Home Stress-Incontinence Stimulator

• The statement: claim 9verified — that Medicare NCD 230.8 covers pelvic floor electrical stimulation with a non-implantable stimulator for stress and/or urge urinary incontinence in cognitively intact patients who failed a documented four-week trial of pelvic muscle exercise training. It came back refuted on 2026-08-30, which is what opened this review; it reads claim 9verified now, and the closure below is why. The tag is named here in words rather than cited in the [claim N: tag] form, because that form reads the tag the candidate file carries today — citing a historical tag through it would make this file assert something the candidate file contradicts, which is the one thing a loop record may never do.
• What the check found: the local CMS Medicare Coverage Database cache, now populated at 646 rows, contains no policy naming E0740. The tag moved from unconfirmed (missing-cache) to refuted on 2026-08-30 when the cache landed. The Verifier's own note says to read the caveat before treating it as a clean kill, and the caveat is the point: the check was a code lookup, and NCD 230.8 is a real national coverage determination that may simply not reach us through E0740.
• Who depended on it: register row A3 Medicare NCD 230.8 covers pelvic floor… — Market Access, Commercial, and Regulatory. The regulatory dependency is the non-obvious one and the expensive one: the intended use was drafted to mirror NCD 230.8's eligibility language, which is documented in the candidate's own indication-scope note.
• What it costs: the narrow indication is currently buying nothing. It was traded for a coverage argument that would be a restatement of existing national policy; if the policy does not reach this code, the indication has been narrowed — shrinking the population in claim 19unconfirmed — in exchange for an evidence saving alone, and Regulatory says there is no further class benefit in narrowing because 876.5320 is already the floor.
• Outcome: commission
• Because: the distinction between unconfirmed and refuted is doing real work here and it is worth one retrieval to resolve. The refutation is a statement about a cache and a code binding, not about whether NCD 230.8 exists or what it says. Commissioning against a refuted claim is usually re-litigating a finding, and this is the exception the protocol's asymmetry allows for: the finding is about our lookup, not about the world. REQ-0001 Does the NCD 230.8 eligibility language… is that commission. If the full text comes back and does not reach this indication, this review reopens and the honest outcome becomes pivot, not absorb.
• **Closed 2026-09-14 — REQ-0001 Does the NCD 230.8 eligibility language… answered, and it answered for the commission rather than against it.** The ledger carries REQ-0001 Does the NCD 230.8 eligibility language… as answered, answered_by: claim 9verified. The retrieval this review bought turned out to be a connector run, not a human errand: a new ncd subcommand addresses CMS Pub. 100-3, NCD Manual ch. 1 pt. 4, by section number, and §230.8 (Rev. 13808, issued 05-29-26) states this register row's statement close to verbatim — non-implantable pelvic floor electrical stimulation is "covered for the treatment of stress and/or urge urinary incontinence in cognitively intact patients who have failed a documented trial of pelvic muscle exercise (PME) training", with the four weeks carried in the manual's definition of a failed trial rather than in the coverage sentence. claim 9verified as of 2026-09-14. **The reopen condition this review wrote for itself did not fire**: the full text came back and it does reach this indication as written, so the outcome stays commission and stays discharged, A3 Medicare NCD 230.8 covers pelvic floor… stays live, and no pivot follows. The narrow indication is buying, at the policy level, exactly what it was traded for.
• **What the closure does not settle — the request's own note is explicit, and it is repeated here so the register is not read as one row more settled than it is.** The NCD names no HCPCS code. The coverage E0740 code-linkage finding that produced the original refutation is not withdrawn by this and is not re-tagged anywhere; it survives as a residual inside a now-verified claim. So three things stay open behind A3 Medicare NCD 230.8 covers pelvic floor…: (i) that this device codes to E0740 and thereby to the NCD — that is claim 45unconfirmed, the DME MAC (durable medical equipment Medicare administrative contractor)/PDAC coding-verification guidance, which no connector in this repo reaches; (ii) that a MAC (Medicare administrative contractor) pays a claim, coverage text being an eligibility statement and never a payment decision; and (iii) anything about app-linked, EMG-gated or RTM-billed variants, §230.8 being a 2006 national policy about the device class. A verified coverage policy and a covered device are two different facts and only the first one is in hand.
• What this review got right and wrong, kept because the next one inherits it. Right: the protocol's asymmetry. The refuted was a finding about our lookup — a code-keyed query over a cache with 356 blank codes columns out of 646 — rather than about the world, which is the narrow exception that lets a commission run against a refutation, and it held. Wrong: this review priced the retrieval as a human errand because CMS returned HTTP 403 on search.aspx?ncdid=232. That URL resolves to NCD 270.2, a different policy entirely. **The transferable part: "only a human can get this" is usually a claim about the address we tried, not about the source** — and the address the claim already cited, the section number, was the one that worked.

