idea-013 · viability deck
Every slide states a condition that would have to be true, then reports where it stands using the candidate file's own claim and its own confidence tag. A deck never upgrades a tag, invents a number, or recommends anything, and there is no ask slide.
The chair: the investor, and the operator who would have to make a valve-plus-cap device sale add up to a business inside an established DMEPOS commodity category.
How to read this: every slide is a condition, not a conclusion. A bracketed claim reference points into knowledge-base/candidates/idea-013.md — the number is the claim's position in its ## Claims list, the tag is copied from it and never adjusted here. A [no claim] marker means nothing in the file speaks to this. No number appears here that is not in a claim, and there is no ask slide. See docs/deck-spec.md.
First build. At stage: verified, only 1 of 9 rubric factors carries any verified claim (Regulatory, scored 1/5) [see the candidate's Gate Check]. This deck reads accordingly thin.
Would have to be true: A specific payer, from a specific budget line, would have to pay for this device — Medicare DMEPOS, a commercial DME benefit, or the patient's own wallet — and that budget line would have to be one this device can actually bill against.
Where it stands: The only budget line named in the file at all is the one for the conventional pouching system this device would displace: HCPCS codes A4361-A4438 ("Ostomy Pouches and Supplies") are an established Medicare DMEPOS billing category [claim 7: unconfirmed] — unconfirmed for a pure tooling reason (three missing local cache files), not a finding that the category doesn't exist. No claim, established or unconfirmed, names a billing code for this candidate's own device shape — a valve-based, non-adhesive intraluminal collection interface. Formalized in this deck's write-back: no HCPCS code distinct from A4361-A4438 currently exists that would let this device be billed under Medicare DMEPOS [claim 18: unverified].
What would settle it: Fetching the three missing local caches (data/hcpcs_level_ii.csv, data/hcpcs_payment_rates.csv, data/medicare_coverage_policies.csv) per data/README.md, then re-running the hcpcs, payment, and coverage connectors against A4361-A4438 and any candidate new code; absent an existing code, a miscellaneous-code (e.g. A9999) billing assessment or a new-code petition path from a DMEPOS coding consultant.
If it's false: There is no payer at all, and the device is a cash-pay product by default — which the file has no price claim to support either (Slide 3 below).
Would have to be true: Code existence, payment amount, and coverage policy would all have to land favorably — and, distinct from all three, PDAC-style coding verification would have to confirm this specific device shape actually fits whatever code applies.
Where it stands:
What would settle it: The same connector/cache work named in Slide 1; a PDAC coding-verification submission once a candidate code is identified — no connector in this repo covers PDAC.
If it's false: The reimbursement pathway used by the conventional product this device would replace does not automatically transfer to it, and every dollar figure downstream is speculative until a code exists.
Would have to be true: Population × incidence of dissatisfaction with current pouching × price × reachable share would all have to be named, so the total is checkable rather than a single headline figure.
Where it stands:
What would settle it: A primary epidemiological dataset broken out by ostomy type and reconstruction status; a stated price or reimbursement model [claim 19: unverified]; a market-sizing pass that decomposes the aggregate $586.6M US ostomy-care-and-accessories estimate [claim 8: unconfirmed] — the market connector itself reports this class of claim unverifiable by design — into ileostomy-specific, standard-anatomy volume rather than citing the aggregate.
If it's false: TAM is a hard kill condition in the rubric (below-floor TAM gates the candidate). It is not evaluable today — not because it has been checked and passed, but because two of its four components (price, reachable share) have no claim at all.
Would have to be true: Whichever regulatory route applies — Class I exempt via an existing CFR section, Class II via De Novo, or Class II via an existing-but-unread regulation — plus prototype and bench-validation work, would have to reach commercial revenue on capital and a timeline this deck could name.
Where it stands: No cost or calendar-time claim exists anywhere in this file — the Scores table records this factor's own gap analysis rather than a guess, and this deck does not improve on it [no claim]. What is checkable is that the regulatory route itself is genuinely unresolved among three candidates: 21 CFR 876.5030 (Class I exempt, continent ileostomy catheter) whose scope for a standard ileostomy is unconfirmed [claim 1: unconfirmed]; 21 CFR 876.5900 (Class I exempt, ostomy pouch and accessories) verified to exist and match the external cap/reservoir component only, not the valve [claim 2: verified]; and 21 CFR 876.5980 (Class II, via the newly-surfaced Waters Ileostomy Valve predicate K813585), which is not exempt and would mean a materially longer, costlier path if it governs [claim 12: refuted] — see Slide 6 for what this refutation does and does not settle. Which route actually applies is exactly the feasibility deck's load-bearing condition; this deck names it as the reason the capital number cannot be estimated at all rather than restating the regulatory analysis.
What would settle it: An FDA pre-submission (Q-sub) to resolve the pathway question; a contract-engineering estimate for prototype and bench-validation cost and calendar time, which cannot be scoped until the pathway (and therefore the evidence burden) is known.
If it's false: If the Class II / De Novo route governs rather than either Class I exempt analog, this candidate's capital and timeline profile changes category entirely, and nothing in the file currently prices that difference.
