Not yet independently audited
This page was written from verified sources but has not yet passed the independent citation audit. Treat its claims as provisional.
Open questions — irritable bowel syndrome¶
Last curated: 2026-10-21 (literature sweep; older dated searches remain dated to their original windows). Stable IDs are OQ-n. OQ-29 was sharpened on 2026-10-21; OQ-4/D5 and OQ-5 retain the 2026-10-14 updates; OQ-13, OQ-15 and OQ-28 retain the 2026-09-30 updates. Dated absences that were re-searched in the window and stayed empty are named in the note at the end. No new question was opened.
Tier 1 = answering it would change practice and a study is designable today. Tier 2 = blocked on tools, numbers, or a Tier-1 answer.
Dots not yet connected¶
Cross-domain junctions where two bodies of evidence in this knowledge base both exist and no study has joined them.
| # | Dot A | Dot B | The missing junction | Powers |
|---|---|---|---|---|
| D1 | Rome III, Rome IV and Rome V select measurably different populations — agreement κ 0.36–0.55, sensitivities 87.5% / 78.9% / 66.1% in the same 726 patients (Staller 2026, PMID 42392123) | Every meta-analysis in this knowledge base pools trials across criteria eras: rifaximin's registration used Rome II (PMID 21208106), ATLANTIS used Rome IV (PMID 37858323), most low-FODMAP trials used Rome III | No trial or IPD meta-analysis has stratified a treatment effect by which criteria set defined the population. Effect estimates are being transported across populations that demonstrably differ | OQ-1 |
| D2 | A validated nine-variable risk score predicts post-infectious IBS with AUC 0.70, separating strata at 10%, 35% and 60% incidence (Thabane 2009, PMID 19568228) | Post-infectious IBS has a 14.5% incidence and OR 4.3 after acute gastroenteritis, i.e. a large, identifiable, temporally defined at-risk population (Porcari 2024, PMID 39013599) | No prevention trial has ever used the score to enrich a high-risk cohort. The only condition in this knowledge base with a datable exposure and a validated pre-symptomatic predictor has no prevention literature (searched 2026-09-02; date-restricted search through 2026-09-30 still returned no prevention trial) | OQ-6 |
| D3 | Low-dose amitriptyline works in primary care (ATLANTIS, IBS-SSS −27.0; PMID 37858323) | Remote CBT works in refractory IBS with roughly double that effect (ACTIB, IBS-SSS −61.6 telephone / −35.2 web at 12 months; PMID 30971419) and hypnotherapy holds 61.8% responders at two years (PMID 40491242) | The two best non-dietary treatments have never been compared head to head, in any setting, and no guideline can therefore sequence them | OQ-3 |
| D4 | Diet beat optimised drug therapy in a specialist clinic (CARIBS, 76%/71% vs 58%; PMID 38643782) and a smartphone FODMAP app beat an antispasmodic in primary care (DOMINO, 71% vs 61%; PMID 35483886) | Every major guideline was written before or alongside these trials and none sequences diet ahead of pharmacotherapy (see wiki/guidelines.md) | No guideline has tested or adopted diet-first sequencing, and no trial has compared a dietary pathway against ATLANTIS-style amitriptyline in the same primary-care population | OQ-7 |
| D5 | 2,797 PubMed records for irritable bowel syndrome AND (microbiome OR microbiota) on 2026-09-02 with reproducible taxon-level differences (Pittayanon 2019, PMID 30940523; Li 2024, PMID 39777150) |
Zero validated microbiome diagnostics; probiotic certainty rated low to very low across almost every analysis of 82 trials (Goodoory 2023, PMID 37541528); FMT meta-analyses disagree on whether an effect exists at all (PMID 39289768 vs PMID 38999862) | Prospective selection is now registered, without outcomes. NCT07872514 (first posted 2026-10-13; n=30; enrolling by invitation; no registry results) selects a probiotic using screening histamine or LPC/LPA for masked n-of-1 crossovers. The question is now whether selection improves outcomes and generalises beyond the pilot. The earlier compositional rifaximin predictor remains unreplicated (Li 2020, PMID 32470562) | OQ-4, OQ-13 |
