Secondary Logo

Journal Logo

Original Articles—Gastroenterology

Recurrent Abdominal Pain: Symptom Subtypes Based on the Rome II Criteria for Pediatric Functional Gastrointestinal Disorders

Walker, Lynn S.*; Lipani, Tricia A.*; Greene, John W.*; Caines, Karen*; Stutts, John; Polk, D. Brent*; Caplan, Arlene; Rasquin-Weber, Andree

Author Information
Journal of Pediatric Gastroenterology and Nutrition: February 2004 - Volume 38 - Issue 2 - p 187-191



Click on the links below to access all the ArticlePlus for this article.

Please note that ArticlePlus files may launch a viewer application outside of your web browser.

Research on pediatric functional gastrointestinal disorders (FGIDs) has focused primarily on recurrent abdominal pain, a common complaint estimated to affect 10% to 15% of school-age children (1,2). In his pioneering studies, Apley (1) defined recurrent abdominal pain (RAP) as three or more episodes of abdominal pain severe enough to interfere with a child's activities and occurring during a period longer than 3 months. Subsequent clinical literature has proposed the existence of several variants of RAP, including nonulcer dyspepsia, irritable bowel syndrome, and functional abdominal pain (3–5). Most empirical investigations continue to treat children who meet Apley's criteria as a single homogeneous group, thereby disguising potential subgroups within a category so broad that it has limited utility for investigating possible differences in pathophysiology and discovering whether patients with different symptom patterns may vary in their course of illness and treatment response (6). This global approach to RAP has been dictated by the absence of biologic markers and the lack of diagnostic symptom criteria to characterize symptom subgroups for clinical research.

Symptom-based diagnostic criteria (the “Rome criteria”) recently have been established for the classification of FGIDs in adults (7) and children (8). This classification system defines FGIDs in terms of symptom constellations in the absence of structural or biochemical abnormalities. By standardizing the diagnostic criteria used in selecting patients for research participation, the Rome criteria may facilitate focused research on these disorders. This study examined the utility of the Rome criteria in identifying distinct symptom profiles among patients meeting Apley's criteria for RAP. We tested the hypothesis that most of these patients have symptom profiles consistent with one or more of the FGIDs defined by the pediatric Rome criteria.



Subjects were parents of new patients referred to the pediatric gastroenterology clinic at Vanderbilt University Medical Center and who met study eligibility criteria: the patients were between the ages of 4 and 17 years and by parent report had experienced abdominal pain at least as frequently as specified in Apley's definition of RAP (i.e., three or more episodes in the previous 3 months). All patients had previously been evaluated at least once by their primary care provider and were referred for subspecialty evaluation. Patients with known organic disease, chronic illness, or a handicapping condition were excluded. Data were obtained for 114 patients with recurrent abdominal pain.


Questionnaire on Pediatric Gastrointestinal Symptoms

The Questionnaire on Pediatric Gastrointestinal Symptoms (QPGS) (9) assesses symptoms associated with FGIDs in children, as specified by the pediatric Rome criteria (8). The QPGS Form A, the parent-report form for parents of children between 4 and 17 years of age, was administered in this study and is available from the first author. This structured questionnaire includes sections assessing children's bowel habits, abdominal pain, and other gastrointestinal symptoms, as well as limitations in activities. It can be completed in approximately 30 minutes. (This questionnaire can be found on the JPGN website as an appendix to this article, click on “Articles Plus” in the Table of Contents.)


After informed consent was obtained and before their children's medical evaluation, parents completed the QPGS in the waiting room. A trained interviewer introduced the questionnaire and was available for assistance. Four physicians, all with subspecialty certification in pediatric gastroenterology, conducted the medical evaluations. The medical evaluation included medical history, review of records from the referring primary care provider, family and social history, review of systems, and complete physical examination. Specific laboratory tests and procedures were conducted at the discretion of the attending physician. One year after the initial medical evaluation, each child's medical record was reviewed for evidence of organic disease associated with abdominal pain.

Data Analysis

Parents' responses on the QPGS were examined to determine which FGID symptom criteria, if any, were met by each child.


Patient Demographic Characteristics

Patients ranged in age from 4 to 17 years (mean = 10.01 years, SD = 3.52 years). Distribution by gender was almost equal (51.3% males). The sample was 89% white, 6% African-American, and 5% other ethnicities.

