idea-009 · feasibility 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: engineering. The people who have to make a short-stretch textile assembly whose printed elongation windows mean the same thing on a 22 cm calf and a 48 cm calf, on strap one and strap six, on day one and after forty washes — and who have to produce the evidence a reviewer would ask for on a route nobody has yet confirmed applies.
How to read this: every slide is a condition, not a conclusion. A bracketed reference points into knowledge-base/candidates/idea-009.md — the number is the claim's position in its ## Claims list, the tag is copied from it and never adjusted here. [no claim] means nothing in the file speaks to the condition. See docs/deck-spec.md.
One thing to hold while reading: this is the only lens with verified claims behind it — all five of them — and every one of them is about the regulation, not about this device. The generic type's class and exemption are verified; this device's membership in that generic type is not.
Would have to be true: The wrap would have to produce a distal-to-proximal decreasing pressure profile on a below-knee limb, inside the 30-50 mmHg band the A6583 descriptor names, when applied by the patient to indicator-aligned strap tension — not when applied by a fitter on a bench.
Where it stands: The target band is taken from a HCPCS descriptor that has never been read against primary CMS text — data/hcpcs_level_ii.csv is absent, so the "30-50 mmHg, below knee" specification the product was deliberately sized to is unverified in both its wording and its numbers [claim 8: unconfirmed]. No claim in the file states a target gradient ratio, an ankle-to-calf pressure step, or a tolerance; the mechanism sentence in the frontmatter describes an intent, not a spec. The requirement as an engineering proposition went on record only with this deck [claim 36: unverified].
What would settle it: One human places the CMS HCPCS Level II quarterly file per data/README.md and connectors hcpcs A6583 --expect "gradient compression wrap with adjustable straps" re-runs unattended, fixing the band and the anatomical scope. The gradient target itself is a design decision to be taken against the recognised consensus documents on compression in lymphedema — a Tier 3 clinical/engineering act, not a lookup.
If it's false: If the descriptor does not say what the file assumes, the product is built to the wrong specification and the reimbursement anchor and the engineering spec fail together — Factor 2 and Factor 1 at once, since the below-knee scope was chosen for the code.
Would have to be true: Sustained gradient compression would have to work as maintenance therapy at the pressures this device can actually deliver, and the inelastic short-stretch construction would have to hold its applied tension across a wear period rather than relaxing into an ineffective wrap by hour six.
Where it stands: The therapeutic mechanism is the least contested thing in the candidate and is still not verified: sustained gradient compression is stated to be established maintenance therapy within complete decongestive therapy, so mechanism risk is low and the material risk is application risk — but no Stage 3 connector answers a mechanism or clinical-risk claim, and the tempting substitutes (literature, trials, literature-detail, trial-detail, recalls) are Stage 0 scans returning a ScanResult that carries no confidence tag and cannot acquire one, so running them produces reading material, never verification [claim 21: unconfirmed]. Tension retention over a wear period is not addressed anywhere in the file and went on record only with this deck [claim 37: unverified]. Note the scope choice that helps here: excluding the intensive decongestive phase removes the therapist-applied multilayer failure mode and leaves an ergonomic one.
What would settle it: Tier 3 clinical review against the consensus documents for the therapy half. The tension-retention half is a bench question and belongs on Slide 5: an instrumented leg form, interface-pressure sensors, and a timed hold under simulated ambulation and limb-volume change.
If it's false: Factor 6 (mechanism and clinical risk), currently blank for want of any verified claim, becomes the problem rather than the reassurance. A wrap that relaxes is not a weaker version of the product — it is an inelastic wrap behaving like a bad elastic one.
Would have to be true: FDA would have to read an inelastic strapped wrap into 21 CFR 880.5780(b) and § 880.9 would have to not fire on an indication of lay self-application — both, not either — for the build to be registration, listing and a records and complaints file rather than a premarket notification with a predicate.
Where it stands: This is where the file's only verified evidence sits, and it is conditional on its face. Verified off primary eCFR text: § 880.5780 exists and is titled "Medical support stocking" [claim 1: verified]; paragraph (b) is Class I, exempt from the premarket notification procedures in subpart E of part 807, and additionally exempt from part 820 CGMP except records and complaint files under § 820.35 [claim 2: verified]; paragraph (a) is Class II and also premarket-exempt, and does not carry the part 820 exemption [claim 3: verified]; product codes FQL and DWL both map to 880.5780 [claim 4: verified], [claim 5: verified].