• The statement: claim 16refuted and claim 17refuted — the two Medicare volume legs at the base of the sizing chain.
• What the check found: against the CMS utilization cache for 2024, CPT (Current Procedural Terminology) 90912 ran 28,093 allowed services against a claimed 200,000 (overstated by 612%), and E0740 ran 4,130 allowed units against a claimed 10,000 (overstated by 142%). Both outside the 25% tolerance. Both are Medicare fee-for-service only, which the Verifier flags as a **floor on US volume, not a market size** — the figures exclude Medicare Advantage, commercial, Medicaid, VA and cash-pay.
• Who depended on it: register rows A4 TAM component 1 and A5 TAM component 2 — Commercial, and through the market model every number downstream of it.
• What it costs: the sizing chain's base is an order of magnitude out, so every band built on it is wrong by that factor. But the finding is narrower than it looks and the narrowness is the whole opportunity: what was refuted is a claim about *Medicare fee-for-service volume*, and what was assumed was that Medicare FFS volume was the right base to size from.
• Outcome: pivot
• Because: the refutation is a real finding about the world and commissioning more work against it would be re-litigating it — 28,093 is 28,093. What is wrong is not the number but which population the opportunity was resting on. Market Access had independently reached the same door from the coverage side: if the national policy does not hold and the Medicare volumes are a floor rather than a market, both chairs are pointing at the commercially-insured population. That is P1 commercially-insured first. The new dependency it creates, claim 32unconfirmed, is registered as A7 A material share of the addressable… and reviewed in IR-3 claim 32 is unconfirmed and P1 now rests on… rather than assumed.
• **Correction, 2026-09-19 — this review states the E0740 figure in the wrong unit, and the error is six-fold.** "E0740 ran 4,130 allowed units" is the connector's own wording, carried in from claim 17refuted and repeated here without being looked at. claim 70verified reads the fuller CMS DME release — "Medicare Durable Medical Equipment, Devices & Supplies — by Geography and Service", same code, same data year — as 4,130 supplier services across 689 beneficiaries, 2,974 claims, 28 suppliers and 553 referring providers. claim 72verified says E0740 is flagged rentable in that release, which is why services exceed beneficiaries sixfold, and that a service count must not be read as a count of devices or of patients. **The number of American women who obtained this class of device through Medicare fee-for-service in 2024 is 689.** Register row A5 TAM component 2 carried the same error and is corrected above; claim 67unconfirmed carries it in its anchor and is now A18 The number of commercially insured women…, with IR-10 claim 67 is unconfirmed against it. The Commercial chair found all three on 2026-09-19 and could fix none of them, because a chair may not write the register — which is the prohibition working, not failing.
• What the correction does to this review, said plainly so it is not read as a reopening. The outcome stays pivot, P1 commercially-insured first is not disturbed, pivot: is not bumped and no chair is shown as stale. The correction makes this review's cause larger rather than smaller: the Medicare fee-for-service floor is six times thinner than the number the pivot was argued over, which strengthens "this is a floor, not a market" instead of weakening it. What it does change is a reading that has already been quoted downstream — the business case's "implied Medicare-FFS device pool, 4,130 × $74.52 ≈ $307,700/year" multiplies a service count on a rentable code by a purchase-style national rate, which is two different units multiplied together, and it should not be quoted again in that form.
• **And the inference this correction must not be allowed to smuggle in, because a unit fix is exactly the moment somebody re-derives it.** $57.89, the realized average allowed amount inside claim 71verified, is not evidence that claim 10verified's $74.52 "is not being realized". One is an average per service across a rental-and-purchase mix for data year 2024; the other is a purchase-style national unadjusted rate labelled **schedule year 2026**. Not commensurable in kind or in time. That error was made in this file's own prose and corrected in four places, and the whole of what the per-beneficiary arithmetic adds is bound as claim 87unconfirmed (A14 The average Medicare fee-for-service…) so a Verifier owns it and this record does not.

• The statement: claim 32unconfirmed — that a material share of the addressable stress-urinary-incontinence population is commercially insured rather than Medicare.
• What the check found: nothing yet. It was surfaced as an assumption and has never been attempted against a source.
• Who depended on it: register row A7 A material share of the addressable… — Commercial and Market Access. Before P1 commercially-insured first nobody did; the pivot is what made it load-bearing, which is exactly the moment to register it.
• What it costs: if it is false, P1 commercially-insured first routes around Medicare and in doing so routes around the market, and the pivot is worse than the shape it replaced.
• Outcome: commission
• Because: unconfirmed here means unasked, not unanswerable — prevalence by payer mix for stress urinary incontinence in women 45+ is a researchable number. REQ-0003 Payer mix for diagnosed stress urinary… is the commission. Recording this as an impact review rather than as a footnote to the pivot is deliberate: a pivot that creates a new load-bearing assumption and does not review it has moved the risk rather than reduced it.

• The statement: claim 21unconfirmed — the FTO (freedom to operate) claim.
• What the check found: no connector call was possible and none was made. The patent primitive takes a specific patent_id and cannot run a landscape search; the claim names no patent number and no assignee, so there was no object to pass it. The claim is unsearched, not merely unverified.
• Who depended on it: register row A6 FTO — no live, enforceable US patent claims… — Legal & IP, which has abstained rather than hold a position on it.
• What it costs: nothing yet, and that is the trap. An unsearched FTO claim costs nothing right up until the point where it costs the programme, and the tag it carries is the same tag a searched-and-inconclusive claim would carry.
• Outcome: commission
• Because: this is a missing capability, not a missing fact — the repo has no landscape search. REQ-0002 A patent landscape over closed-loop… asks Research for a CPC-class landscape, which is the shape of work that can answer it. Absorbing it would mean deciding FTO does not matter to a device whose differentiator is its mechanism.

• The statement: claim 6unconfirmed — that a device with this intended use is eligible for 510(k) clearance and requires neither a De Novo nor a PMA.
• What the check found: openFDA publishes no De Novo endpoint and this repo holds no cached copy of FDA's De Novo listing, so the Verifier could not attempt it and explicitly declined to substitute a web search. What is established — claim 2verified carries submission_type_id 1, and several KPI clearances exist — establishes that the product code is a 510(k) code. It does not establish that this device, with this intended use, is substantially equivalent to any of them.
• Who depended on it: register row A2 A device with this intended use is eligible… — Regulatory, which holds a conditional verdict resting on exactly this.
• What it costs: a De Novo route adds 12-18 months and a new evidence burden ahead of any revenue, which moves Factor 7 as much as Factor 1. It does not change P1 commercially-insured first — the payer-route pivot is orthogonal to the clearance route — which is why this review is separate from IR-2 claims 16 and 17 came back refuted rather than folded into it.
• Outcome: commission
• Because: this is a missing cache, not a finding — the same shape as IR-4 claim 21 is unconfirmed and the opposite of IR-2 claims 16 and 17 came back refuted. The De Novo listing is a file FDA publishes and this repo does not yet fetch, so the honest work is to fetch it; Regulatory's [C1] names the alternative (a pre-submission meeting response) which settles the substantial-equivalence half that no listing can answer. No request is raised against Research for the first half, because engine/scripts/fetch_data_file.py is the tool for a missing cache and that is a job, not a question.

