The newly introduced Rome V criteria for irritable bowel syndrome may miss about one-third of patients who would meet a clinical reference standard for irritable bowel syndrome (IBS), according to a diagnostic accuracy study conducted in a UK secondary-care clinic. The diagnostic accuracy study included consecutive adults aged 16 years and older referred with suspected IBS to a specialist IBS clinic at Leeds Teaching Hospitals NHS Trust, Leeds, UK, from Sept. 22, 2016, to June 18, 2024.
The findings, published in The Lancet Gastroenterology & Hepatology, represent the first independent validation of the Rome V criteria since their introduction in May 2026. Although the revised criteria were intended to better distinguish irritable bowel syndrome (IBS) from centrally mediated abdominal pain syndrome, the investigators found that they identified a different group of patients than the Rome III and Rome IV criteria and excluded many patients who would likely still be diagnosed with IBS in routine clinical practice, the authors said.
"IBS remains a clinical diagnosis," first author Kyle Staller, MD, MPH, of the division of gastroenterology at Massachusetts General Hospital and Harvard Medical School, Boston, told GI & Hepatology News. “The Rome criteria are valuable because they help us standardize that diagnosis, but they are not a substitute for clinical judgment. Rome V is an important step in the continuing evolution of IBS diagnosis, but our findings suggest that some refinement may be needed, particularly around how we separate IBS from continuous abdominal pain syndromes. For patients, the diagnosis has real meaning. It validates the symptom pattern, reduces the impulse toward endless testing, and allows us to focus on treatments that can improve quality of life.”
Dr. Staller and colleagues evaluated the diagnostic performance of the Rome V criteria in 726 consecutive adults referred to a UK specialist clinic with suspected IBS between 2016 and 2024. Before undergoing guideline-recommended testing to rule out organic gastrointestinal disease, patients completed standardized symptom questionnaires. The investigators performing the diagnostic evaluations were blinded to the questionnaire results. The reference standard defined IBS as lower abdominal pain or discomfort associated with changes in stool frequency or stool form in patients without evidence of organic gastrointestinal disease after standardized testing. The investigators then compared the diagnostic performance of the Rome V criteria with that of the Rome IV and Rome III criteria using this reference standard.
Among the 726 patients with complete Rome V data, 417, or 57%, met the new diagnostic criteria. However, 590 patients met the reference standard for IBS, and only 390 of them also met the Rome V criteria, resulting in a sensitivity of 66%. The Rome V criteria had a specificity of 80%. By comparison, the Rome IV criteria achieved a sensitivity of 79% and a specificity of 81%, while the Rome III criteria had a sensitivity of 88% and a specificity of 75%.
The main reason patients were missed was the new Rome V requirement that abdominal pain not be continuous. Among the 200 patients who met the reference standard but did not meet the Rome V criteria, 71% were excluded because they reported abdominal pain every day of the week.
Agreement between the Rome V criteria and earlier diagnostic definitions was also limited. Agreement with the Rome III and Rome IV criteria ranged from fair to moderate, suggesting that the new criteria identify a different group of patients than the earlier versions.
Dr. Staller noted that the current analysis did not use the official Rome V questionnaires, which are now available for use in research settings. “Some subtle differences between the questionnaires used in the current study and the official questionnaires—particularly in regard to the definition of continuous abdominal pain—may affect the reported performance of the Rome criteria,” he said. “Nevertheless, these findings may have important implications for the way in which these new criteria are applied.”
Adding the Rome V supportive criterion that abdominal pain or discomfort in women should not occur only during menstruation had little effect on overall diagnostic performance. Only 11% of women who met the Rome V criteria reported symptoms occurring exclusively during their menstrual periods. Including this criterion slightly improved specificity but reduced sensitivity even further.
Patients who met the Rome V criteria also differed clinically from those identified by the earlier definitions. Compared with patients who met the Rome III or Rome IV criteria, they had lower IBS Severity Scoring System scores, indicating milder disease on average. They also reported lower levels of anxiety, depression, and extraintestinal somatic symptoms, suggesting that the Rome V criteria may preferentially identify patients with less severe illness. The authors suggested the criteria might perform better in primary care than in a secondary-care referral setting, but emphasized that this has not yet been demonstrated.
Only about 5% of patients who met any of the three Rome criteria were ultimately found to have an organic gastrointestinal disease after diagnostic testing. The most common alternative diagnosis was bile acid diarrhea, followed by exocrine pancreatic insufficiency, microscopic colitis, and Crohn’s disease.
For clinicians, the findings suggest that patients should not be considered unlikely to have IBS simply because they do not meet the Rome V criteria, particularly if they have continuous abdominal pain. The relatively low sensitivity and higher negative likelihood ratio indicate that many patients with clinically confirmed IBS may no longer meet the formal Rome V definition despite having symptoms consistent with the disorder.
Dr. Staller reported research support from Ardelyx and consulting relationships with AbbVie, Ardelyx, Atmo, Ferring, Gemelli, Laborie, Mahana, Mindset Health, Salix, and Takeda. The remaining authors reported no competing interests.
Expert Insight
GI & Hepatology News asked Dr. Staller to elaborate on the study findings.
Looking at the complete results, what surprised you most?
Dr. Staller: The most surprising finding was that Rome V was less sensitive than either Rome III or Rome IV. Among patients who met our reference standard for IBS, only about two-thirds met Rome V criteria, compared with nearly 79% for Rome IV and 88% for Rome III. The specificity of Rome V was reasonable, but it was not meaningfully better than Rome IV.
The other striking finding was why patients might not meet the new criteria. More than 70% of the patients who met the reference standard for IBS but did not meet Rome V criteria failed to do so because they reported continuous abdominal pain every day. That raises a very practical question: are these patients truly different, or are we asking the question in a way that patients interpret differently than clinicians intend?
Most gastroenterologists recognize that IBS symptoms can cluster over several days, and during those flares patients may experience pain that feels essentially continuous. That is not necessarily the same thing as centrally mediated abdominal pain syndrome. So the Rome V effort to distinguish IBS from continuous pain syndromes is conceptually important, but the way it is operationalized may be too blunt.
How might the findings influence clinical practice?
Dr. Staller: I would not interpret these findings as saying that Rome V should be ignored. Rather, they suggest that Rome V should be applied thoughtfully, especially around the issue of continuous abdominal pain.
For the practicing gastroenterologist, the key message is that not meeting Rome V criteria should not automatically end the diagnostic conversation. In our study, the negative likelihood ratio for Rome V was relatively high, meaning that a negative Rome V result only modestly reduced the probability that a patient truly had IBS. In plain terms, some patients who fail Rome V criteria will still look very much like IBS patients in the clinic.
The findings also reinforce that a positive diagnosis of IBS can still be made safely with a careful history, attention to alarm features, and limited, guideline-consistent testing. Only 5% of patients meeting Rome V, Rome IV, or Rome III criteria had organic disease identified, with bile acid diarrhea being the most common alternative diagnosis. That should remind us not to over-test everyone, but also not to forget targeted evaluation in the right clinical context, particularly in IBS-D.
What gaps remain, and what research should be done next?
Dr. Staller: The biggest gap is generalizability. This was a single-center secondary-care study in the UK, in a relatively homogeneous population. Rome V now needs to be tested in other settings: primary care, tertiary referral centers, community gastroenterology practices, and geographically and racially diverse populations.
The second major gap is the language around continuous abdominal pain. A binary yes/no item may not capture the distinction clinicians are trying to make between IBS flares, overlapping pain amplification, and centrally mediated abdominal pain syndrome.