None of that has yet been shown to apply to this device, and the same verified text is what raises the doubt. Both paragraphs identify a device "constructed of elastic material and designed to apply controlled pressure to the leg", while this candidate's mechanism is expressly inelastic short-stretch — its whole therapeutic premise. And § 880.9 requires a premarket notification anyway where a device "is intended for lay use where the former intended use was by health care professionals only", or operates using a "different fundamental scientific technology", against an intended_use: of self-application without a fitter. Whether FDA reads this device into the generic type is an FDA judgment no connector answers, and it is unconfirmed — not refuted, and not settled [claim 6: unconfirmed]. The Verified Composite holds Factor 1 at 2/5 for exactly this reason. A third fork sits alongside: a pressure-measurement claim on the indicators — the tempting one, since they are the differentiator — could pull the product off the exempt route entirely, which is why the design intent is that the indicator is a strap self-application aid and never a patient measurement [claim 6: unconfirmed]. Whether human-factors validation alone can carry a lay-use indication on this route went on record only with this deck [claim 38: unverified]. What the exemption does not excuse — establishment registration and device listing — is unread, because the regulation 807.20 probe hit the eCFR outage [claim 22: unconfirmed].
What would settle it: A 513(g) Request for Information to FDA, or Tier 3 regulatory-counsel review of the draft labelling — the two routes the file names, neither an agent action, and neither reachable by any connector now or after the caches land. One cheap adjacent step: browse the FDA product-code classification database by regulation number 880.5780 and by panel HO by hand, since the --device-name probe form is known to miss records that exist. Re-run regulation 807.20 for free once eCFR recovers.
If it's false: This is not a slower build, it is a different one — a premarket notification, a predicate hunt in a database where exempt competitors structurally do not appear [claim 7: unconfirmed], performance and human-factors evidence to a reviewer's standard, and full QMSR rather than records and complaints only. The file states the consequence for Factor 7 in the same terms: a different business, not a worse row [claim 22: unconfirmed].
Would have to be true: The printed elongation window would have to align at the same strap tension across the entire labelled limb-circumference range, across manufacturing tolerance on the strap web, and across each of the several straps — and the pressure that tension produces would have to be the same on a large calf as on a small one, which is not what geometry does for free.
Where it stands: This is the product's only differentiator and it is unevidenced in every direction the file looks. As an asset it is asserted, not demonstrated — the IP half of the moat cannot resolve while no patent search has been performed [claim 20: unconfirmed]. As an instrument it is unvalidated: no claim states a tolerance, a calibration method, or a circumference range [claim 39: unverified]. As a label it is a live regulatory question, since a measurement claim about the patient may not be a "stocking" at all [claim 6: unconfirmed]. And as a user interface it depends on a patient with hand arthritis reading it correctly, which the file records as having zero evidence on either side [claim 23: unconfirmed]. Laplace's law is the physics problem underneath — the same tension over a smaller radius is a higher pressure — and no claim in this file addresses it [no claim].
What would settle it: A bench calibration study on an instrumented leg form across the labelled circumference range with interface-pressure sensors at ankle, gaiter and calf, repeated across production strap lots. This is the first thing to build a rig for and it costs nothing that a clinical study costs.
If it's false: The differentiator becomes a decoration: the straps still work like every incumbent's straps, the moat argument in the viability deck loses its only mechanism [claim 20: unconfirmed], and the desirability crux gets harder rather than easier, because a patient with no reliable feedback is applying an unknown pressure to their own leg.
Would have to be true: Before any patient wears this, a bench rig would have to show two things: that indicator-aligned straps produce the intended gradient across the labelled circumference range, and that the tension survives a wear period — with a documented arterial-insufficiency contraindication and ABI screening statement already drafted into the labelling.
Where it stands: No bench evidence exists in the file and the Diligence Package section is empty [no claim] — that section is Stage 7 and is filled by a human, not by an agent. The two bench propositions are on record only through this deck [claim 36: unverified], [claim 37: unverified], [claim 39: unverified]. The labelling half is stated but unverified: the arterial-insufficiency contraindication and ABI screening statement are described as mandatory labelling content rather than optional, and the file records that the PAD hazard is worse on this candidate than on fitter-applied incumbents precisely because the intended use is self-application without a fitter — the same fact that makes § 880.9 live on Slide 3 [claim 21: unconfirmed], [claim 6: unconfirmed].
What would settle it: Build the rig and run it — interface pressure at ankle, gaiter and calf on a range of leg forms, indicator-aligned by an untrained applier, held under simulated ambulation. Then Tier 3 clinical and regulatory labelling review of the contraindication and ABI language against the recognised consensus documents. Nothing downstream of this slide is an agent's to greenlight, and no agent in this repo authorises a human study.
If it's false: There is nothing to show a fitter, nothing to show a reviewer if the exempt route does not hold, and nothing to put in front of the patients the desirability deck's load-bearing condition needs — the donning sessions cannot measure applied pressure if the indicator has no calibrated meaning yet.
Would have to be true: The wrap would have to be manufacturable from short-stretch textile at DMEPOS unit economics, and it would have to keep both its compression performance and its printed calibration through the labelled service life — including repeated home laundering by the patient — while being donned on a bathroom floor by someone with limited hip flexion who may not be able to reach their own foot.