| D6 | Enterochromaffin-cell activation is sufficient to produce persistent visceral hypersensitivity in mice without inflammation, and the circuit is tonically engaged in females (Bayrer 2023, PMID 36949192; Venkataraman 2025, PMID 41411420) | IBS is 1.5-fold commoner in women (OR 1.46; PMID 32702295) and 5-HT3 antagonists were historically licensed by sex — alosetron for women, ramosetron initially for men | No human study has tested whether the estrogen-responsive L-cell → PYY → EC-cell → serotonin circuit operates in people, or whether it explains the sex ratio or predicts 5-HT3 antagonist response | OQ-14 |
| D7 | A warm practitioner relationship raises adequate relief from 44% to 62% (Kaptchuk 2008, PMID 18390493), and open-label placebo does not differ significantly from blinded placebo (d=0.10; Lembo 2021, PMID 33605656) | 54% of patients' spontaneous comments about their clinicians are negative, and only 38.3% report receiving comprehensive information against 96% who want it (Halpert 2011, PMID 21561228; Halpert 2010, PMID 19513835) | No implementation trial has tested whether deliberately delivering the relational component at scale improves outcomes, despite its randomised effect exceeding most licensed drugs | OQ-8 |
| D8 | Bile acid diarrhoea affects 28.1% of IBS-D and responds to sequestrants (RR 1.50 stool consistency, 2.80 frequency; PMID 25913530, 41090475); exocrine pancreatic insufficiency affects 5–6.1% and responds to enzymes (PMID 19835990, 35704255) | IBS-D drugs are licensed and prescribed without testing for either, and the IBS-D guideline contains no strong recommendation at all (Lembo 2022, PMID 35738725) | No trial has randomised a test-first pathway against empirical IBS-D treatment. Two treatable mimics with double-digit and single-digit prevalence are routinely untested | OQ-9 |
| D9 | Blinded FODMAP reintroduction shows an inert glucose control triggers symptom recurrence in 26% of exposures (Van den Houte 2024, PMID 38401741) | Dietary restriction in IBS is widespread, dietitian-scarce, and carries measurable microbial costs — reduced Bifidobacterium, Actinobacteria and butyrate, with the trialists advising against strict long-term use (Wilson 2020, PMID 32433273) | Nobody has quantified how much of real-world dietary restriction is nocebo-driven, or tested a blinded-reintroduction protocol against open reintroduction for long-term dietary breadth | OQ-10 |
| D10 | 34.6–55.3% of people with IBS also meet criteria for functional dyspepsia, and the overlap independently predicts worse symptoms, more consultations, more drugs and worse mood at 12 months (Andreev 2022, PMID 36286762; Barberio 2022, PMID 33839276) | No IBS treatment trial retrieved in this build measured, stratified on, or reported concurrent functional dyspepsia | Over half of every IBS trial population probably has an unmeasured prognostic factor, which the trial's own authors have flagged as having "important implications for future treatment trials" | OQ-11 |
| D11 | 38% of tertiary-care IBS patients had contemplated suicide because of their bowel symptoms, with depression scores not explaining the variance (Miller 2004, PMID 15625650) | Enacted stigma is experimentally higher for IBS than for IBD or asthma when only the label varies (Taft 2017, PMID 27501483), and invalidation causally links to suicidality, healthcare avoidance and diagnostic delay across 151 qualitative reports (Bontempo 2025, PMID 40310228) | No study has tested whether reducing stigma or invalidation changes suicidality, healthcare avoidance or diagnostic delay in IBS. The 2004 figure still has no comparable adult multi-tier replication (re-searched through 2026-09-30). A paediatric school cohort is a different outcome and design (Li W 2026, PMID 42787978) | OQ-15 |
| D12 | 81.8% of people with IBS report overall work impairment and 85.6% presenteeism, costing an estimated 72–188 million UK working hours a year (Goodoory 2022, PMID 35794733) | Work productivity appears only as a secondary endpoint in industry drug trials (e.g. PMID 25237424, 33769858) | No IBS trial has used work or social function as its primary endpoint, so no treatment has ever been selected for its effect on the outcome patients and payers care most about | OQ-16 |
Tier 1 — designable today, would change practice¶
OQ-1. Are treatment effects criteria-dependent?¶
Rome III, Rome IV and Rome V identify overlapping but distinguishable populations: in 726 secondary-care patients, sensitivities were 87.5%, 78.9% and 66.1% with agreement κ 0.36–0.55 (Staller 2026, PMID 42392123), and Rome III → Rome IV halves community prevalence (Palsson 2020, PMID 31917991). Every network meta-analysis in this knowledge base pools across these eras. Designable now: individual-participant-data meta-analysis of existing trials, re-applying each criteria set to baseline symptom data and testing effect-modification. → diagnosis-and-rome-criteria, placebo-response-and-trial-design
