Irritable bowel syndrome (IBS), functional constipation, and diarrhea are a group of functional gastrointestinal disorders that are common in Western countries. Depending on the different diagnostic criteria used, population selection, and data sources, IBS has been reported in between 17% and 22% of the population (1–3), functional constipation in between 3% and 17% (3,4), and functional diarrhea in between 2% and 18% (3,4). Functional gastrointestinal disorders do not appear to damage the intestinal tract permanently, nor do they lead to more serious diseases, such as cancer. The majority of people with functional gastrointestinal disorders can manage their symptoms by modifying their diet, stress management, and prescribed medications. The health-related quality of life is impaired in people with functional gastrointestinal disorders, however, which results in increased absenteeism from school or work, reduced job opportunities, and reduced social interaction.
Functional gastrointestinal disorders comprise a common group of medical problems for adolescents (5) as well as adults. Data from studies conducted in adults have been used to guide the treatment of children with functional gastrointestinal disorders (6). Adolescence is a unique time in human development in terms of both physiology and psychology. People in this age group differ from adults in many aspects, including their social groups and environmental situations (7). Adolescents in modern society face many health issues, particularly in the areas of mental, emotional, and social health. To date, most of the literature about functional gastrointestinal disorders comes from Western industrialized societies, but IBS appears to be common in adolescents in Asia. A history of dysentery and the abuse of analgesics, such as nonsteroidal anti-inflammatory drugs (NSAIDs), were highly correlated with the symptoms of IBS. The putative mechanism of IBS caused by dysentery is as follows: the external infection may increase the synthesis and release of adrenal cortex hormone releasing factor (CRF); the increased function of CRF neurons in nerve centers continues after the initial infection with dysentery in some patients, and promotes the continual increase in gastrointestinal movement and secretion (8). It is valuable to investigate normal bowel habits, prevalence rates, and risk factors of functional gastrointestinal disorders to understand disorders in bowel motility (9); however, there are few data in the adolescent age group.
To date, no surveys have been performed to investigate normal bowel habits, prevalence rates, and factors associated with functional gastrointestinal disorders in Chinese adolescents according to the Rome III criteria. The aims of the present study were to investigate the normal patterns of bowel habit and assess the prevalence rates and associated factors of functional gastrointestinal disorders based on the Rome III criteria in Chinese adolescents.
PATIENTS AND METHODS
Shanghai is a modern metropolis in southern China. There are 18 senior high schools and 20 junior high schools in the Jiading district, and 16 senior high schools and 19 junior high schools in the Putuo district. For the present study, we selected 6 senior high schools and 6 junior middle schools at random from within the Jiading and Putuo districts of Shanghai between September and October 2008. Eight classes from each of the selected senior high schools (grades 10–12) were selected at random, and 9 classes were selected from each of the selected junior middle schools (grades 7–9). The mean number of class-bound clusters of children was 40 (SD 5.82, range 34–45). Every student in each of the randomly selected classes was included in the study, giving a recruitment rate of 100%. No student was excluded from the study after the medical assessment. Overall, 3671 students were recruited into the study. The study was approved by the medical ethics committee of Xinhua Hospital and the Jiading and Putuo medical bureaus.
The study used clustered sampling with a stratified, randomized approach. Using the general formula n = t2pq/d2 and data from a preliminary investigation, where t = 1.96 (error of the first kind), p = 13.25% (estimated prevalence), q = 1 – p, and d = 15% × p (permissive error), we determined that the optimal number of students for the present study was 1118. To ensure reliability, the calculated sample size using the above formula needed to be multiplied by a design effect of 2.5. Therefore, we increased the minimum sample size requirement to 3000 students and proceeded to recruit 3671 students into the study.
Before implementing the study, the purpose of the research project was explained to the principals and teachers from each of the target schools, and the students who were eligible to participate in the study were provided a letter for their parents. In this letter, we explained the objectives of the project and parents were also told that participation in the survey was voluntary; an informed consent form was then signed. Members of the Jiading and Putuo district education bureaus joined with doctors and graduates of the gastrointestinal department and the pediatric department who had been schooled in the relevant techniques to train the teachers in charge of the selected classes to administer the questionnaire. Students then completed the self-reported questionnaire under the direction of the trained teachers during classes of cultural subjects for health. Data in all of the questionnaires that were returned were checked carefully, and the test-retest reliability was established to be 97.8%.