• The statement: claim 14unconfirmed — that pelvic floor muscle status qualifies as "musculoskeletal system status" for the purposes of billing CPT 98977, which is what puts this device on the RTM path at all.
• What the check found: nothing determinative. claim 13unconfirmed establishes that 98977 exists and describes device supply for remote therapeutic monitoring of the musculoskeletal system; whether pelvic floor sits inside that category is a coding question nobody has retrieved an answer to.
• Who depended on it: register row A8 Pelvic floor muscle status qualifies as… — Market Access, via its [C2]. It arrived in the register through that binding rather than through the front end, which is the mechanism working as designed.
• What it costs: the money model has two legs, a device sale and a recurring RTM stream. If this is false the second leg closes and the opportunity becomes a one-time sale of a $350-$700 device — which is a different business, not a smaller one. Note that this cost is independent of P1 commercially-insured first: it bites the same way under Medicare and under commercial payers.
• Outcome: commission
• Because: unasked rather than unanswerable, and cheap to ask — CPT Assistant guidance or any payer policy naming pelvic floor within the RTM musculoskeletal category settles it. Market Access's [C2] already names that source. Folded into REQ-0001 Does the NCD 230.8 eligibility language…'s scope rather than raised separately: it goes to the same department, against the same coding database, in the same sitting.
• **Correction, 2026-09-14 — this commission has no vehicle and this review wrote it that way.** REQ-0001 Does the NCD 230.8 eligibility language… is now answered, and its question was "Does the NCD 230.8 eligibility language reach this indication as written?" It never carried the RTM question at all, so folding IR-6 claim 14 is unconfirmed and the… into its scope folded it into a request that could not have answered it, and the request closed without touching claim 14unconfirmed. The outcome stays commission because it was the right outcome; what was wrong was the vehicle, and it is named here rather than quietly re-pointed. claim 14unconfirmed is unattempted, A8 Pelvic floor muscle status qualifies as… is live, and Market Access's [C2] still binds it — the chair re-stated on 2026-09-14 and carried [C2] forward unchanged, adding that §30.1.1's home-biofeedback exclusion is "an adjacent CMS position on the same organ and the same home setting, which is a reason to check this rather than assume it." REQ-0023 Does pelvic floor muscle status fall inside… is what now carries this commission — front end → research, raised 2026-09-14, settling "pelvic floor muscle status qualifies as musculoskeletal system status for the purposes of billing RTM device-supply code 98977." It is scoped to CMS's own published commentary (the CY2022 PFS (physician fee schedule) final rule that created the RTM family, the later RTM provisions, and the CMS RTM FAQ) and explicitly not to AMA CPT descriptor text, which is AMA-licensed and permanently unverifiable here per claim 13unconfirmed — scoping it that way is what makes it answerable at all rather than a second request that cannot be filled. It is raised by the front end and not by Market Access because the front end is what stranded it. The lesson is narrow and worth keeping: **folding a commission into another review's request is only honest if the request's question carries it**, and REQ-0001 Does the NCD 230.8 eligibility language…'s never did.

• The statement: claim 55unconfirmed — that this device, an intravaginal pelvic floor electrical stimulator with integrated surface-EMG and a paired app, obtains DME MAC/PDAC coding verification to HCPCS E0740, so that NCD 230.8's coverage reaches it *as billed*.
• What the check found: the Verifier read it as a compound claim and checked each half cold rather than re-running what this file already holds. E0740 exists (hcpcs E0740 → verified) and NCD 230.8 covers non-implantable pelvic floor electrical stimulators on the stated criteria (ncd 230.8 → verified). Neither establishes the join. **The unresolved half is not a database fact at all** — whether this device obtains PDAC coding verification is a future regulatory-administrative decision, and no connector in this repo reaches PDAC coding-verification determinations. The two other PDAC claims in the file are unconfirmed on the identical structural gap.
• Who depended on it: register row A9 This device — Market Access and Commercial. It arrived through Market Access's [C4], and it is the residual IR-1 claim 9 came back refuted's closure explicitly left open: a verified coverage policy and a covered device are two different facts, and IR-1 claim 9 came back refuted only ever had the first.
• What it costs: the narrow indication was traded for a coverage argument. IR-1 claim 9 came back refuted bought back the policy half; this is the half that says the policy reaches this product. If the device codes to a miscellaneous code instead of E0740, the DME benefit does not reach it, the coverage argument stops at the class boundary, and the indication is again buying an evidence saving alone. Market Access has already named this as the one finding that would take it past blocks to a pivot — *because it would be about the world and not about our lookup.*
• Outcome: commission
• Because: unconfirmed here is a gap in the check, and the gap has two different shapes that must not be collapsed. The retrievable shape — what the PDAC code-review article and any DME MAC LCD (local coverage determination) actually say about a stimulator with integrated surface-EMG and app connectivity — is a document pull, and REQ-0019 Retrieve is already that commission, raised by the chair on 2026-09-14 and open. This review adopts it rather than raising a second request against the same documents. The unretrievable shape is the future event, and no request can settle it: PDAC verification of a device that does not exist yet is not a fact anyone can look up, and a commission that pretended otherwise would be a request nobody could fill. **So what REQ-0019 Retrieve can return is the criteria and the precedent, not the answer** — and the honest reading of a clean REQ-0019 Retrieve is that the claim becomes better-founded, never verified. Absorbing was available and is wrong: A9 This device has two chairs behind it and is the load-bearing half of the argument the indication was bought for.