OQ-2. What is the placebo response made of, and can any design separate it from drug effect?¶
Placebo response is 27.3% for global improvement, 34.4% for pain and 17.9% for the FDA composite (Bosman 2021, PMID 33765447) — and falls to 4.3% for the composite when the responder window tightens from 6 to 9 of 12 weeks (Barberio 2022, PMID 34425274). Its components are separable and dose-like (Kaptchuk 2008, PMID 18390493), and open-label placebo is statistically indistinguishable from blinded placebo (Lembo 2021, PMID 33605656). Designable now: a four-arm trial of drug vs blinded placebo vs open-label placebo vs no treatment, powered on the drug-vs-OLP contrast. No such trial was retrieved on 2026-09-02, and a date-restricted PubMed search through 2026-09-30 returned none. → placebo-response-and-trial-design
OQ-3. Neuromodulator or brain–gut behavioural therapy first, in primary care?¶
ATLANTIS established low-dose amitriptyline as effective second-line treatment in 55 English general practices (IBS-SSS −27.0; PMID 37858323, 39397570). ACTIB established telephone and web CBT in refractory IBS with roughly double the effect (−61.6 and −35.2 at 12 months; PMID 30971419), and remote delivery is statistically indistinguishable from face-to-face (Tao 2026, PMID 41505702). One is cheap and available; the other is more effective and rationed. A date-restricted PubMed search from 2026-09-02 through 2026-09-30 still retrieved no head-to-head trial. Designable now: a pragmatic three-arm primary-care trial with a health-economic endpoint. → gut-brain-neuromodulators, psychological-therapy
OQ-6. Can post-infectious IBS be prevented?¶
A nine-variable score derived and split-sample-validated in 1,368 exposed patients discriminates at AUC 0.70 and separates strata with 10%, 35% and 60% post-infectious IBS (Thabane 2009, PMID 19568228). Incidence after acute gastroenteritis is 14.5% with OR 4.3 (Porcari 2024, PMID 39013599). Designable now: score-enriched randomised prevention of any candidate intervention during or immediately after acute enteritis. No such trial was retrieved on 2026-09-02, and a date-restricted PubMed search through 2026-09-30 returned no prevention trial. → post-infectious-ibs
OQ-7. Should diet be sequenced before drugs?¶
Two independent randomised trials in different countries, settings, comparators and delivery modes both favour diet: CARIBS 76%/71% vs 58% against optimised medical treatment (Nybacka 2024, PMID 38643782) and DOMINO 71% vs 61% against otilonium bromide in primary care (Carbone 2022, PMID 35483886), with adherence 94% vs 73%. No guideline sequences diet first. Designable now: a pragmatic primary-care trial of a dietary pathway versus low-dose amitriptyline, the two interventions with the best primary-care evidence. → dietary-therapy, guidelines
OQ-8. Does delivering the relational component of care at scale improve outcomes?¶
The randomised effect of a warm, attentive practitioner relationship — 62% vs 44% adequate relief, against 28% for observation alone (Kaptchuk 2008, PMID 18390493) — exceeds the drug–placebo difference of every licensed IBS drug in this knowledge base. Patients report the opposite of what the trial delivered: 54% negative provider comments, 38.3% receiving comprehensive information against 96% wanting it (Halpert 2011, PMID 21561228; Halpert 2010, PMID 19513835). Designable now: cluster-randomised implementation trial of a structured consultation intervention, with IBS-SSS as the endpoint. → placebo-response-and-trial-design, quality-of-life-and-stigma
OQ-9. Would a test-first pathway beat empirical treatment in IBS-D?¶
Bile acid diarrhoea affects 28.1% (22.6–34.0) of IBS-D and responds to sequestrants with moderate certainty (Slattery 2015, PMID 25913530; Dilmaghani 2025, PMID 41090475); exocrine pancreatic insufficiency affects 5–6.1% and responds to enzyme replacement (Leeds 2010, PMID 19835990; Olmos 2022, PMID 35704255); microscopic colitis accounts for the entire excess organic yield in diarrhoeal phenotypes and is missed for want of biopsies in one in four cases (Asghar 2022, PMID 32882424). Designable now: randomise test-first versus empirical IBS-D management on symptom and cost endpoints. → differential-diagnosis-and-exclusion, diarrhoea-predominant-pharmacotherapy
