7 – 7.4) 3.3 (1.3 – 8.5) Coughing 2.0 (0.98–4.2) 1.7 (0.6–4.4) 1) Reference group (OR = 1) is those coded 0 in each factor together into a main effect model, adjusted for age and ASQ symptom severity and consulting behaviour vs CSQ factors sex. The OR for these factors remained significant, as shown in Table 3. From the ASQ, the mean grades of GI symptom severity for affirmative symptoms were analyzed against the three Consulters versus non consulters final CSQ factors from Table 3 (psychological illness, Consulters and non-consulters among those with persist- somatic illness, ache/fatigue) and for age, sex and consult- ent FGID were compared regarding their complaints, as ing behaviour, as shown in Table 4. The analysis showed reported in the CSQ. The proportion (0–1.0) per an obviously higher risk of increased GI symptom severity complaint is shown in Figure 4. There were no statistically for consulters (OR 12.3) and for psychological illness (OR significant differences between the consulters and non- 4.5), while somatic illness and ache/fatigue had a low risk, consulters for any of the complaints (adjusted for age and with the 95% CI close to 1.0. From the ASQ, consulting sex, p > 0.0016). behaviour was analyzed for the CSQ factors psychological illness, abdominal illness, age and sex (see Table 5). The analysis showed a greater chance of being a consulter for abdominal illness (OR 2.0) and psychological illness (OR 2.2). Page 8 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Consulters Figure 4 Proportion of complaints from the Complaint Score Questionnaire among those with FGID, divided into Consulters and Non- Proportion of complaints from the Complaint Score Questionnaire among those with FGID, divided into Con- sulters and Non-Consulters. Proportion (0–1.0) of complaints from the Complaint Score Questionnaire (CSQ) among those with FGID, divided into Consulters and Non-Consulters (n = 141). None of the variables showed a significant difference between Consulters and Non-Consulters for P values less than 0.0016, tested by a sex and age adjusted logistic regression. Table 4: Odds ratios of graded GI symptom severities in the Discussion ASQ for consulting, psychological illness, somatic illness and This study shows that among subjects with longstanding ache/fatigue factors. Odds ratios (OR, with 95% confidence FGID, there is a remarkably high prevalence of psycholog- intervals (CI)) of graded (0,1,2) GI symptom severity in the ASQ ical illness and also of non-GI somatic complaints. These for consulting, psychological illness, somatic illness and ache/ fatigue factors, age and sex, for both FGID and SSF (n = 232). are present regardless of whether the subjects have con- Ordinal logistic regression. sulted their doctor about their GI problems, and are more severe in subjects with persistent FGID. Although FGID Variable OR (CI) was more common in women, the consultation rates in sufferers were similar for males and females and were not Psychological illness low 1 age-related. Only about a quarter of the sufferers had high 4.5(2.4–8.4) Somatic illness low 1 never consulted their doctor. high 2.0(1.1–3.8) Ache/fatigue low 1 We consider that our findings can be generalized to the high 2.1(1.1–4.2) whole population, as the study groups were sampled from Consulters no 1 a well-defined and thoroughly investigated population, yes 12.3(6.3–23.9) most of whom had participated in prior studies [28,29]. Age (continuous) 0.96 (0.94–0.99) The original study base was formed in 1988 from the Sex female 1 male 0.9 (0.5–1.7) Swedish National Population register, which guarantees complete coverage of all citizens. There were no differ- ences in age, gender or education level between the study Page 9 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 Table 5: Odds ratio of consulting for abdominal complaints in the Subjects with FGID were on average younger than con- ASQ, 1995, for psychological illness, abdominal illness, age and trols without FGID, which may be expected as the preva- sex, for both FGID and SSF. Odds ratio (OR, with 