• The statement: claim 56unconfirmed — that at least two named commercial payer medical policies cover non-implanted pelvic floor electrical stimulation for stress urinary incontinence, on eligibility criteria a device with this intended use would meet.
• What the check found: nothing, and the Verifier was precise about why. No named commercial payer was checked, or could be: coverage reads data/medicare_coverage_policies.csv, the CMS NCD/LCD index, and ncd reads CMS Pub. 100-3. Both are Medicare-only instruments, and neither touches commercial payer medical policy. This is unreachable in this repo, not unanswered in it.
• Who depended on it: register row A10 At least two named commercial payer medical… — Market Access and Commercial. It is P1 commercially-insured first's launch route. The chair reported the whole area as entirely unsearched on 2026-09-14, and the Stage 3 pass confirms that reading structurally rather than taking it on the chair's word.
• What it costs: P1 commercially-insured first moved this opportunity from Medicare fee-for-service to the commercially-insured population. If commercial payers class non-implanted pelvic floor stimulation as investigational, the pivot delivers a reachable population and no payment mechanism — a market without a payer, which is the failure mode P1 commercially-insured first was supposed to escape, arrived at from the other side. Note this is the only register row whose failure would reopen the pivot itself rather than a number under it.
• Outcome: commission
• Because: unconfirmed from a connector that was never going to reach the source is the clearest case the protocol's asymmetry names — a gap in the check, not a finding about the world. Three national payers publish their medical policies by policy number, so this is a document pull with a known address, and REQ-0020 Retrieve is already that commission, open since 2026-09-14 against Research. This review adopts it. What it adds is the reading that a negative return settles the claim just as well: three payers calling the class investigational is a finding about the world, and the honest outcome next time would be pivot, not another commission.
• Note, 2026-09-19 — this review's heading is now out of date, and A10 At least two named commercial payer medical… is corrected above. claim 74verified and claim 75verified put Aetna CPB 0223 into this file: intravaginal pelvic floor electrical stimulation found medically necessary for stress, urgency or mixed urinary incontinence, on indication and sex alone, with no prior failed-exercise requirement, no minimum duration, no NCD cross-reference and no home-versus-office distinction. P1 commercially-insured first's launch route has one payer behind it, not none, and on criteria looser than the national Medicare policy the intended use was drafted to mirror. The outcome does not move and neither does the claim. claim 56unconfirmed asks for two named policies; claim 76verified records that the second payer checked removed the therapy from Cigna MCP 0160 in January 2025 and that E0740 appears nowhere in the current version, so that document establishes no position in either direction. One of two. REQ-0020 Retrieve is still the commission and the third national payer is still the cheapest thing that could move it. What changed is the shape of the cost if this fails, and A10 At least two named commercial payer medical… now carries it: not "a market with no payer" but a single-payer route, where the one policy covering the class is also the only evidence that anyone covers it. A second gap opened underneath the good news, and it is REQ-0045 Does a commercial medical policy finding…, raised by Commercial on 2026-09-19: a medical policy finding a therapy medically necessary is not the same fact as that plan's durable medical equipment benefit paying for a home-dispensed device, which is precisely the residual claim 75verified names when it records silence on site of care. That request is against Market Access and is not behind a register row yet, because the claim it would settle has not been written — if Market Access answers it by binding a condition, the row arrives here the way A9 This device–A17 A prescription home pelvic floor electrical… did.
• **One thing the good news does not buy, recorded because the temptation is to spend it on A16 This candidate's intended use.** claim 74verified's permissiveness is what makes claim 89unconfirmed worth registering, and it is also one policy. A single payer's criteria are not "what the commercial market requires", and A16 This candidate's intended use's proposition is written against the best-evidenced commercial policy for that reason — the wording is a limit on the claim, not a hedge in it.

• The statement: claim 57unconfirmed — that commercial payer allowed amounts for a non-implanted pelvic floor electrical stimulator are set independently of the Medicare DMEPOS national unadjusted amount for E0740, rather than benchmarked to it.
• What the check found: the Verifier routed it to adoption, which returned unverifiable — the claim asserts what payers would do and needs primary research. The routing is the finding worth reading: adoption was the closest available tool, and the reason it was the closest available tool is that **no connector in this repo reaches commercial payer fee-schedule methodology at all.** The Verifier's note says the claim stays unconfirmed permanently absent a human-attached payer rate study.
• Who depended on it: register row A11 Commercial payer allowed amounts for a… — Market Access and Commercial, through Market Access's [C6].
• What it costs: P1 commercially-insured first's economics. If commercial DME allowables are expressed as a percentage of, or a floor against, the Medicare DMEPOS schedule, then $74.52 anchors the commercial route too and P1 commercially-insured first has collapsed back into the economics it was chosen to escape. The $350–$700 net-revenue-per-system assumption would then need a buyer who is not a payer at all — cash-pay, or an employer — which is a third shape this opportunity has not been framed as.
• Outcome: commission
• Because: unreachable-by-connector is not the same as unanswerable, and conflating them is how a permanent unconfirmed gets quietly treated as a soft no. Payers publish the basis on which an allowable is set, in the reimbursement schedules that accompany the policies REQ-0020 Retrieve already retrieves, and the basis is what this claim is about — not a rate. REQ-0022 Read against the same commercial payer… is that commission, front end → research, open since 2026-09-14, and deliberately separate from REQ-0020 Retrieve: one asks whether the class is covered, the other asks what the allowable is anchored to, and only the second decides whether P1 commercially-insured first's economics differ from the route it replaced. **What this review must not do is record the adoption routing as a reason to expect nothing.** A claim the Verifier calls permanently unconfirmed will never carry verified — and the work behind REQ-0022 Read against the same commercial payer… still changes what this opportunity is worth, which is the distinction between a tag and a decision.