OQ-10. How much of dietary "trigger" identification is nocebo?¶
In blinded reintroduction, an inert glucose control triggered symptom recurrence in 26% of exposures, against fructans 56% and mannitol 54%, with a mean of only 2.5 ± 2 genuine triggers per patient (Van den Houte 2024, PMID 38401741). Real-world restriction is far broader. Designable now: randomise blinded versus open reintroduction, with dietary breadth, nutritional adequacy and microbial outcomes at 12 months. → dietary-therapy, red-flags-and-safety-concerns
OQ-11. Does concurrent functional dyspepsia modify IBS treatment response?¶
Overlap affects 34.6% (28.2–41.4) pooled and 55.3% of a Rome IV cohort, and independently predicts more severe symptoms, more consultations, more drugs and worse mood at 12 months (Andreev 2022, PMID 36286762; Barberio 2022, PMID 33839276). No IBS trial retrieved measures it. Designable now: re-analyse existing trial datasets for overlap status as an effect modifier; prospectively stratify future trials. → overlap-with-functional-dyspepsia
OQ-12. Should ondansetron be licensed for IBS-D?¶
It ranks first for stool consistency among 5-HT3 antagonists (SUCRA 0.98) in a 21-trial, 10,421-patient network (Rokkas 2021, PMID 34276193), and meta-analysis of three trials gives NNT 9 for the FDA composite and NNT 5 for stool response — though no pain benefit (RR 0.95, 0.74–1.20) (Gunn 2023, PMID 36866724). It is generic, cheap, and recommended by the Mexican position statement (Remes-Troche 2025, PMID 40307155) while licensed for IBS nowhere. TRITON recruited 80 of a planned 400. Designable now: an adequately powered publicly funded trial. → diarrhoea-predominant-pharmacotherapy
OQ-17. Should AGA's recommendation against SSRIs be revised?¶
AGA conditionally recommends against SSRIs in both IBS-C and IBS-D at low certainty (Chang 2022, PMID 35738724; Lembo 2022, PMID 35738725). The 2025 updated meta-analysis, published after both, finds SSRIs superior to placebo for abdominal pain (RR 0.74, 0.56–0.99; 7 RCTs, 324 patients) and explicitly highlights their potential (Khasawneh 2025, PMID 40258375). Designable now: an adequately powered SSRI trial with abdominal pain as the primary endpoint would settle it; the confidence interval currently touches unity. → gut-brain-neuromodulators, guidelines
OQ-18. Is peppermint oil's effect real, and is the endpoint or the drug at fault?¶
Peppermint oil ranked first for global symptoms in a 51-trial network (RR 0.63, 0.48–0.83; Black 2020, PMID 31859183) and its updated meta-analysis gives NNT 4 (95% CI 2.5–71) (Ingrosso 2022, PMID 35942669) — yet the largest and most rigorous single trial missed both co-primary regulatory endpoints while improving the same symptoms as continuous secondaries (Weerts 2020, PMID 31470006). Designable now: a trial powered for both dichotomous and continuous endpoints, prespecifying the comparison between them. → antispasmodics-and-peppermint
OQ-19. How much of the historical antispasmodic effect is molecule and how much is label?¶
A 2×2 factorial trial found mebeverine itself did nothing (23.4% vs 22.0%, OR 1.08, 0.59–1.99) while the label "mebeverine" nearly doubled treatment success (31.6% vs 14.1%, OR 2.84, 1.52–5.34, p=0.001) (Rexwinkel 2025, PMID 40074185). The trial was in adolescents. Meanwhile antispasmodics carry a pooled RR of 0.68 (0.57–0.81) from 22 mostly old, mostly poorly blinded trials (Ford 2008, PMID 19008265). Designable now: replicate the labelling design in adults, for an antispasmodic and for a neuromodulator. → antispasmodics-and-peppermint, placebo-response-and-trial-design
OQ-20. Should eluxadoline remain licensed?¶
Pancreatitis accounted for 16.4% of 597 post-marketing adverse-event reports, with 53 hospitalisations and two reported fatalities, against 0.2–0.5% for comparator agents (Gawron 2018, PMID 28804032); all ten sphincter-of-Oddi spasm events in pooled trial safety data occurred in patients without a gallbladder (Cash 2017, PMID 27922029). The NNT is roughly 8 (Brenner 2019, PMID 31356229). Designable now: a formal benefit–risk re-assessment using current pharmacovigilance data; the raw material exists. → red-flags-and-safety-concerns, diarrhoea-predominant-pharmacotherapy