We developed the 45-item questionnaire for IBS in adolescents based on previous international studies and Rome III criteria (5,10). To make the questionnaire specific for China, we integrated certain cultural characteristics and special situations, which incorporated differences in basic social demographic characteristics. Based on the structured (nominal and focus group) methods, the theory, and methodology, the questionnaire was designed and 247 middle school students were studied to evaluate the questionnaire. The questionnaire for IBS in adolescents was designed with 5 categories of questions: food habits (eg, How often do you eat cellulose, protein, spicy, fried foods, and drinks containing caffeine? [marked as: Usually, often, or occasional/no]); diagnostic criteria (Rome III criteria, stool consistency was questioned and recorded according to a Bristol type); evoking agents (including gastrointestinal infections and antibiotic and analgesic use); gastrointestinal symptoms (including dyspepsia, the feeling of foreign matter in the pharynx, heartburn, hiccups, and decreased appetite); and psychosomatic symptoms (including insomnia, fatigue, emotions, and means of entertainment, such as playing football, table tennis, swimming, singing, and dancing). Cronbach alpha estimations of reliability on the 5 categories were 0.668, 0.752, 0.824, 0.749, and 0.824, respectively, and the cumulative variance of the principal components was 70.72%. The intraclass correlation coefficients for the 3 main dependent variables (IBS, functional constipation, and functional diarrhea) ranged from 0.84 to 0.91.
The 23 possible factors associated with functional gastrointestinal disorders were conceptually grouped into 5 domains: demographic variables; food habits; evoking agents; gastrointestinal symptoms; and psychosomatic symptoms. Table 1 demonstrates the recording of each of the 23 possible factors associated with functional gastrointestinal disorders.
Definition of IBS
IBS was diagnosed and classified into 4 subgroups according to the Rome III criteria (10). IBS was defined as recurrent abdominal pain or discomfort for at least 3 days per month in the last 3 months associated with 2 or more of the following: improvement in abdominal pain or discomfort after defecation, onset associated with a change in frequency of stool, and onset associated with a change in form (appearance) of stool. The criteria were fulfilled for the last 3 months with the onset of symptoms at least 6 months before diagnosis. Students with medical records that documented major abdominal surgery, peptic ulcer, ulcerative colitis, diabetes mellitus, or hyperthyroidism were not classified with IBS, and were excluded from the study. Subjects were classified into 4 subgroups: if the stools were lumpy or hard in at least 25% of defecations, or there were no loose or watery stools occurring in at least 75% of stools, the subject was identified to be having the constipation-predominant IBS subgroup; the diarrhea-predominant IBS subgroup was identified if stool was loose or watery in at least 25% of defecations, or there were no lumpy or hard stools in at least 75% of stools; the alternating diarrhea and constipation-predominant IBS subgroup was identified if stool was lumpy or hard as well as loose or watery in at least 25% of defecations; and the undetermined-predominant IBS subgroup was identified if subjects did not fulfill any of the above criteria for the constipation, diarrhea, and alternating diarrhea and constipation-predominant IBS subgroups.
Definition of Functional Constipation
The diagnostic criteria for functional constipation must include 2 or more of the following: straining during 25% or more of defecations; lumpy or hard stools in at least 25% of defecations; the sensation of incomplete evacuation for at least 25% of defecations; the sensation of anorectal obstruction for 25% or more of defecations; manual maneuvers to facilitate defecation for 25% of bowel movements or more; and/or fewer than 3 defecations per week. In this condition, loose stools are rarely present without the use of laxatives. There are insufficient criteria for IBS. Criteria are fulfilled for diagnosis if present for the last 3 months with the onset of symptoms at least 6 months before the diagnosis (10).
Definition of Functional Diarrhea
The diagnostic criteria for functional diarrhea were loose or watery stools without pain occurring in at least 75% of stools. This criterion was fulfilled for the last 3 months with symptom onset at least 6 months before diagnosis (10).
Descriptive statistics (frequencies, proportions, means, and standard deviations [SD]) were calculated for all of the items of the questionnaire, and the 95% confidence intervals (CI) for the proportions were estimated by standard methods. Bivariate comparisons between study groups were made using statistical tests that were appropriate for the level of measurement of the dependent variables, including χ2 and t tests. The logistic regression analysis was conducted to assess the association with factors that contributed to IBS, functional constipation, and functional diarrhea. Unadjusted OR and 95% CI for IBS, functional constipation, and functional diarrhea were calculated using univariate logistic regression. Multivariate logistic regressions adjusted for age and sex were used to test the factors that contributed to IBS, functional constipation, and functional diarrhea. Statistical tests of regression estimates or OR were based on Wald statistics. The alpha level of significance was set at 0.05 and all P values that were calculated were 2-tailed. The data were processed, and a statistical analysis was performed with SAS 9.1 programs (SAS Institute, Cary, NC).