• The statement: claim 67unconfirmed — that the number of commercially insured women aged 45+ who would be dispensed a prescription home pelvic floor NMES device each year is materially larger than "the 4,130 annual Medicare fee-for-service E0740 allowed units", and large enough to carry a standalone hardware business.
• What the check found: two different things, and only one of them is in the tag. The Verifier routed the claim to market "<claim text>" --jurisdiction US, which returned unverifiable — "has no free, authoritative, machine-queryable source. Stays unconfirmed permanently unless a human attaches a licensed report." No connector in this repo reaches commercial-payer utilization or insurance-mix data. That is the IR-9 claim 57 came back unconfirmed shape exactly. **The second finding is the Commercial chair's, made cold on 2026-09-19, and no tag can show it: the claim's anchor is misread.** The claim's own note asserts that "the anchor figure (4,130 Medicare FFS E0740 allowed units, 2024) is itself already independently verified elsewhere in this file via procedures E0740". What is verified elsewhere in this file is claim 70verified — 4,130 supplier services across 689 beneficiaries — and claim 72verified, which says E0740 is rental-flagged and that a service count must not be read as a count of devices or of patients. The comparator this claim says commercial volume is "materially larger than" is six times the number of women it was taken to mean.
• Who depended on it: register row A18 The number of commercially insured women… — Commercial, and through it every unit-volume line in knowledge-base/commercial/idea-001.md, including the "implied Medicare-FFS device pool, 4,130 × $74.52 ≈ $307,700/year" the chair names as two different units multiplied together. The row is new and the dependency is not: A18 The number of commercially insured women… was registered on 2026-09-19, five days after the claim was written and six after P1 commercially-insured first superseded A4 TAM component 1 and A5 TAM component 2 without saying what would carry the size chain's volume. No assessment pillar cites claim 67unconfirmed, which is correct and deliberate — knowledge-base/assessments/idea-001.md puts the load-bearing population statement on claim 32unconfirmed under B2 What the manufacturer collects per… and keeps claim 70verified as context — so nothing in the assessment moves on this review, and that is the register catching something the assessment was right not to carry.
• What it costs: the units leg of P1 commercially-insured first's size chain, which is to say the chain has no volume under it. A7 A material share of the addressable… is a share of a population, and a share multiplied by nothing is nothing. Beyond that, the anchor defect costs something a later commission would otherwise pay for twice: **a claim whose comparator is wrong by a factor of six cannot be verified as written even if the research lands**, because a clean answer would be measured against the wrong number. And the direction is easy to get backwards, so it is stated: 689 rather than 4,130 makes "materially larger" easier to satisfy and the hardware business harder, because the floor a commercial population has to beat is smaller while the evidence that anyone buys this device at all is six times thinner than this file has been reading it.
• Outcome: commission
• Because: unconfirmed here is a gap in the check twice over and neither half is a finding about the world — nobody has measured commercial volume and found it small; nobody has measured it. Absorbing was available and is wrong: this is the one leg of the revenue chain that P1 commercially-insured first created and then left unregistered, and absorbing it would mean deciding the opportunity does not rest on how many units it sells. Pivoting is wrong for the same reason IR-3 claim 32 is unconfirmed and P1 now rests on… was a commission and not a pivot — the statement can still be settled, and a pivot away from an unasked question moves the risk rather than reducing it. The vehicle had to be new. REQ-0003 Payer mix for diagnosed stress urinary… settles payer mix and never a unit count; REQ-0020 Retrieve and REQ-0022 Read against the same commercial payer… are coverage and allowable. Folding this into any of them would repeat IR-6 claim 14 is unconfirmed and the…'s error exactly — a commission folded into a request whose question could not carry it — and that error cost this file five days and a request that closed without touching the claim. REQ-0046 Build the market model this candidate has… is the commission, front end → commercial, raised 2026-09-19: build the market model this candidate has never had, and rebuild the units leg on the unit basis claims 70 and 72 establish.
• **What the commission may not be asked to do, and this is the half that is this record's own rule rather than the request's.** It must not correct claim 67unconfirmed. Claims are append-only, claim 67unconfirmed carries a Verifier's tag against its existing wording, and editing a tagged claim's statement so it matches a later finding would leave a tag attached to a sentence nobody checked — the one rule, broken by tidying. The restatement goes in as a new appended claim through the ordinary path, unverified, and REQ-0046 Build the market model this candidate has… says so in its question. Until that lands, A18 The number of commercially insured women… stands on claim 67unconfirmed as written and this review is the record of why the row and the claim do not say the same thing.
• What the honest reading of a clean REQ-0046 Build the market model this candidate has… is, said now rather than when it arrives. A market model is a decomposition with a band and a coverage line, not a source. It can type the gaps and file the researchable ones; it cannot make market's verdict go away, and the comparative half of this claim may well stay unconfirmed permanently for the same reason A11 Commercial payer allowed amounts for a…'s does. That is not a reason to skip it — IR-9 claim 57 came back unconfirmed's lesson holds, unreachable-by-connector is not unanswerable and a permanent unconfirmed must not be quietly read as a soft no — but a reader should expect this row to become better-founded, not verified.