Tier 2 — blocked on tools, numbers, or a Tier-1 answer¶
OQ-4. Has the microbiome literature produced anything clinically actionable?¶
Blocked on the gap between composition and intervention. Taxon-level differences replicate (Pittayanon 2019, PMID 30940523; Li 2024, PMID 39777150) but 40% of the underlying studies did not report whether cases and controls were comparable on age and sex; probiotic certainty is low to very low across 82 trials (Goodoory 2023, PMID 37541528); and the field's two most recent strain-specific syntheses disagree about Escherichia strains (PMID 37541528 vs PMID 41682832). A mechanism-selected pilot is now registered: NCT07872514 assigns probiotics using screening histamine or LPC/LPA and tests them against placebo in n-of-1 crossovers (planned n=30; enrolling by invitation; no results, API checked 2026-10-14). The remaining question is prospective clinical utility and external validation, rather than whether such selection is being attempted. → microbiome
OQ-5. What are the long-term harms of restrictive dietary therapy?¶
Low FODMAP reduces faecal Bifidobacterium, Actinobacteria and butyrate, and its trialists state "strict long-term use should not be advised" (Wilson 2020, PMID 32433273; Halmos 2015, PMID 25016597). Nutritional adequacy is preserved after reintroduction (O'Keeffe 2018, PMID 28707437). Risk scores years after dietitian-led FODMAP education are common — any tool 38%, EAT-26 7%, NIAS/ARFID 24%, ORTO-7 21% in 74 patients a mean of 7.1 years later — but multivariable analysis tied ARFID and orthorexia risk to psychological distress and symptom severity, not to measured FODMAP intake, self-perceived diet pattern or time since education, and no new eating-disorder diagnosis or exacerbation was recorded (Silva 2026, PMID 42712718). The new Rouen retrospective study separates NIAS-Fr screening (155/326; 47.5%) from a study-defined DSM-5-compatible classification (109/326; 33.4%; Table 2), without specialist diagnostic confirmation (Methods §2.3; Mikhael-Moussa 2026, PMID 42823900). Neither study measures controlled incident eating disorders after dietary treatment. That sharpens the question. It does not show that the diet caused disordered eating, and it is not a controlled incidence cohort (Scarlata 2025, PMID 39681226; Atkins 2023, PMID 36782302). Cross-links anorexia nervosa. → dietary-therapy, red-flags-and-safety-concerns
OQ-13. Why does one FMT trial report 89% response and two report none?¶
El-Salhy 2020 (PMID 31852769) found 23.6% / 76.9% / 89.1% response across placebo / 30 g / 60 g from a single well-characterised donor; Lahtinen 2020 (PMID 32343000) and Yau 2023 (PMID 37667968) were negative on their primary endpoints. Meta-analyses disagree on whether any effect exists (RR 1.44, 0.88–2.33, very low certainty — PMID 39289768 — versus OR 0.46, 0.33–0.64 — PMID 38999862). The Norwegian donor-versus-autologous phase 3 has now reported and does not resolve the positive trial. REFIT2 (NCT04691544) found 40% versus 38% on a ≥75-point IBS-SSS reduction at 90 days after one rectal enema (absolute difference 1.9 percentage points, 95% CI −8.1 to 12.0; p=0.76; Johnsen 2026, PMID 42785338). On 2026-09-30 the registry record was still active, not recruiting, with no results posted. The abstract does not report a ≥50-point result, the route and the responder threshold differ from El-Salhy, and no trial has randomised recipients among multiple allogenic donors (re-searched through 2026-09-30). Sharpened 2026-10-28. A Czech three-arm crossover trial using pooled multi-donor material from bacteriome-preselected donors was null on IBS-SSS (effect 2.5 points, 95% CI −33.6 to 38.6). This was despite measured engraftment of donor-derived strains (Cinek 2026, Research Square doi:10.21203/rs.3.rs-11289849/v1, preprint, not peer reviewed; registry match NCT04899869). The question now has a second half: if donor strains engraft and symptoms do not move, what did the El-Salhy single-donor material deliver that engraftment does not? → microbiome
OQ-14. Does the enterochromaffin-cell circuit operate in humans, and does it explain the sex ratio?¶
Mouse work shows EC-cell activation is sufficient for persistent visceral hypersensitivity without inflammation and that an estrogen-responsive L-cell → PYY → EC-cell → serotonin pathway heightens gut sensitivity in females (Bayrer 2023, PMID 36949192; Venkataraman 2025, PMID 41411420). Blocked on human-tissue and in-vivo methods. Would, if translated, explain both the sex ratio and the mechanism of 5-HT3 antagonists. → brain-gut-axis-and-visceral-hypersensitivity