Results of Medical Evaluation

Review of medical records indicated that 7 of the 114 patients with RAP had organic disease that could explain their abdominal pain. Diagnoses included gastroesophageal reflux disease (n = 4), gallstones (n = 2), and inflammatory bowel disease (n = 1). These patients were excluded from all additional analyses.

Frequency of Functional Gastrointestinal Disorders

Table 1 shows, for each FGID associated with abdominal pain, the number and percent of patients with RAP without identifiable organic disease who met the symptom criteria for that FGID. Symptom criteria were met for irritable bowel syndrome (44.9%), functional dyspepsia (15.9%), functional abdominal pain (7.5%), and abdominal migraine (4.7%). Thus, in this sample of patients without identifiable organic etiology for RAP, 73% (n = 78) had symptoms consistent with the Rome criteria for an FGID associated with abdominal pain. Some of these patients (11.2%; n = 12) also met the symptom criteria for an FGID not associated with abdominal pain. Specifically, five patients who met the symptom criteria for irritable bowel syndrome (IBS) also met the symptom criteria for functional fecal retention. In addition, three patients who met the symptom criteria for abdominal migraine also met the symptom criteria for cyclic vomiting (the Rome criteria do not specify conditions under which a diagnosis of abdominal migraine would take precedence over a diagnosis of cyclic vomiting or vice versa).

Number and percent of 107 patients with RAP who met Rome II symptom criteria for FGIDs associated with abdominal pain

Subgroups of Functional Dyspepsia

Patients who met the symptom criteria for functional dyspepsia were further classified into subgroups of ulcer-like dyspepsia, dysmotility-like dyspepsia, and unspecified dyspepsia, based on the Rome criteria. Of the 17 patients who met symptom criteria for functional dyspepsia, most met the criteria for either ulcer-like dyspepsia (n = 11) or unspecified dyspepsia (n = 5). Only one child met the criteria for dysmotility-like dyspepsia.

Subgroups of Irritable Bowel Syndrome

We explored the QPGS data to determine whether patients meeting the symptom criteria for IBS could be grouped into subtypes of constipation-predominant versus diarrhea-predominant IBS. Patients with fewer than three bowel movements per week or whose usual stool consistency was “hard” or “very hard” were considered to have constipation-predominant IBS. Patients whose usual stool consistency was described as “mushy” or “watery” were considered to have diarrhea-predominant IBS. Using these definitions, 28 of the 48 (58.3%) patients with IBS were constipation predominant and 7 (14.6%) were diarrhea predominant.

Patients Not Meeting Rome Criteria for any FGID

Approximately one-quarter of the patients did not meet the Rome criteria for any FGID. Analysis of their QPGS data revealed several reasons that they did not meet criteria for an FGID: (1) 10 patients had too few symptoms to meet the criteria for IBS but had to be excluded from functional abdominal pain (FAP) because their pain was sometimes associated with bowel symptoms (e.g., pain relief with defecation) or with physiological events (e.g., eating, menses); (2) in 8 patients, abdominal pain episodes were less frequent than the “12 weeks of continuous or nearly continuous abdominal pain” required by the Rome criteria for FAP; (3) the abdominal pain episodes of 9 patients were consistent with the criteria for abdominal migraine, but they did not meet the additional Rome criteria for two or more associated features (e.g., headache during episodes, family history of migraine); and (4) 2 patients were excluded for miscellaneous reasons.


This study provides the first systematic empirical evidence that RAP, as defined by Apley (1), includes children with symptoms consistent with the Rome symptom criteria of several FGIDs. After a medical evaluation that yielded no evidence of organic disease, more than two-thirds of cases met the Rome symptom criteria for one of the FGIDs associated with abdominal pain. Irritable bowel syndrome and functional dyspepsia were the most common FGIDs. Additional research is needed to evaluate the extent to which these FGIDs can be reliably identified in children, may be characterized by different pathophysiological and psychosocial processes, and may benefit from different treatment approaches.

Consistent with studies linking RAP to IBS (5,10), nearly half of the patients with RAP in this study met the symptom criteria for IBS. Moreover, the relative frequency of the FGIDs in this sample reflects what is known about the prevalence of these conditions in the general pediatric population (e.g., the frequencies for IBS and dyspepsia were greater than that for abdominal migraine). Thus, the symptom clusters identified by the Rome classification system appear to reflect prior observations. In future clinical research on RAP, the Rome classification system may be useful for describing symptom heterogeneity within RAP samples and for selecting patients with similar symptom profiles for in-depth investigation. Whether the classification system has clinical utility remains to be investigated.