• The statement: claim 33unconfirmed — that closed-loop EMG-gated stimulation produces a better continence outcome than open-loop stimulation or pelvic muscle exercise alone, by a margin a clinician would act on.
• What the check found: the Stage 3 pass of 2026-08-28 ran adoption against it and returned unverifiable on the "a clinician would act on" half, recording that no primary research on clinician response margin exists in this repo; on the comparative-efficacy half it named the Cochrane review as the closest available evidence and stopped there. No connector here reaches a comparative-efficacy margin, so the tag is a gap in the instrument rather than a finding about the world. **What is new on 2026-09-19 is not a tag but a search.** The Clinical chair ran a PubMed window from 2000 for EMG-triggered pelvic floor stimulation and got status: none; ran a narrower window and got one 2004 German review of anal incontinence; ran a 111-study registry sweep of electrical stimulation in urinary incontinence and found not one closed-loop or EMG-gated pelvic floor study among the forty returned; and reports that the AJOG 2026 systematic review could not pool the 32 open-loop RCTs it did find. The statement has moved from unchecked to unattested, and the front end records that as the chair's reading rather than as a tag, because it is not this department's to issue and a position is not evidence.
• Who depended on it: nothing in this register, which is the finding. Pillar B1 The discriminating trigger works in the… — fatal — has rested on this claim in the assessment since 2026-09-14 while no register row named it. Behind it now: A28 Closed-loop EMG-gated stimulation produces…, the front end's own row; A19 No randomised comparison of closed-loop…, the Clinical chair's [C1], which is the statement that nobody has measured what A28 Closed-loop EMG-gated stimulation produces… asserts; every deck that has called the closed loop the differentiator; and the predicate argument itself, which distinguishes this device from InTone, InToneMV and InWave by the gating and by nothing else.
• What it costs: not the pathway — a KPI 510(k) against those predicates does not require this question to be answered, and that is the trap rather than the relief. What it costs is a cleared, covered device with no evidenced reason for a clinician to choose it over a therapy already on the market under three other clearances, in a modality whose 56-trial Cochrane comparison against pelvic floor muscle training found no difference on low-quality evidence. The chair has priced the answer: two arms, gated against open-loop on the same probe and the same envelope, n≈120, twelve-week endpoint to match the only posted result in the space, twelve to eighteen months, roughly $1.5-3M, and no sponsor anywhere in this file.
• Outcome: commission
• Because: unconfirmed here is a gap in the check and not a finding about the world — no instrument in this repo reaches a comparative-efficacy margin — and the protocol's asymmetry routes that to a commission. The other two outcomes were available and both were declined on the record. Absorb is not available: the dependants are named above and none of them survives the claim being false, which is what fatal means in the assessment. **Pivot was considered and declined**, for two reasons that are different and both sufficient: there is no failed statement to pivot on, a tag held for want of an instrument being the opposite of a finding; and the shape option now on the table arrived from a chair that explicitly declined to choose it, so choosing it here would be the front end making a chair's call over a silent Technology chair and an un-restated Regulatory one. It is recorded as an option under ## Pivot options on the table instead, with its evidence named. What is commissioned is the only part of this any artifact in this repo can reach: REQ-0051 Hand-search the included-study tables of…, front-end → research, a hand-search of the included-study tables of PMID 40120731 and PMID 29271482 for any arm gated on the patient's own voluntary surface EMG. A connector sweep establishes that no such trial is indexed under that description; it does not establish that none sits inside a pooled review under another one, and the difference between those two sentences is the whole of what this commission buys. It settles the existence question — claim 91unconfirmed, A19 No randomised comparison of closed-loop… — and it does not settle the margin. The margin is Study 2, it is funded band, it has no owner, and this review says so plainly rather than letting a cheap commission stand in for an expensive answer.

The 2026-09-19 Stage 3 batch — nine claims, seven reviews, and why they are grouped that way

Claims 91–99 were written by the Clinical & Scientific chair on 2026-09-19 and put through a Verifier the same day. All nine came back unconfirmed; none came back refuted. They are answered below in seven reviews rather than nine, because nine ceremonies over one Stage 3 pass would bury the thing that actually distinguishes them. The groupings are stated so they can be argued with:

• IR-14 claims 93 and 94 groups claims 93 and 94 — both are pillar B4 Prescribers will send a woman home with a…'s guideline pair, both ask what guidance the launch jurisdiction actually carries about conservative therapy for stress urinary incontinence, and both are settled by the same unread document set. One answer covers them.
• IR-18 claims 98 and 99 came back exact and are… groups claims 98 and 99 — both are ClinicalTrials.gov reads through trial-detail, both came back numerically exact, and both are held at unconfirmed by the same property of the instrument rather than by anything found. Same cause, same answer.
• Everything else is answered on its own, including claims 91 and 96, which both turn on the same trial (PMID 42190320 / NCT06419517 (a ClinicalTrials.gov registration number)) and are deliberately not grouped: the read left claim 91unconfirmed standing exactly as worded and left claim 96unconfirmed weaker than the chair stated it. Those are opposite directions out of one document, and grouping them would average them.

The distinction that does the routing work in all seven. unconfirmed after a real attempt is not the same thing as unconfirmed because no instrument exists. Three of these lines (92, 95, 99, and the endpoint half of 98) are held below verified by a property of this repo — literature-detail, trials and trial-detail return a ScanResult, a vocabulary deliberately disjoint from confidence tags with no to_claim_line() renderer, so there is no code path from those connectors to verified — and UK guidance has no typed connector at all. Where that is the whole of the gap, the tag understates the evidence and the honest outcome is absorb. Where the chair's reading outran its source, that is a finding about this file and it is said in those words. Where a real question remains open, it is commissioned.