95% lence of dyspepsia and IBS is generally higher in younger confidence intervals (CI) of consulting for abdominal complaints age groups [32]. This study was not a case control study, in the ASQ for 1995 for psychological illness, abdominal illness, age and sex for both FGID and SSF (n = 232). Logistic regression. however, but a study of all subjects with FGID and SSF Somatic illness and ache/fatigue were excluded in the final model within the cohort. Any differences caused by this grouping because they showed no significance in the prior step. strategy were controlled for in the analysis by gender- and age-adjusted logistic regression. There was a particular Variable OR (CI) association between FGID and psychological illness, Psychological illness low 1 although "fibromyalgia-like" symptoms (ache and High 2.2 (1.2–4.0) fatigue) [49] and other somatic complaints were also Abdominal illness low 1 common. Previous outpatient studies have shown that High 2.0 (1.1–3.8) IBS is associated with psychiatric illnesses such as depres- Age (continuous) 1.0 (0.99–1.0) sion [34], dysthymia [35] and anxiety [36], and similar Sex female 1 findings are reported for dyspepsia [50,51]. Greater men- Male 0.7 (0.40–1.3) tal pathology has been reported for consulters, i.e. patients, than non-consulters with IBS [44,45,52], but this has not been demonstrated convincingly in dyspepsia subjects [53]. and the sample groups or between the population sam- Healthcare seeking behaviour is complex and until now it ples from 1988 and 1995. Also, the proportions of those has been studied largely in patient samples. Nyrén et al. reporting symptoms explicable in terms of an organic dis- found that patients with non-ulcer dyspepsia had an ease have been shown to be insignificant [13]. Thus, the excessive need for sick leave compared with ulcer patients, FGID subjects and the symptom-free subjects would seem but that the predominant reason for leave was related to to be representative. musculoskeletal rather than abdominal symptoms [41]. Kettell et al. [42] found that severity of abdominal pain The validity of the research tool is a potential source of and anxiety about the seriousness of their condition were bias. However, for symptom reporting, only well-vali- important factors in patients consulting for IBS [42]. dated questionnaires were used. Both the ASQ and the There seem to be no differences in the use of healthcare CSQ have been adequately validated [28-30]. The psycho- services or co-morbidity status between the year before logical illness factor identified in this study embraces and the year after diagnosis of IBS [43,46]. symptoms of both "neurotic" and "personality" dimen- sions, listed in Table 3, and the outcome seems plausible. Although this study was focused on total co-morbidity, in The questionnaire used to assess medical knowledge was terms of consultation, other factors associated with FGID simple and straightforward, with kappa values per ques- consultation were also considered. The total consultation tion 1 = 0.70, 2 = 0.89, 3 = 0.47, 4 = 0.78, 5 = 0.80, 6 = pattern for the subjects will be published elsewhere. In 0.70 when repeated within a week, considered acceptable essence, effective treatment of patients with FGID involves for all with some reservation in the 0.47 case [53]. not only addressing GI discomfort, but also considering mental and somatic complaints such as depression and The definitions of dyspepsia and IBS used in this study exhaustion. were those used in the original study from 1988 [31], when the Rome II criteria [11] were unavailable. We opted Conclusion to retain our original study definitions despite ongoing In the present study, psychological illness proved to be an changes. important co-morbidity factor among subjects with FGID, and the severities of the two were linked. We cannot con- Our definition of dyspepsia was more restricted in terms clude anything about the cause of the relationship. The of symptoms than the Rome II definition, but wider in presence of psychological illness was also associated with terms