Unexpectedly, fewer than 8% of patients with RAP met the Rome symptom criteria for FAP, and 27% of patients with RAP did not meet the symptom criteria for any FGID associated with abdominal pain. These findings highlight the difficulty of defining symptom criteria for nonspecific abdominal pain in children. The Rome criteria define FAP as at least 12 weeks of continuous or nearly continuous abdominal pain associated with some loss of daily functioning and unrelated to physiologic events, such as eating or defecation. Thus, a child with persistent daily abdominal pain that sometimes was associated with defecation would not meet the symptom criteria for FAP. In the current sample, several children had enough IBS-like symptoms to be excluded from FAP but not enough to meet the criteria for IBS. Consequently, although their abdominal pain was of sufficient concern to merit evaluation in a tertiary care center, they were among those patients in the sample who did not meet the symptom criteria for any FGID defined by the Rome criteria.

The criteria of impairment present another difficulty. The Rome criteria for FAP are similar to Apley's criteria in that pain must be “severe enough to interfere with activities.” (1) We have argued elsewhere (6) that this requirement confounds severity and impairment. Neither Apley's criteria nor the Rome criteria have a means for classifying children whose abdominal pain, because of children's coping efforts or parental pressure, does not interfere with daily activities. Furthermore, neither Apley nor the Rome criteria clearly specify the extent or frequency of activity interference required to meet the impairment criterion, although the wording of the criteria suggests that impairment must be substantial. In most cases of FAP in this study, parents reported that abdominal pain interfered with their children's activities no more frequently than “one to three times a month.” This finding suggests that functional abdominal pain may occur with only minimal interruption of children's activities.

Several limitations of the study must be considered in interpreting the results. The findings are based on patients presenting with abdominal pain and cannot be generalized to an unselected population of patients presenting to a pediatric gastroenterology clinic. For example, the low incidence of dysmotility-like dyspepsia in our sample may reflect the fact that patients with dyspepsia characterized by nausea without abdominal pain were not eligible for the study because they did not meet Apley's criteria for RAP. Another study limitation is that participants were recruited from a tertiary care setting and probably had more severe symptoms than those in a primary care setting. It also is possible that the medical evaluation failed to identify existing organic disease. However, results of a prior prospective study suggest that less than 5% of patients would be likely to receive an organic diagnosis in the year after a medical evaluation for RAP at this center (11). It may be important that some patients in this study who met the symptom criteria for functional dyspepsia were treated with H-2 blockers and, in the absence of endoscopy, it is not known whether they in fact had significant organic disease. Finally, children's symptoms were assessed by parents and may not reflect children's own perceptions of their symptoms. For example, parents may have difficulty assessing whether abdominal pain or nonpainful discomfort (e.g., nausea, bloating) is more bothersome to the child; this judgment is required by the Rome criteria to discriminate ulcer-like dyspepsia from dysmotility-like dyspepsia. Future research should compare parents' and children's reports about children's gastrointestinal symptoms and should identify factors, such as child age, that determine whose report is more valid and reliable. In addition, symptom report on the QPGS should be validated against the clinical interview.

It would be premature to recommend changes in the Rome symptom criteria on the basis of this single report. Nonetheless, the current findings highlight areas for further investigation. For example, since children whose symptoms were partly consistent with IBS could not be classified as having IBS or FAP, it may be useful to modify the criteria for FAP so that children are not excluded if they have some symptoms of IBS but do not meet the full criteria for IBS. Moreover, it may be reasonable to adopt the distinction, developed in the Rome II criteria for adults, between functional abdominal pain syndrome and unspecified functional abdominal pain. A diagnosis of unspecified functional abdominal pain could be applied when pain does not interfere significantly with activities or when the child has subclinical levels of IBS-like symptoms. The next revision of the Rome criteria should specify criteria for functional constipation and functional diarrhea in school-age children (the criteria for preschoolers are already established). Although not necessarily the cause of abdominal pain, constipation often has been observed in children with RAP (12).

Ideally, diagnostic criteria meet the needs of both clinical practice and research. This presents a challenge because these needs may be at odds with each other. For example, practitioners may prefer criteria that are sufficiently flexible to diagnose atypical cases. Researchers may require criteria that are sufficiently strict to achieve homogeneous samples for research. Thus, a practitioner might diagnose IBS in a patient whose symptom duration was somewhat less than the 3-month criterion, but a researcher would exclude this patient from a study of IBS.