One thing that did not happen, recorded because the assessment said in advance that it would. Pillar B1 The discriminating trigger works in the… carries a stated trigger: a refuted on claim 96unconfirmed sends it to blocking. claim 96unconfirmed did not come back refuted. The trigger did not fire, no pillar is blocking, and nothing below changes a standing, an exposure or a posture — those are rendered from the tags and the tags are the Verifier's.

• The statement: claim 91unconfirmed — that no randomised comparison of closed-loop EMG-gated intravaginal electrical stimulation against open-loop intravaginal electrical stimulation in women with stress urinary incontinence exists in the published literature or in the ClinicalTrials.gov registry, so the margin claim 33unconfirmed asserts has never been measured in any population.
• What the check found: the near-miss this record flagged on the morning of 2026-09-19 was run down, and it does not break the claim. literature-detail 42190320 returned the European Journal of Obstetrics, Gynecology, and Reproductive Biology 2026 paper behind NCT06419517 at abstract depth, and its comparator is verbatim **"PFMT plus sham stimulation"**, with methods reading "(1) PFMT + sEMG-triggered electrical stimulation, or (2) PFMT + placebo stimulation". A sham comparator measures the therapy against nothing; it does not measure the gated form against its ungated form, which is the only comparison claim 33unconfirmed's margin lives in. The token-miss correction this record recorded also checked out exactly: literature "EMG-triggered electrical stimulation stress urinary incontinence" → none, literature "sEMG-triggered electrical stimulation" → found, 2 records. The registry sweep on the gated intervention returns total 1, and that one is NCT06419517. So the tag is held at unconfirmed for the reason IR-11 claim 33 is unconfirmed already named — an absence cannot be established by term-matched sweeps that return ScanResults — and not because anything was found against it.
• Who depended on it: register row A19 No randomised comparison of closed-loop…, the Clinical chair's [C1]; register row A28 Closed-loop EMG-gated stimulation produces… and claim 33unconfirmed, which is the statement of the margin itself; pillar B1 The discriminating trigger works in the…, graded fatal, through its cost band; and the predicate argument, which distinguishes this device from InTone, InToneMV and InWave by the gating and by nothing else. A19 No randomised comparison of closed-loop… was written in the direction where the expensive branch is the row being true, and that is the branch the read leaves standing.
• What it costs: the cheap escape is closed. A19 No randomised comparison of closed-loop… recorded that if a randomised closed-loop against open-loop comparison existed somewhere nobody here had looked, B1 The discriminating trigger works in the…'s cheapest test would drop from a trial to a literature read. It does not exist in the one place it could plausibly have been hiding, so the Clinical chair's Study 2 — two arms, gated against open-loop on the same probe and the same envelope, n≈120, twelve-week endpoint, twelve to eighteen months, roughly $1.5–3M — stays the only instrument that reaches the margin, and it still has no sponsor in this file. What is new and is worth a line on its own: REQ-0052 PMID 42190320's first half asked whether the comparator was open-loop or sham, and the Stage 3 read of this claim has now answered it from the abstract — sham. The request stays open on its second half, the population question, which is IR-13 claim 96 came back weaker than the chair… below.
• Outcome: commission
• Because: unconfirmed here is a gap in the instrument and not a finding about the world, and the protocol routes that to a commission. Absorb is not available: A19 No randomised comparison of closed-loop…, A28 Closed-loop EMG-gated stimulation produces… and a fatal pillar all depend on this line and none of them survives it being wrong in the expensive direction. Pivot is not available and is declined on the record: nothing failed. The claim came back stronger than it went in — the single most on-point trial in the world was read and its comparator is inactive placebo — and pivoting off a statement that survived its check would be a redesign wearing a learning's clothes. What is commissioned is already raised and is unchanged by this read: REQ-0051 Hand-search the included-study tables of…, front-end → research, the hand-search of the included-study tables of PMID 40120731 (32 RCTs) and PMID 29271482 (Cochrane, 56 RCTs) for any arm gated on the patient's own voluntary surface EMG. A connector sweep establishes that no such trial is indexed under that description; only a hand-search establishes that none sits inside a pooled review under another one. It settles the existence question and it does not settle the margin — the margin is Study 2, it is funded band, and this review will not let a cheap commission stand in for an expensive answer.

