idea-013 · narrative
A synthesis across this candidate's claims, dossier and decks — every statement here traces to something that already exists elsewhere in the knowledge base. A narrative never invents a fact, upgrades a tag, or recommends anything.
How to read this. This is a synthesis, not a new source of truth: every claim it cites carries the tag the candidate file gives it right now, copied not paraphrased, and every line quoted from a deck is quoted verbatim. See docs/narrative-spec.md.
What this was built from. The candidate file as it stood on 2026-09-29 (29 claims), after the Verifier's second pass of the same date, and all three decks, each rebuilt 2026-09-29 and each reported current (not STALE) by check_decks.py at build time. The candidate's domains: and seeds: frontmatter are both empty, so no dossier and no seed was read, and none is cited below.
This is a soft silicone one-way valve that an adult with a mature, standard (not surgically reconstructed) ileostomy would seat at their own stoma each day, held by a low-profile collar instead of an adhesive wafer and draining into a small cap they empty by hand — a proposed alternative to the continuous-wear adhesive pouch. It does not exist as a prototype claim 13unconfirmed. Two of the premises it was generated on have since been refuted: an "ileostomy valve" was cleared by FDA (US Food and Drug Administration) in 1982 under a Class II (FDA device class II) code the file had not considered claim 12refuted, and the adjacent leak-alert product it positions against is US-cleared, not UK-only claim 4refuted. Only two of 29 claims are verified, and neither speaks to whether the device works, sells or is wanted: a pouch classification that reaches the external cap only claim 2verified and a repo-state check that no ostomy dossier exists claim 14verified. The second Verifier pass put real Medicare code descriptors and allowed amounts into the file — the codes this device could plausibly be billed under are allowed at roughly $2-$5 a unit — while holding every one of those claims at unconfirmed claim 18unconfirmed. The candidate sits at status: queued-research.
The problem. People living with an ostomy report leakage across a wide published range — 18% to 76%, about 37% of those weekly, and over 81% of those with leakage also reporting peristomal skin irritation — but the claim names no PMID (PubMed identifier), so nothing has been run against it claim 6unconfirmed. The population behind that pain is stated for any ostomy, not ileostomy: 725,000 to 1,000,000 Americans and about 100,000 new ostomy surgeries a year, from patient-organisation materials rather than a primary dataset claim 5unconfirmed. That an ileostomy-only count exists in a primary source is itself only on record, unchecked claim 26unverified.
The mechanism. Per the candidate's frontmatter, the valve sits at the stoma orifice and mucocutaneous junction rather than being advanced into the bowel, and is mechanical rather than absorptive. The Generator's reasoning is that this avoids the retained-foreign-body and obstruction risk of an intraluminal device while taking on its own failure modes — occlusion by particulate effluent, leakage past the seal, irritation at the collar — none of which has been characterised, because nothing has been built claim 13unconfirmed. The only human precedent in the file is a Silastic-tube-and-balloon "indwelling ileostomy valve device" studied in 14 patients whose continent-ileostomy nipple valve had failed — a different device and a different population, read at abstract level; the claim dates it 1980 and the PubMed record 1981 claim 11unconfirmed.
Why this shape. The file places the design between two neighbours. Ampa Medical's InterPoc is an absorbent device advanced into the small bowel and removed by a string; its description and pilot outcome have no connector, and no 510(k) (premarket notification) or De Novo (De Novo classification request) record was found for it, which is consistent with a device still in development claim 3unconfirmed. Coloplast's Heylo is a leak-alert sensor for the existing pouch; the claim's own statement that no US clearance was found is wrong — it was cleared as K252140 (an FDA 510(k) clearance number) on 2025-10-06, indicated for people with an ileostomy or colostomy aged 12 and older claim 4refuted. The file also notes that claim 4refuted carries a US jurisdiction tag on UK content, a mismatch it leaves for the Generator/Scorer.
Why now. No macro-trend seed is linked to this candidate, and no claim in the file speaks to timing. This narrative has no "why now" to report.