Would have to be true: Something beyond the mechanical valve concept itself — a patent position, a manufacturing know-how advantage, a coding requirement a competitor can't meet — would have to keep a fast-follower from copying the design once it is on the market.
Where it stands: No claim in the file addresses defensibility directly — the Scores table leaves Moat/defensibility blank rather than guessed. The closest lever is regulatory rather than commercial, and it cuts against the candidate: this deck's own Slide 1 shows no distinguishing billing code exists yet, so there is no coding barrier to a fast-follower either. Freedom to operate is the sharper version of this question and belongs to the feasibility deck's Slide 7, referenced rather than duplicated here: a live, recently-issued US patent on a "continent ostomy valve and method of use" is a named, unread potential blocker for any mechanical valve at a stoma, including this candidate's own mechanism [claim 9: unconfirmed].
What would settle it: See the feasibility deck's FTO slide for what would settle the patent question; a manufacturing-complexity assessment of the valve+cap assembly for the know-how question — no claim in the file addresses this.
If it's false: A device this file itself frames as differentiated mainly by design choice (valve vs. absorptive, mechanical vs. sensor) rather than by IP or coding is exactly the shape a fast-follower can replicate cheaply.
Would have to be true: The field would have to be open enough, or this candidate's positioning distinct enough from existing players, for a new entrant to have room.
Where it stands: This slide is where the Verifier pass produced the file's single most consequential finding, and it cuts directly against the frontmatter's own framing. The candidate's predicate_or_analog field states plainly that "no cleared 510(k) or De Novo predicate for this exact device shape... was identified." That premise is refuted: a search for "ileostomy valve" returns a real, cleared device — WATERS ILEOSTOMY VALVE (Waters Instruments, Inc.), cleared 1982-04-09, K813585, product code KNT, Class II, current regulation 21 CFR 876.5980 [claim 12: refuted]. Two things this does not settle: whether K813585 covered a standard (non-surgically-reconstructed) ileostomy — no 510(k) summary exists for a filing this old, so its Indications for Use cannot be read from any source this repo reaches — and whether product code KNT/876.5980 remains the governing classification, since the device predates 876.5030 by seven months. Separately, two adjacent competitors are named: Ampa Medical's InterPoc, an investigational absorptive intraluminal device with no US FDA clearance found (a different mechanism, positioned deliberately against by this candidate) [claim 3: unconfirmed], and Coloplast's Heylo, a leak-notification sensor accessory — this claim's own text disclaims US clearance, but that disclaimer is itself refuted: Heylo was cleared 2025-10-06, K252140, predating this candidate's generation date and therefore findable [claim 4: refuted].
What would settle it: A human/regulatory read of what K813585's Indications for Use actually covered — the single highest-value unread document in this file; an openFDA classification/510(k) search for any other stoma-valve devices under product code KNT since 1982; a clinicaltrials.gov check on InterPoc's development stage.
If it's false: Read this inverted — if a Class II predicate already existed for a device this close in concept, the "no predicate" framing this candidate was generated under was wrong from the start, and the actual competitive/regulatory field is more crowded and more consequential than the frontmatter states.
Would have to be true: There would have to be one piece of evidence, cheap to obtain, that moves the most claims off unconfirmed at once.
Where it stands: Two candidates for cheapest-next-test, and neither has been attempted. The three HCPCS/payment/coverage claims are blocked purely on missing local cache files [claim 7: unconfirmed] — a fetch-and-rerun, not new research. Reading K813585's actual scope (Slide 6) is blocked on the absence of a posted 510(k) summary for a 1982-era filing and would need a human FOIA-style request or FDA pre-submission correspondence rather than a connector call.
What would settle it: engine/scripts/fetch_data_file.py for the three DMEPOS caches, per data/README.md; then a human pull of K813585's underlying file, or an FDA pre-submission that asks the classification question directly.
If it's false: Nothing here is false — this slide is a work order. The candidate is queued-research, and this names the two lowest-cost items on that queue.
Every condition above with nothing verified behind it — the [no claim] markers, plus the unverified claims this deck itself put on record. Read this slide first.
unverifiable by the market connector's
own design; permanently so absent a human-attached licensed report.
Structural note: two of the file's four resolved (non-unconfirmed) claims are refuted, and both cut against this candidate's own framing — the "no predicate exists" premise [claim 12: refuted] and the "no US clearance for Heylo" premise [claim 4: refuted]. A refuted claim is the most decision-relevant thing in this file, and both belong on this deck rather than buried in the candidate's Verified Composite.
If only one thing from this chair could be checked: whether this device could be sold or billed at a price the DME/insurance payer or the patient would actually pay — since no HCPCS code exists for its own shape [claim 18: unverified] and no price figure exists anywhere in the file [claim 19: unverified], the entire revenue side of this business is currently unstated rather than merely unconfirmed.
Every other viability question — capital to first dollar, moat, competitive crowding — assumes a revenue line exists to defend. None of them can be meaningfully weighed against a number that has never been written down. This is also the sharpest place this lens diverges from feasibility's load-bearing condition (which regulatory route governs): even a fast, cheap regulatory path produces a device with no named way to get paid for.
Naming it is not a recommendation, a gate, or a kill.