of abdominal location, as not only epigastric but a greater need for medical consultation. Fear of life-threat- also mid abdominal symptoms were included. The IBS ening illness has been reported to be an important reason definition used was in accordance with the Rome II crite- for consultation in FGID [1], and this worry and anxiety ria [22]. Consequently, we consider the overall prevalence needs to be taken into account when attempting to man- of FGID in this study and the concordance on an individ- age FGID successfully. Somatic co-morbidity was found to ual level to be applicable within today's definitions [11]. be a less important reason for consultation, although a high proportion of subjects with FGID (77%) in our study Page 10 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 14. Agreus L: Socio-economic factors, health care consumption were taking analgesics while only 32% used specific gas- and rating of abdominal symptom severity. A report from trointestinal medication. the abdominal symptom study. Fam Pract 1993, 10:152-63. 15. Haycox A, Einarson T, Eggleston A: The health economic impact of upper gastrointestinal symptoms in the general popula- Competing interests tion: results from the Domestic/International Gastroenter- The author(s) declare that they have no competing ology Surveillance Study (DIGEST). Scand J Gastroenterol Suppl 1999, 231:38-47. interests. 16. Westbrook JI, McIntosh J, Talley NJ: Factors associated with con- sulting medical or non-medical practitioners for dyspepsia: Authors' contributions an Australian population-based study. Aliment Pharmacol Ther 2000, 14:1581-8. TÅ participated in the study design and checking the cod- 17. Nyren O, Lindberg G, Lindstrom E, Marke LA, Seensalu R: Eco- ing of the data, carried out the construction of the "medi- nomic costs of functional dyspepsia. Pharmacoeconomics 1992, cal knowledge questionnaire" including a test-retest 1:312-24. 18. Stanghellini V: Relationship between upper gastrointestinal survey, distributed and collected the questionnaires, symptoms and lifestyle, psychosocial factors and comorbid- called the participants to his health centre, supported the ity in the general population: results from the Domestic/ International Gastroenterology Surveillance Study other health centers personnel, performed all the statisti- (DIGEST). Scand J Gastroenterol Suppl 1999, 231:29-37. cal analysis, participated in writing the manuscript, and 19. Haug TT, Mykletun A, Dahl AA: Are anxiety and depression submitted the final manuscript. LA carried out the study related to gastrointestinal symptoms in the general population? Scand J Gastroenterol 2002, 37:294-8. design and contributed to the statistical analysis and the 20. Koloski NA, Talley NJ, Boyce PM: Epidemiology and health care writing of the manuscript. KS contributed to checking the seeking in the functional GI disorders: a population-based study. Am J Gastroenterol 2002, 97:2290-9. coding of the data and participated in the study design. SJ 21. Agreus L, Svardsudd K, Nyren O, Tibblin G: Reproducibility and contributed help with the statistical analysis. validity of a postal questionnaire. The abdominal symptom study. Scand J Prim Health Care 1993, 11:252-62. 22. Agreus L, Talley NJ, Svardsudd K, Tibblin G, Jones MP: Identifying Acknowledgements dyspepsia and irritable bowel syndrome: the value of pain or This work was supported by the personnel of the Primary Health Care discomfort, and bowel habit descriptors [In Process units of Östhammar, the Uppsala County Council and Anna Hedin Ph D, Citation]. Scand J Gastroenterol 2000, 35:142-51. National Institute of Public Health, Sweden. We also thank Madeline Frame 23. Agréus L, Svärdsudd K, Tibblin G, Nyrén O: The epidemiology of dyspepsia: Symptom clusters and demographic for assistance with the manuscript preparation. characteristics. Gastroenterology 1991, 100:A1. 