Demographic Characteristics of the Cohort
The age of the 3671 students included in the present study ranged from 12 to 18 years. The senior high school students (grades 10–12) included 419 boys and 554 girls (n = 973) with a mean age of 16.45 ± 1.27 years; the junior middle school students (grades 7–9) comprised 1280 boys and 1418 girls (n = 2698), with a mean age of 13.65 ± 1.41 years. The ratio of boys to girls across all of the grades in the present study was 1:1.16.
Frequency of Defecation
Of the entire cohort, 3232 students (88.05%) reported a stool frequency of between 1 or 2 times per day and once every 2 days. Furthermore, 2278 students (62.06%) reported a stool frequency of between 1 and 2 times per day. Nine hundred fifty-four students (25.99%) reported having only 1 bowel movement every 2 days. The majority of girls had significantly fewer stool frequencies than boys (Table 2, Fig. 1).
Seven hundred thirty students (334 boys, 396 girls) had symptoms of IBS as defined by the Rome III criteria. The prevalence of IBS in the total cohort was 19.89% (95% CI 18.60–21.18); the incidence of IBS was slightly higher among girls (20.08%) than boys (19.66%), but this difference was not significant (P > 0.05).
Students with IBS were subdivided into 4 categories of IBS. Among the students who fulfilled the Rome III criteria, 147 (20.14%) were constipation predominant, 135 (17.76%) were diarrhea predominant, 75 (10.27%) had alternating diarrhea and constipation, and 373 (51.10%) were placed in the undetermined subgroup of IBS.
Factors That Contributed to IBS
The IBS (Rome III, n = 730) and non-IBS (n = 2941) groups were compared to determine the probable factors that contribute to IBS. We found that gastrointestinal tract infection (OR 2.26), analgesic abuse (OR 1.25), air swallowing to terminate hiccups (OR 1.28), fatigue (OR 1.15), and depression (OR 1.36) were factors that were significantly associated with IBS after an appropriate adjustment for age and sex (P < 0.05; Table 3).
Nine hundred fifteen students (362 boys and 553 girls) had symptoms of functional constipation as defined by the Rome III criteria. The prevalence of functional constipation was 24.93% (95% CI 23.53–26.33), and the incidence of functional constipation was significantly higher among girls (28.04%) than boys (21.31%; P < 0.05).
Factors That Contribute to Functional Constipation
Nine hundred fifteen students were defined as having functional constipation according to the Rome III criteria, and 2756 students were defined as having nonfunctional constipation. These 2 groups were compared to determine the probable factors that contributed to functional constipation. We found that fried food (eg, the regular consumption of fried fish, chicken, and meat; OR 1.68), air swallowing to terminate hiccups (OR 1.21), anxiety (OR 1.12), and depression (OR 1.57) were factors that were significantly associated with functional constipation in our cohort after adjustment for age and sex (P < 0.05; Table 4).
One hundred ninety-nine students (98 boys and 101 girls) had symptoms of functional diarrhea as defined by the Rome III criteria. The prevalence of functional diarrhea was 5.42% (95% CI 4.69–6.15), and the incidence of functional diarrhea was slightly higher among boys (5.77%) than girls (5.12%), although this difference was not significant (P > 0.05).
Factors That Contributed to Functional Diarrhea
One hundred ninety-nine students were defined as having functional diarrhea according to the Rome III criteria and 3472 students were defined as having nonfunctional diarrhea. These 2 groups were compared to determine the probable factors that contributed to functional diarrhea. We did not find that any factors were significantly associated with functional diarrhea.
Of all of the students that fulfilled the criteria for functional gastrointestinal disorders, 16.85% of the students with IBS, 9.39% of those with functional constipation, and 11.02% of students with functional diarrhea had visited a physician for abdominal pain and altered bowel habits.
Functional gastrointestinal disorders are recognized to be complexes of symptoms that are characterized by abdominal pain and disturbed bowel action. The prevalence rates of functional gastrointestinal disorders have important implications for the allocation of resources and planning of medical services. The prevalence of functional gastrointestinal disorders varies between countries and within different populations in the same country. Adolescents in China devote a lot of time to study and many of them attend private cramming schools on weekends to do well in the entrance examinations of prestigious senior high schools and universities in Shanghai. To address the differences between adults and adolescents, we incorporated national conditions, cultural situations, and social structures in China in a questionnaire regarding functional gastrointestinal disorders that would be specific for Chinese adolescents. The data gathered in the present epidemiological study are, therefore, more objective and relevant for adolescents in this cultural context.