A large body of clinical and empirical research has accumulated since Apley first identified RAP half a century ago (13). An important next step is to investigate subgroups of patients with similar symptom profiles to evaluate whether they may have different underlying etiologic mechanisms that eventually could become the basis for an etiology-based classification system. The pediatric Rome criteria may facilitate this research by providing the means to categorize the FGIDs using standardized symptom-based diagnostic criteria. By allowing positive identification of patients with IBS, the Rome criteria for adults have contributed to the discovery of altered visceral perception in some of these patients (14,15). Similarly, the pediatric Rome criteria already have been used in clinical research to distinguish IBS from functional dyspepsia, resulting in the finding that these groups may differ in visceral sensitivity (16). Ultimately, characterization of pediatric FGIDs by Rome or other criteria will be useful only if additional empirical research demonstrates that diagnostic subgroups can be reliably identified, have etiologic specificity, and benefit from different treatments.


The authors thank the staff of the Pediatric Gastroenterology, Hepatology and Nutrition Clinic at Vanderbilt University, including Cheryl Little, MD; Cathy Arthur, MD; Stanley McQuiston, RN, CPNP; Kimberly Isenberg, CPNP; and Carol Collett, RN, for assistance with medical evaluations, and Beverly French and Deborah Simpson for assistance with patient scheduling. The authors also thank Courtney Dickens for administering research protocols.


1. Apley J. The Child With Abdominal Pains. London: Blackwell, 1975.
2. Oster J. Recurrent abdominal pain, headache, and limb pains in children and adolescents. Pediatrics 1972; 50:429–35.
3. Barr R. Recurrent abdominal pain. In: Levine MD, Carey WB, Crocker AC, et al., eds. Developmental-Behavioral Pediatrics. Toronto: WB Saunders; 1983:521–8.
4. Boyle J. Recurrent abdominal pain: an update. Pediatr Rev 1997; 18:310–20.
5. Hyams JS, Treem WR, Justinich CJ, et al. Characterization of symptoms in children with recurrent abdominal pain: resemblance to irritable bowel syndrome. J Pediatr Gastroenterol Nutr 1995; 20:209–14.
6. Von Baeyer C, Walker LS. Children with recurrent abdominal pain: research criteria for selection of subjects. J Dev Behav Pediatr 1999; 20:307–13.
7. Drossman DA, ed. The Functional Gastrointestinal Disorders. Diagnosis, Pathophysiology, and Treatment: A Multinational Consensus. Boston: Little, Brown; 1994:1–23.
8. Rasquin-Weber A, Hyman PE, Cucchiara S, et al. Childhood functional gastrointestinal disorders. Gut 1999: 45(SII):60–8.
9. Walker LS, Caplan-Dover A, Rasquin-Weber A. Manual for the Questionnaire on Pediatric Gastrointestinal Disorders. Nashville, TN: Department of Pediatrics, Vanderbilt University School of Medicine, 2000.
10. Walker LS, Guite JW, Duke M, et al. Recurrent abdominal pain: a potential precursor of irritable bowel syndrome in adolescents and young adults. J Pediatr 1998; 132:1010–5.
11. Walker LS, Garber J, Van Slyke DA, et al. Long-term health outcomes in patients with recurrent abdominal pain. J Pediatr Psychol 1995; 20:233–45.
12. Edwards MC, Finney JW, Bonner M. Matching treatment with recurrent abdominal pain symptoms: an evaluation of dietary fiber and relaxation treatments. Behav Therapy 1991; 22:257–67.
13. Walker LS. The evolution of research on recurrent abdominal pain: history, assumptions, and a conceptual model. In: McGrath PJ, Finley GA, eds. Progress in Pain Research and Management. Vol. 13. Seattle, WA: International Association for the Study of Pain Press; 1999:141–72.
14. Naliboff BD, Munakata J, Fullerton S, et al. Evidence for two distinct perceptual alterations in irritable bowel syndrome. Gut 1997; 41:505–12.
15. Mertz H, Naliboff B, Munakata J, et al. Altered rectal perception is a biologic marker of patients with irritable bowel syndrome. Gastroenterology 1995; 109:40–52.
16. Van Ginkel R, Voskuijl WP, Benninga MA, et al. Alterations in rectal sensitivity and motility in childhood irritable bowel syndrome. Gastroenterology 2001; 120:31–8.

Functional gastrointestinal disorders; Irritable bowel syndrome; Recurrent abdominal pain

Supplemental Digital Content

© 2004 Lippincott Williams & Wilkins, Inc.