24. Talley NJ: Working Team Report. Functional Dyspepsia: A References Classification with Guidelines for Diagnosis and 1. Jones R, Lydeard SE, Hobbs FD, Kenkre JE, Williams EI, Jones SJ, Jones Management. Gastroenterology International 1991, 4:145-160. , Repper JA, Caldow JL, Dunwoodie WM, Bottomley JM: Dyspepsia 25. Tibblin G, Tibblin B, Peciva S, Kullman S, Svardsudd K: "The Gote- in England and Scotland. Gut 1990, 31:401-5. borg quality of life instrument" – an assessment of well-being 2. Agreus L, Svardsudd K, Talley NJ, Jones MP, Tibblin G: Natural his- and symptoms among men born 1913 and 1923. Methods tory of gastroesophageal reflux disease and functional and validity. Scand J Prim Health Care Suppl 1990, 1:33-8. abdominal disorders: a population-based study. Am J 26. World Health Organization: WHO/46 Press Release: World Health Gastroenterol 2001, 96:2905-14. Organization 1997. 3. Agreus L: Natural history of dyspepsia. Gut 2002, 50:iv2-9. 27. StataCorp: Stata Statistical Software: Release 8.0 College Station, TX: 4. Talley NJ, Zinsmeister AR, Schleck CD, Melton LJ III: Dyspepsia and Stata Corporation; 2003. dyspepsia subgroups: a population-based study. Gastroenterol- 28. Agréus L, Svärdsudd K, Nyrén O, Tibblin G: Reproducibility and ogy 1992, 102:1259-68. validity of a postal questionnaire. The abdominal symptom 5. Talley N, Boyce P, Jones M: Identification of upper and lower study. Scand J Prim Health Care 1993, 11:252-62. gastrointestinal symptom groupings in an urban population. 29. Agréus L: The Abdominal Symptom Study. An Epidemiologi- Gut 1998, 42:690-95. cal Survey of Gastrointestinal and Other Abdominal Symp- 6. Agréus L, Talley NJ: Dyspepsia: current understanding and toms in the Adult Population of Östhammar, Sweden. In PhD management. Annu Rev Med 1998, 49:475-93. thesis Uppsala University, Department of Family Medicine; 1993. 7. McDougall NI, Johnston BT, Kee F, Collins JS, McFarland RJ, Love AH: 30. Tibblin G, Cato K, Svärdsudd K: Goteborg quality of life study of Natural history of reflux oesophagitis: a 10 year follow up of men born in 1913 and 1923 – age, sex, job satisfaction and its effect on patient symptomatology and quality of life. Gut cardiovascular diseases. Scand J Prim Health Care 1990, 1996, 38:481-6. 1(Suppl):39-45. 8. Agreus L, Borgquist L: The cost of gastro-oesophageal reflux 31. Agréus L, Nyrén O, Svärdsudd K, Tibblin G: Ont i magen-En epi- disease, dyspepsia and peptic ulcer disease in Sweden. Phar- demiologisk studie om bukbesvär i Östhammars kommun. macoeconomics 2002, 20:347-55. Svenska Läkarsällskapets Hygea 1990, 99(1):90. 9. Leong SA, Barghout V, Birnbaum HG, Thibeault CE, Ben-Hamadi R, 32. Agréus L, Svärdsudd K, Talley NJ, Jones MP, Tibblin G: Natural his- Frech F, Ofman JJ: The economic consequences of irritable tory of gastroesophageal reflux disease and functional bowel syndrome: a US employer perspective. Arch Intern Med abdominal disorders: a population-based study. Am J 2003, 163:929-35. Gastroenterol 2001, 96:2905-14. 10. Drossman DA, Corazziari E, Thompson WG, Talley NJ, Whitehead 33. Gupta S, Masand PS, Kaplan D, Bhandary A, Hendricks S: The rela- W: The Functional Gastrointestinal Disorders. Second edition. tionship between schizophrenia and irritable bowel syn- Degnon Associates, McLean, Va, USA; 2000. drome (IBS). Schizophr Res 1997, 23:265-8. 11. Talley N, Stanghellini V, Heading R, Koch K, Malagelada J, Tytgat G: 34. Masand PS, Kaplan DS, Gupta S, Bhandary AN, Nasra GS, Kline MD, Functional Gastroduodenal Disorders. Gut 1999, 45(Suppl Margo KL: Major depression and irritable bowel syndrome: is 11):1137-42. there a relationship? J Clin Psychiatry 1995, 56:363-7. 12. Whitehead WE, Gibbs NA, Li Z, Drossman DA: Is functional dys- 35. Masand PS, Kaplan DS, Gupta S, Bhandary AN: Irritable bowel syn- pepsia just a subset of the irritable bowel syndrome? Baillieres drome and dysthymia. Is there a relationship? Psychosomatics Clin Gastroenterol 1998, 12:443-61. 1997, 38:63-9. 13. Agreus L, Svardsudd K, Nyren O, Tibblin G: Irritable bowel syn- drome and dyspepsia in the general population: overlap and lack of stability over time. Gastroenterology 1995, 109:671-80. Page 11 of 12 (page number not for citation purposes) BMC Medicine 2005, 3:8 http://www.biomedcentral.com/1741-7015/3/8 36. Lydiard RB, Fossey MD, Marsh W, Ballenger JC: Prevalence of psy- chiatric disorders in patients with irritable bowel syndrome. Psychosomatics 1993, 34:229-34. 