Using our modified questionnaire, we gathered data that showed that the prevalence rates of IBS and functional constipation among students were 19.89% and 24.83% respectively, which confirmed that IBS and functional constipation were common in Chinese adolescents. Our results were consistent with previous epidemiological surveys in Asia (25.7%) (11) and the West (8%–44%) (4,12), which had indicated the prevalence of IBS to be higher in children and adolescents than in adults. The differences in the prevalence of functional gastrointestinal disorders between adolescents and adults may be associated with different populations and environments used in these studies. Girls have a significantly lower bowel frequency than boys, which is similar to the results of previous epidemiological surveys in adults (13). At present, it is not possible to conclude whether this difference in sex ratios was a biological phenomenon or caused by socio cultural reasons. Possible reasons may be related to the menstrual cycle and differential hormonal ratios between sexes, or partly because of the fact that girls may be more willing to report their symptoms to doctors (14). Also, boys may have more accurate memories of painless symptoms (15). The difference in sex ratios with regard to morbidity in the present study provided further ideas for future studies.
The etiology and pathogenesis of functional gastrointestinal disorders are uncertain. An evaluation of the factors that contributed to IBS provided insight into the character of these disorders. In the present study, we found that students with IBS had not only lower gastrointestinal symptoms of abdominal pain and disturbed bowel action but also upper gastrointestinal symptoms that included hiccups that were terminated by air swallowing. A history of dysentery and the abuse of analgesics, such as NSAIDs, were greatly correlated with symptoms of IBS.
Certain external pathogens, endotoxins, and other cell factors damage the mucous membrane of the intestine directly and increase its permeability. The immune cells involved in inflammation may be activated by foods, bacterial antigens, or neuropeptide receptors and act on intestinal function through cell and inflammation factors (15). Many patients with IBS took analgesics to reduce abdominal pain or other aches. Whether this was the reason or the result of IBS merits further study. Therefore, to lower the incidence of IBS in adolescents, it may be important to control dysentery and ban the use of analgesics without a prescription.
We also explored possible contributing factors of functional constipation and found that an excessive intake of fried food was significantly associated with this disorder. High lipids in fried food could induce intestine strong contraction and spasm. In addition, industrialization and urbanization are changing the Chinese lifestyle in many ways. As a result, Chinese students face a fast-paced and greatly competitive school environment, which places them under a high level of psychological stress. Our study also indicated that minimizing psychological distress could improve the symptoms of IBS and functional constipation. Unfortunately, we found no factors that were significantly related to the etiology of functional diarrhea to form the basis of future studies.
The prevalence of functional diarrhea was the lowest for functional bowel disorders in adolescents and there was no significant difference in sex ratios. This could be associated with improved living standards, refined food, and the improved knowledge of hygiene in high schools.
The rates of health seeking in IBS, functional constipation, and functional diarrhea were 16.85%, 9.39%, and 11.02% respectively, which were lower than the results from previous epidemiological surveys carried out in adults in China (16). This suggested that the behavior of health seeking in adolescents had its own characteristics, and it may be necessary to design medical strategies in accordance with these manifestations.
In China, the epidemiological studies with regard to functional gastrointestinal disorders have been performed mainly with cohorts of adults. Previous studies on bowel function in adolescents were sparse and focused mainly on specific bowel functions rather than on establishing the characteristics of normal bowel habits in this age group (5,12). A better understanding of the range of bowel patterns is important to enable disorders of bowel motility to be understood. Using a self-reported questionnaire, we undertook the present cross-sectional survey in adolescents in Shanghai, and found that 87.32% of students reported having a stool frequency of between 1 and 2 times per day and once every 2 days. The majority of girls had significantly fewer stool frequencies than boys. These figures were similar to those recorded in previous surveys that were conducted in the West. Surveys from Western countries found that the normal stool frequency fell within the limits of between 3 times per week and 3 times per day in most apparently healthy subjects, and that women had fewer bowel movements than men (13,17). The similarity of these cross-cultural surveys supported the view that the definitions of functional constipation and functional diarrhea as stool frequency lying outside this normal stool frequency range were valid, and that a different stool frequency between sexes may be associated with different levels of hormones, which deserved further study.
In conclusion, we carried out an epidemiological survey of functional gastrointestinal disorders in Chinese adolescents in line with the Rome III criteria, which enabled the normal bowel frequency in adolescents to be defined. This provides objective evidence for further basic and clinical research. There were some limitations to the present survey. Because our results are limited to Shanghai, which is a greatly developed city in southern China, inferences to the rural areas of southern China and other regions could not be made. Consumption of high protein, high fiber, and fried food was not quantified. In addition, gastrointestinal tract infections (virus? or bacterium?) and analgesic use (NSAIDs? or narcotics?) were not defined or quantified. Psychological distress such as depression and anxiety were measured with students’ self-reports, and not with a validated instrument. Therefore, these most probable risk factors deserved to be investigated further.