The regulatory story. Three routes are carried and none is confirmed. The continent-ileostomy-catheter regulation (21 CFR (Title 21 of the Code of Federal Regulations) 876.5030) exists and is Class I and exempt, but describes a device inserted to empty a surgically created ileal pouch, and whether it reaches a standard ileostomy is an FDA judgement the text does not settle claim 1unconfirmed. The ostomy pouch regulation (21 CFR 876.5900) is Class I and exempt with product codes EXB, EZQ and EZS — but reaches the external cap, not the indwelling valve claim 2verified. Against both, the WATERS ILEOSTOMY VALVE (K813585, Waters Instruments, cleared 1982-04-09) sits under product code KNT, Class II, 21 CFR 876.5980; no 510(k) summary exists for a filing that old, so what it was cleared for cannot be read claim 12refuted. Whether the Waters device is still listed as marketed is on record, unchecked claim 25unverified. The cheapest reading — that the whole device fits an existing Class I exempt classification — is also on record, unchecked claim 28unverified.
The money story. The Verifier's second pass, with the HCPCS (Healthcare Common Procedure Coding System), fee-schedule, utilization and coverage caches available for the first time, changed what the file can say about price without changing any tag. The ostomy code range the file first named is wrong at its upper bound — A4438 (a HCPCS code for a supply) is a nerve-stimulator clip — while A4416, a closed one-piece pouch, is allowed at $3.93 non-rural in the July 2026 DMEPOS (durable medical equipment, prosthetics, orthotics and supplies) release claim 7unconfirmed. No code names a device of this shape: A5081 "Stoma plug or seal, any type" and A5055 "Stoma cap" are the live candidates, A5082 and A5083 are restricted to a continent stoma the intended use excludes, A4421 is the miscellaneous fallback, and only a PDAC (Pricing, Data Analysis and Coding contractor) coding verification decides; every candidate code is allowed at about $2-$5 per unit claim 18unconfirmed. Whether a coverage policy exists that would reach the device is on record, unchecked claim 24unverified, and whether anyone would pay more than supply rates is an assumption no connector can answer claim 19unconfirmed. The one headline market figure, about $586.6 million for 2025, is an aggregator estimate the market connector marks unverifiable by design; the one primary volume in the file (A4416: 1,014,232 Medicare FFS (fee-for-service) services across 2,587 beneficiaries) carries a Verifier flag to check the unit definition before anyone multiplies it claim 8unconfirmed.
IP (intellectual property). US 11,395,757 B2, "Continent Ostomy Valve and Method of Use", now resolves: filed 2019-08-30, naive expiry 2039-08-30. Its in-force status, grant date and claims are unread, and a dates lookup is never an FTO (freedom to operate) finding claim 9unconfirmed. The adjacent leak-sensor patents are cited without numbers claim 10unconfirmed. Whether this design could itself be patented over the 2019 patent and the 1981 device is on record, unchecked claim 27unverified.
The people. Every stakeholder claim the decks sent to the Verifier came back unconfirmed on 2026-09-29, because the adoption connector is unverifiable by design: patients willing to self-insert daily claim 15unconfirmed, WOC nurses willing to teach it claim 16unconfirmed, patients preferring a valve and cap past a trial period claim 17unconfirmed, and patients able to do it safely unsupervised claim 21unconfirmed. Two newer ones are unchecked: accepting several emptyings a day and overnight claim 22unverified, and choosing a valve and cap over adding a cleared leak alert to their current pouch claim 23unverified. The movement from unverified to unconfirmed on these is not progress on desirability; it is the file recording that only a human-attached study will move them.
The build. Manufacturability, sterilisation or reprocessing, shelf life and a unit cost that fits supply-rate reimbursement have no connector claim 20unconfirmed. Whether a non-adhesive collar can keep the valve seated and sealed through daily activity and peristalsis is on record, unchecked claim 29unverified.
What the file's own record says about where it stands. The Verified Composite and Gate Check sections were written 2026-08-30, before the second Verifier pass. They still describe the HCPCS caches and patent key as missing and read claims 7 and 9 as blocked on tooling; that description is now out of date, though neither claim's tag moved. This narrative repeats no factor score from those sections and does not recompute one; that is Stage 4's and Stage 5's work, not this document's.
Two claims are verified. Neither is evidence about a user, a price, or whether the mechanism works.
knowledge-base/domains/ — a repo-state
fact, which is also why this narrative cites no dossier claim 14verified
Refuted — the two premises that did not survive.
Regulatory — unconfirmed or unchecked.