OQ-15. Does reducing stigma or invalidation change outcomes?¶
Enacted stigma is experimentally higher for IBS than for IBD or asthma with only the label varying (Taft 2017, PMID 27501483); familiarity and empathy correlate with lower stigma; invalidation is causally linked to shame, suicidality, healthcare avoidance and diagnostic delay (Bontempo 2025, PMID 40310228). Blocked on the absence of a validated stigma-reduction intervention and on the fact that the 38% tertiary-care suicidality figure (Miller 2004, PMID 15625650) still has no comparable adult multi-tier replication (re-searched through 2026-09-30). A prospective school cohort of 92,120 adolescents found adjusted odds of 1.27 (1.18–1.37) for suicide risk and 1.21 (1.06–1.38) for attempts (Li W 2026, PMID 42787978). That result is paediatric, and the outcome is not bowel-specific contemplation. → quality-of-life-and-stigma, red-flags-and-safety-concerns
OQ-16. Should work and social function be primary trial endpoints?¶
81.8% overall work impairment, 85.6% presenteeism, 72–188 million UK hours lost annually (Goodoory 2022, PMID 35794733), replicated in Sweden (Frändemark 2018, PMID 30254230). Blocked on the absence of an agreed, responsive work-function endpoint validated for IBS trials — and on regulators' current composite-responder framework (placebo-response-and-trial-design). → quality-of-life-and-stigma
OQ-21. What distinguishes the third of patients who are not rectally hypersensitive?¶
At the optimal barostat cut-off, 63.5% of IBS patients and 6.6% of controls are hypersensitive, with no significant difference between subtypes (Ludidi 2012, PMID 22591192). Roughly 36% of criteria-positive patients are normosensitive by the field's most-used mechanistic measure. Blocked on the absence of a clinically deployable sensitivity assay; no treatment trial has ever stratified on barostat status. → brain-gut-axis-and-visceral-hypersensitivity
OQ-22. Is bile acid diarrhoea a differential or a mechanistic subtype of IBS-D?¶
A quarter to a third of IBS-D patients test positive and respond to sequestrants, yet there is no reference standard — Valentin and colleagues could estimate diagnostic yield only, not sensitivity or specificity, "relative to a gold standard test" (PMID 26347530) — and SeHCAT is unavailable across most of the world. Blocked on a reference test. → differential-diagnosis-and-exclusion
OQ-23. Does SIBO exist as a distinct entity, and can any threshold define it?¶
Breath-test SIBO is present in 35.5% of IBS patients and 29.7% of controls; lactulose testing yields 3.6-fold higher prevalence in patients and 7.6-fold higher in controls than glucose testing; jejunal culture gives 4% (Shah 2020, PMID 31913194; Ford 2009, PMID 19602448). North American (≥20 ppm) and Asia-Pacific (≥12 ppm) thresholds disagree and only the latter tracked symptom severity prospectively (Loh 2026, PMID 42235989). The consensus that set the North American thresholds attracted a published conflict-of-interest challenge (Maltz 2017, PMID 29215615). Blocked on a test that separates cases from controls. → rifaximin-and-the-sibo-question
OQ-24. Why does rifaximin retreatment keep the pain benefit but lose the stool benefit?¶
Retreatment gives 38.1% vs 31.5% overall (p=0.03) and 50.6% vs 42.2% for abdominal pain (p=0.018), but 51.8% vs 50.0% for stool consistency (p=0.42) (Lembo 2016, PMID 27528177), while the drug's stool-microbiota effect is "modest, largely transient" (Fodor 2019, PMID 29708822). Blocked on small-bowel rather than stool sampling. → rifaximin-and-the-sibo-question
OQ-25. Can pain relief be pharmacologically separated from secretion in IBS-C?¶
MD-7246 (ileocaecal-release linaclotide) improved abdominal pain versus placebo while producing bowel results similar to placebo, with placebo-equivalent diarrhoea rates (Chey 2021, PMID 33065589); PEG 3350 did the reverse, improving bowel frequency with no pain benefit versus placebo (Chapman 2013, PMID 23835436). Blocked on whether any sponsor develops a pain-only secretagogue to registration. → constipation-predominant-pharmacotherapy
OQ-26. Is prevalence rising, and is the answer real or methodological?¶