Where it stands: The manufacturing side is described only in shape: on the exempt route the FDA cost is registration, listing and QMS compliance rather than a submission, so first revenue is gated by textile tooling and fitter training rather than FDA review — stated deliberately without a dollar or month figure, and conditional on the exempt route holding [claim 22: unconfirmed]. The QMS burden itself is the one favourable thing here that is verified, and only if the device is inside paragraph (b): that paragraph exempts the device from part 820 CGMP except records and complaint files under § 820.35 [claim 2: verified] — and note that paragraph (a) does not, so a single sentence of venous marketing copy costs the QMS exemption as well as Class I [claim 3: verified]. Durability and laundering are unaddressed in the file and went on record only with this deck [claim 40: unverified]. The use environment is the desirability crux seen from the engineering chair, with zero evidence on either side [claim 23: unconfirmed].
What would settle it: Wash-and-wear cycling against the labelled service life with pressure and indicator-alignment re-measured at intervals; hook-and-loop peel and shear retention after cycling; a real textile quote for the tooling estimate the file correctly refuses to invent. Human-factors observation in the actual environment covers the last part and is the same session as the desirability deck's Slide 6.
If it's false: A wrap whose indicators drift after washing is worse than one with no indicators, because it is confidently wrong. Factor 7 absorbs the tooling half, and the records-and-complaints scope of the QMS obligation stops being an advantage the moment paragraph (b) stops applying.
Would have to be true: A strap-mounted mechanical tension or elongation indicator on a limb compression wrap would have to be practisable without infringing an unexpired claim held by one of the four incumbents — and if a design-around were needed, it would have to not be the product, since the indicator is the differentiator.
Where it stands: Unexamined, not clear — and the file is emphatic that the difference matters. No patent search was performed, no patent number is named anywhere to look up, and connectors patent takes a patent_id it does not have; the Verifier did not invent one [claim 19: unconfirmed]. Separately, PATENTSVIEW_API_KEY is unset, so even a named number would return unconfirmed for an infrastructure reason that could be misread as a clean field. An absence of found patents is not freedom to operate. The concentration risk is named and unchanged: the only differentiator sits in exactly the narrow art four long-established incumbents — medi GmbH, Essity/BSN, Sigvaris, Solaris — would occupy [claim 20: unconfirmed]. The proposition an FTO search would actually test went on record only with this deck [claim 41: unverified].
What would settle it: A human or Scout assignee sweep in USPTO Patents Public Search and Espacenet against those four assignees, with CPC class A61F13/08 and keyword sets on "adjustable compression wrap", "strap tension indicator", "elongation indicator" and "sub-bandage pressure" — discovery first, because no connector here performs it and the API key alone buys nothing. Each number found then goes through connectors patent US<number> for family, legal status and expiry. The definitive answer is a Tier 3 legal FTO opinion at Stage 7; nothing an agent produces is one.
If it's false: FTO is the rubric's only automatic kill regardless of other scores, and the Gate Check records the condition as cannot be evaluated with zero patents searched. A blocking claim on the indicator does not slow the build — it removes the reason to build.
Every condition above with nothing verified behind it. This deck is the only one of the three that cites verified claims at all, and the boundary is precise: what is verified is what the regulation says — its text, its two paragraphs' classes and exemptions, and two product-code records [claim 1: verified], [claim 2: verified], [claim 3: verified], [claim 4: verified], [claim 5: verified]. What is not verified is that any of it applies to this device.
regulation 807.20; free once eCFR recovers.
unverifiable by connector.
If only one thing from this chair could be checked: whether FDA reads an inelastic adjustable-strap wrap, indicated for lay self-application without a fitter, into 21 CFR 880.5780(b) — or whether § 880.9 requires a premarket notification anyway [claim 6: unconfirmed].
This one rather than the others because it is the only condition that changes what gets built rather than how well. The verified evidence is real and favourable about the regulation: paragraph (b) is Class I, exempt from premarket notification under subpart E, and exempt from part 820 CGMP except records and complaints [claim 2: verified]. But the same verified text says "constructed of elastic material" and this device is short-stretch by design, and § 880.9 names lay use where the former intended use was professional-only as a case where notification is still required — against an indication that says "without a fitter". Both readings are FDA judgments, both are unconfirmed, and neither is reachable by any connector in this repo, which distinguishes this from every other gap on the deck: the CMS blanks are a missing file, the eCFR blanks are an outage, the FTO blank is an unrun search. This one is a question only FDA or counsel can answer, it sits upstream of the QMS scope on Slide 6, the evidence burden on Slide 5 and the whole capital argument on the viability deck, and the two routes to it — a 513(g) Request for Information or Tier 3 regulatory-counsel review of the draft labelling — are both human acts.
Naming it is not a recommendation, a gate, or a kill.