Money — unconfirmed or unchecked.
Competitors and IP — unconfirmed or unchecked.
Mechanism and build — unconfirmed or unchecked.
People — unconfirmed or unchecked.
Dossier context. None. The candidate links no dossier, and none exists for this clinical area claim 14verified.
All three decks exist, all three were rebuilt on 2026-09-29 against the candidate as it stands now, and check_decks.py reports all three current. They agree on one thing without qualification: none of their conditions has a verified claim behind it, and the only verified claims any of them cite (claims 2 and 14) are about a classification and a missing dossier. Where they part is on which unknown the others depend on — and each deck names a different one.
Desirability. Written from the chair of the ileostomy patient and the WOC nurse who would train them. Its conditions — that patients would self-insert daily, that nurses would teach it, that emptying a cap several times a day and at night beats a pouch, and that replacing the pouch beats adding a leak alert — are all unconfirmed because no connector can answer them, or freshly on record and unchecked claim 15unconfirmed claim 16unconfirmed claim 22unverified claim 23unverified. It treats the refuted Heylo premise as the reason the comparison a patient would actually make has changed claim 4refuted. Its view is that every other slide assumes a patient who has already agreed to handle a valve at their stoma every day.
"whether adults with a mature, standard ileostomy would be willing to learn and keep performing daily self-insertion and self-removal of a valve at their own stoma, in exchange for less pouch bulk and less adhesive on their skin" — the load-bearing condition of the desirability deck (built 2026-09-29).
Viability. Written from the investor and operator. It agrees with desirability that the people questions are the ones no connector will settle, but its centre of gravity is the number the second Verifier pass produced: every code this device could plausibly be billed under is allowed at about $2-$5 a unit claim 18unconfirmed. Population, route and moat, on this deck's reading, matter only if someone pays more than that claim 19unconfirmed. It also carries the two refutations as competitive facts — a 1982 cleared ileostomy valve whose market status is unchecked claim 12refuted claim 25unverified, and a US-cleared leak alert claim 4refuted. It is the one deck that sets a ceiling the feasibility deck has to build under.
"whether ileostomy patients or their payer would pay a price for this device high enough to sustain a specialty product, above the roughly $2-$5 per unit Medicare allows for the ostomy codes it could be billed under" — the load-bearing condition of the viability deck (built 2026-09-29).
Feasibility. Written from engineering. It disagrees with the other two about order: the route decides how much evidence must be built, but the seal decides whether there is a device to build evidence for. If a collar without adhesive cannot hold the valve sealed, the device reintroduces the leakage it was meant to remove, and the desirability questions are being asked about the wrong product claim 29unverified claim 13unconfirmed. Here the lenses pull against each other directly: feasibility needs a unit cost that fits under supply-rate reimbursement claim 20unconfirmed, while viability's condition is that the device earns more than supply rates — the same $2-$5 figure is a ceiling for one deck and a floor to clear for the other. It also shares with desirability a blank neither can fill yet: the reservoir capacity and emptying interval that would turn "several times a day" claim 22unverified into a number are in no claim.
"whether a low-profile external retention collar, without an adhesive wafer, can hold a soft silicone one-way valve seated at the stoma orifice of a mature ileostomy through normal daily activity and peristalsis without dislodgement and without effluent leaking past the valve seal" — the load-bearing condition of the feasibility deck (built 2026-09-29).
Is there a version of this device — one whose non-adhesive collar actually holds the valve sealed at a standard ileostomy — that adults with an ileostomy would insert every day and that a patient or payer would pay for above the roughly $2-$5 per unit Medicare allows for a stoma plug, seal or cap?
Each deck's condition assumes the other two. The seal claim 29unverified is the premise of the patient's willingness claim 15unconfirmed, and the patient's willingness is the premise of anyone paying a premium over supply rates claim 19unconfirmed claim 18unconfirmed — while the regulatory route that sizes the cost of finding out is itself unsettled by a refuted premise claim 12refuted. None of the three halves has a verified claim behind it, and none of the stakeholder halves can be moved by a further Verifier pass; only bench work and conversations with patients, nurses and payers reach them.
Naming this question is not a recommendation, a gate, or a kill. Which part of it to look at first, if any, is a human's call.
_shared/plain-language.md