Meta-regression shows "a trend toward increased prevalence in recent years" (Ballena-Caicedo 2025, PMID 41488823), while Rome IV and Rome V global surveys six years apart return near-identical aggregate DGBI prevalence (40.5% vs 40.9%; Sperber 2026, PMID 42613194), and the US 6.1% figure was collected during the COVID-19 pandemic with that caveat stated by its authors (Almario 2023, PMID 37595647). Blocked on repeated surveys using one fixed criteria set. → epidemiology-and-burden
OQ-27. Why is Rome IV prevalence higher than Rome III in one 2025 meta-analysis and half of it in three others?¶
Ballena-Caicedo 2025 (PMID 41488823) reports 17.14% vs 13.21%; Oka 2020 (PMID 32702295), Sperber 2021 (PMID 32294476) and Palsson 2020 (PMID 31917991) all report Rome IV at roughly half Rome III. This is a direct unresolved contradiction, not a nuance. Blocked on a reconciliation analysis nobody has performed. A Chilean community study using the official Spanish Rome IV questionnaire returned 25.7–30.9% (von Muhlenbrock 2026, PMID 42928586). That is a single-criteria figure, so it does not test the Rome III-versus-IV direction, but it widens the spread of Rome IV estimates. The Latin American regional databases (SciELO, LILACS) returned HTTP 403 on 2026-10-28, so whether this figure is typical for the region cannot be assessed from the sources we can reach. → epidemiology-and-burden
OQ-28. Is there excess mortality in IBS?¶
Two cohorts now report no overall excess. Claims-identified IBS in NIH–AARP had a time-varying inverse association that the authors did not read as protective (Gutiérrez-Torres 2025, PMID 41518229; figures are on the epidemiology page and were not re-fetched on 2026-09-30). Rome II IBS in the Tohoku Medical Megabank Miyagi cohort had an all-cause HR of 0.94 (0.68–1.29) over a median 6.7 years; participants with IBS who had missed a health checkup in the previous year had HR 2.11 (1.06–4.19) against participants without IBS who had attended (Nakaya 2026, PMID 42702587). Rome II is not Rome IV or Rome V. A criteria-defined mortality follow-up on those later criteria was not in the date-restricted search through 2026-09-30. The suicidality signal (Miller 2004, PMID 15625650) is a different endpoint. → epidemiology-and-burden
OQ-29. Should the Rome IV pain-frequency threshold be relaxed, and what will Rome V do to trial populations?¶
A simple modification — pain only, but frequency relaxed from ≥1 day/week back to 3 days/month — outperformed unmodified Rome IV (sensitivity 90.2% vs 82.1% at identical specificity 85.1%; Goodoory 2025, PMID 39466700), and Rome V did not adopt it. Rome V's own first accuracy study gives sensitivity 66.1% (Staller 2026, PMID 42392123). In one Chinese tertiary clinic, after organic exclusion, Rome IV captured 238/685 (34.7%) of suspected IBS against 443 (64.7%) for Rome V and 533 (77.8%) for the Rome clinical criteria, and the paper reports that 90 (20.3%) of those identified by the clinical criteria were missed by Rome V, mainly on duration or frequency, with a burden comparable to Rome-V-only patients and lower than Rome IV (Wu 2026, PMID 42754968). That is one clinic's case-mix, not an accuracy study. Blocked on multi-centre validation outside Leeds and on a trial that stratifies a treatment effect by criteria set. A date-restricted PubMed search from 2026-10-14 through 2026-10-21 still found none. NCT07882953 (first posted 2026-10-20; not yet recruiting; planned n=180) uses Rome V subtype eligibility for an open-label SIMBA-capsule plus rifaximin protocol; that is an eligibility rule, not a criteria-stratified efficacy analysis. → diagnosis-and-rome-criteria
OQ-30. Should the non-IBS functional bowel disorders be abolished into an IBS spectrum?¶
Latent-class analysis of 558 people meeting Rome IV criteria for a non-IBS functional bowel disorder found poor correlation between clusters and diagnostic labels, classified 75% as "mild IBS" using an IBS-derived model, and found one third fluctuating into IBS within 12 months (Black 2022, PMID 35531932). Rome V retained the separate categories (Sperber 2026, PMID 42613194). Blocked on whether any nosological body acts on the finding. → diagnosis-and-rome-criteria
Cross-condition junctions¶
| Junction | Other condition | Why |
|---|---|---|