37. Sullivan G, Blewett AE, Jenkins PL, Allison MC: Eating attitudes and the irritable bowel syndrome. Gen Hosp Psychiatry 1997, 19:62-4. 38. Walker EA, Gelfand AN, Gelfand MD, Green C, Katon WJ: Chronic pelvic pain and gynecological symptoms in women with irri- table bowel syndrome. J Psychosom Obstet Gynaecol 1996, 17:39-46. 39. Woodman CL, Breen K, Noyes R Jr, Moss C, Fagerholm R, Yagla SJ, Summers R: The relationship between irritable bowel syn- drome and psychiatric illness. A family study. Psychosomatics 1998, 39:45-54. 40. Sperber AD, Atzmon Y, Neumann L, Weisberg I, Shalit Y, M Abu- Shakrah, Fich A, Buskila D: Fibromyalgia in the irritable bowel syndrome: studies of prevalence and clinical implications. Am J Gastroenterol 1999, 94:3541-6. 41. Nyrén O, Adami HO, Gustavsson S, Lööf L: Excess sick-listing in nonulcer dyspepsia. J Clin Gastroenterol 1986, 8:339-45. 42. Kettell J, Jones R, Lydeard S: Reasons for consultation in irritable bowel syndrome: symptoms and patient characteristics. Br J Gen Pract 1992, 42:459-61. 43. Ruigomez A, Wallander MA, Johansson S, Garcia Rodriguez LA: One- year follow-up of newly diagnosed irritable bowel syndrome patients. Aliment Pharmacol Ther 1999, 13:1097-102. 44. Ballenger JC, Davidson JR, Lecrubier Y, Nutt DJ, Lydiard RB, Mayer EA: Consensus statement on depression, anxiety, and func- tional gastrointestinal disorders Irritable bowel syndrome, anxiety, and depression: what are the links? J Clin Psychiatry 2001, 62(Suppl 8):48-51. 45. Lydiard RB: Irritable bowel syndrome, anxiety, and depres- sion: what are the links? J Clin Psychiatry 2001, 62(Suppl 8):38-47. 46. Whitehead WE, Palsson O, Jones KR: Systematic review of the comorbidity of irritable bowel syndrome with other disor- ders: what are the causes and implications? Gastroenterology 2002, 122:1140-56. 47. Jarrett ME, Burr RL, Cain KC, Hertig V, Weisman P, Heitkemper MM: Anxiety and depression are related to autonomic nervous system function in women with irritable bowel syndrome. Dig Dis Sci 2003, 48:386-94. 48. Aro P, Ronkainen J, Storskrubb T, Bolling-Sternevald E, Talley NJ, Agréus L: Prevalence of symptoms and upper endoscopic find- ings in a random adult population. A report from Kalixanda study. Gastroenterology 2001, 120(Suppl 1):A-231. 49. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Gold- enberg DL, Tugwell P, Campbell SM, Abeles M, Clark P: The Amer- ican College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Cri- teria Committee. Arthritis Rheum 1990, 33(2):160-72. 50. Talley NJ, Fung LH, Gilligan IJ, McNeil D, Piper DW: Association of anxiety, neuroticism, and depression with dyspepsia of unknown cause. A case-control study. Gastroenterology 1986, 90(4):886-92. 51. Kane F Jr, Strohlein J, Harper RG: Nonulcer dyspepsia associated with psychiatric disorder. South Med J 1993, 86(6):641-6. 52. Drossman DA, McKee DC, Sandler RS, Mitchell CM, Cramer EM, Lowman BC, Burger AL: Psychosocial factors in the irritable bowel syndrome. A multivariate study of patients and non- patients with irritable bowel syndrome. Gastroenterology 1988, 95(3):701-8. 53. Sackett DL: A primer on the precision and accuracy of the clin- Publish with Bio Med Central and every ical examination. JAMA 1992, 267:2638-44. scientist can read your work free of charge "BioMed Central will be the most significant development for Pre-publication history disseminating the results of biomedical researc h in our lifetime." The pre-publication history for this paper can be accessed Sir Paul Nurse, Cancer Research UK here: Your research papers will be: available free of charge to the entire biomedical community http://www.biomedcentral.com/1741-7015/3/8/prepub peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 12 of 12 (page number not for citation purposes)

Journal

BMC MedicineSpringer Journals

Published: May 13, 2005

There are no references for this article.