| Restrictive dietary therapy and disordered eating | anorexia nervosa | OQ-5; risk scores years after FODMAP education tracked distress and symptom severity rather than FODMAP intake (Silva 2026, PMID 42712718). The grey area between adaptive restriction and eating pathology remains unresolved (Scarlata 2025, PMID 39681226) |
| Chronic overlapping pain, central sensitisation, and the "contested illness" experience | fibromyalgia | The invalidation meta-synthesis pools IBS with fibromyalgia, ME/CFS, endometriosis and long COVID across 151 reports and 11,307 individuals (Bontempo 2025, PMID 40310228); both conditions share nociplastic-pain framing and a stigma problem |
| Shared genetic architecture with mood and anxiety disorders | depression, generalized-anxiety-disorder | rg > 0.5 between IBS and anxiety, neuroticism and depression, attributed to shared pathogenic pathways rather than causation in either direction (Eijsbouts 2021, PMID 34741163) |
| Post-infectious symptom persistence after a dated exposure | ptsd | Not mechanistic — methodological: both fields study persistent symptoms after a datable event, and IBS has the validated pre-symptomatic risk score (Thabane 2009, PMID 19568228) that prevention research in other post-event conditions lacks |
Note to the next sweep (2026-09-30). Date-restricted PubMed searches from 2026-09-02 through 2026-09-30 still returned no prevention trial in post-infectious IBS, no four-arm open-label-placebo drug trial, no head-to-head neuromodulator-versus-behavioural-therapy trial, no criteria-stratified treatment analysis, no Rome V society guideline, and no phase 3 of statin or methanogen inhibition. The eating-disorder absence is narrowed, not closed: Silva 2026 (PMID 42712718) measured risk scores, not incident diagnoses in a controlled cohort. Mortality now has a Rome II community cohort (Nakaya 2026, PMID 42702587) plus the claims-identified NIH–AARP result already on the epidemiology page; a Rome IV or Rome V mortality follow-up is still absent. The 2004 multi-tier adult suicidality survey still has no replication; the new adolescent cohort is not that study. REFIT2's abstract was used without the full text. Re-run each remaining absence before repeating it.
2026-10-14 handoff. The German 2021 and Canadian 2019 guidelines were historical search misses, now corrected in the guideline registry; look for successors to those documents. OQ-4/D5 now track the registered metabolite-selected probiotic pilot (NCT07872514), rather than treating prospective selection as absent. OQ-5 now distinguishes screening, a study-defined DSM-5-compatible classification, specialist diagnosis, and incident diet-attributable disease. REFIT2 full text remained inaccessible; its priority registry record still has no results. No question is closed. Regional transportability cannot be settled by this sweep: SciELO and LILACS could not be searched (HTTP 403, per SOURCES.md); CENTRAL was also unavailable.
2026-10-21 handoff. OQ-29 now records NCT07882953 as a Rome V eligibility protocol, not a criteria-stratified treatment trial. Date-restricted PubMed searches in this window still returned no post-infectious prevention trial, no four-arm open-label-placebo drug trial, no head-to-head neuromodulator-versus-behavioural-therapy trial, no adult multi-tier suicidality replication, no Rome V society management guideline, and no German S3 or CAG successor. Rimegepant posted a negative primary pain analysis (NCT06221111; n=24 analysed). REFIT2 full text remained blocked. SciELO returned HTTP 403; LILACS timed out; CENTRAL was not searched.
2026-10-28 handoff. OQ-13 is sharpened by a null multi-donor FMT preprint with engraftment. OQ-27 gains a Chilean Rome IV data point. A Japanese coeliac screen (1 of 583; PMID 42904019) adds a low-prevalence geography to the universal-serology disagreement on the differential page. Date-restricted PubMed searches in this window (2026-10-21 to 2026-10-28) still returned no post-infectious prevention trial, no open-label-placebo trial, no head-to-head neuromodulator-versus-behavioural-therapy trial, no Rome V management guideline, no German S3 or CAG successor, and no Rome IV or Rome V mortality follow-up. A two-author Gastroenterology item titled "Irritable bowel syndrome and suicide risk" (Shiha 2026, PMID 42906826) has no abstract and was not read. Whether it is primary data bearing on D11 and OQ-15 is unknown; check it before repeating the suicidality absence.