• The statement: claim 96unconfirmed — that women who cannot actively contract the pelvic floor, the population NICE NG123 1.4.10 names, cannot generate the voluntary surface EMG a closed-loop EMG-gated stimulator needs to trigger on, so this candidate's architecture structurally excludes the one population with a positive guideline recommendation for its modality.
• What the check found: the retrievable half holds exactly — NG123 1.4.10 was read verbatim from NCBI Bookshelf NBK542416 and the population description in the claim is 1.4.10's own. Everything that carries the consequence came back the other way. A cross-check run for claim 91unconfirmed surfaced NCT06419517, published as PMID 42190320, whose registered inclusion criterion is verbatim *"Adult women (18 years old - 50 years old), with a **Modified Oxford Score, determined by digital palpation, **of 0 - 1"*. A Modified Oxford Score of 0 is the absence of a palpable voluntary contraction — operationally what NG123 calls "cannot actively contract". A completed, double-blind, randomised, multicentre trial enrolled that band into an sEMG-triggered stimulation arm and reported gains from it (+0.48 Oxford points at 10 weeks, p=0.033; −1.51 leakage episodes/week at 12 months, p=0.001). That is the first evidence in either direction anyone here has put on this line and it points against the claim's physiological premise. Beyond that, the premise was searched and is unmeasured: three literature sweeps for surface EMG in women unable to contract returned six off-point records and two nones. It was correctly not recorded as a refutation, for three reasons that are gaps rather than hedges: the inclusion criterion is a range, 0–1, and neither source reports how many enrolled participants were at its floor; neither source reports a trigger-detection failure rate, threshold or exclusion — the exact quantity this claim turns on; and enrolment is 50 ACTUAL in the registry against 42 allocated in the publication, which makes per-band counts unreadable from either document. The claim's second step is worse off still: "excludes" treats a trigger threshold as fixed, and nothing establishes that such a device could not ship an open-loop mode for exactly this population.
• Who depended on it: register row A24 Women who cannot actively contract the…, the Clinical chair's [C6]; pillars B1 The discriminating trigger works in the… and B4 Prescribers will send a woman home with a…, both graded fatal; and — A24 Women who cannot actively contract the… says this in its own words — the shape of the opportunity, because A24 Women who cannot actively contract the… is the row that says the differentiator and the indication point in opposite directions. A24 Women who cannot actively contract the… currently holds the distinction between "very weak" and "cannot actively contract" as the whole of the row. That distinction survives, but it has narrowed from a recruitment phrase to a registered ordinal criterion, and it now needs a count rather than a reading.
• What it costs: in the direction the evidence points, nothing — it costs this file an argument it was making against itself, which is the cheapest thing it has been offered all day. In the other direction it is unrepairable by evidence, and A24 Women who cannot actively contract the… says so: once established it is a property of the architecture, and the answers are a design change or a different shape. What the pass actually changes is the price of finding out: before it, the question needed a new study reporting voluntary sEMG amplitude against palpation grade; after it, the question may need one institutional fetch of a paper that has already been written.
• Outcome: commission
• Because: the routing rule is explicit that commissioning against a refuted claim is usually re-litigating a finding — and this claim is not refuted. It is unconfirmed with an on-point document that nobody has been able to open, which is the textbook case for a commission: the statement can still be settled, and the thing that would settle it is named. Absorb is not available — two fatal pillars and the shape question itself depend on this line. Pivot is not available: no statement failed, and a pivot taken *because the evidence went against a claim this file was using to argue against its own architecture* would be the front end deciding a physiological question no source has answered. The commission is REQ-0063 Obtain the full text of PMID 42190320, front-end → research: the full text of PMID 42190320, reporting (a) the baseline Modified Oxford distribution across allocated participants, specifically how many were graded 0 rather than 1, and (b) any reported trigger-detection performance — threshold, failure rate, any participant who could not drive the trigger. REQ-0052 PMID 42190320 stays open on its population half and is the same question asked of the Clinical chair. One thing this review states plainly for whoever reads pillar B1 The discriminating trigger works in the… next: **the assessment's stated trigger for B1 The discriminating trigger works in the… going blocking was a refuted on this claim, and it did not fire.** What did happen is that the claim is now *weakly contradicted by an on-point randomised trial* rather than merely unmeasured, and a human weighing B1 The discriminating trigger works in the… should read it as weaker than the chair stated, not stronger. That is a reading of the evidence, not a tag, and no tag moved here.

• The statement: two, grouped because they are one question asked from two ends — *does the standard of care in the launch jurisdiction generate the documented four-week exercise failure this candidate's intended use requires, and does any US guidance name this modality at all?* claim 93unconfirmed — that NCD 230.8's four-week definition of a failed pelvic muscle exercise trial is shorter than the first-line duration clinical guideline recommends (NICE NG123 1.4.4, at least 3 months supervised), so a woman who satisfies the intended-use precondition has not completed a guideline-standard trial. claim 94unconfirmed — that no United States clinical guideline recommends electrical stimulation as a treatment option for stress urinary incontinence.
• What the check found: they came back very differently and the grouping must not hide it. claim 93unconfirmed holds on both numbers and the gap it describes is wider than it claims. The US half went through a typed connector and returned verified: ncd 230.8 --jurisdiction US, CMS Pub. 100-3 ch. 1 pt. 4 (Rev. 13808, issued 05-29-26) §230.8, four weeks exact and verbatim. NG123 1.4.4 was read verbatim from NBK542416 — "at least 3 months' duration", carrying [2019], a recommendation the update actually revisited. And the two trials differ in kind as well as length: NCD 230.8 requires "an ordered plan of pelvic muscle exercises" and says nothing anywhere about supervision, while 1.4.4 requires the three months to be supervised. So the gap is 4 unsupervised weeks against 12 supervised ones, not 4 against 12. The residual is jurisdictional, and it is the reason this is not a formality: the line carries jurisdiction: US while half its evidence is UK guidance with no standing in US practice, and "the minimum duration that clinical guideline recommends" turns out to mean one guideline, in another regulatory system. **claim 94unconfirmed is the most overstated line in this set.** Its specific half is exact — AUA (American Urological Association)/SUFU Guideline Statement 11 verbatim, its three options, the ± biofeedback qualifier and the Expert Opinion grade all confirmed, and no electrical stimulation option in it or in the earlier counselling statement. Its headline half outran the document three separate ways. (a) **The guideline is not silent on electrical stimulation.** Its Future Directions section says verbatim that these therapies *"may offer some benefit in index SUI patients seeking non-surgical treatment"*, reviews the evidence including the Cochrane review this file already carries, and concludes that *"current data does not suggest superiority of these new emerging technologies in comparison to established non-invasive therapies such as PFME"*. Non-superiority on immature evidence is a weaker and different fact from an absence, and it is the same document pointing the other way. (b) The document's own title is *"Surgical Treatment of Female Stress Urinary Incontinence"* — a guideline scoped to surgery, carrying exactly one non-surgical statement at Expert Opinion grade, is the wrong instrument for establishing what US guidance says about a conservative modality. (c) "No United States clinical guideline" is a universal quantifier resting on one document: ACOG (American College of Obstetricians and Gynecologists), AUGS (American Urogynecologic Society) and the AUA/SUFU overactive bladder guideline were not read.
• Who depended on it: register rows A21 Medicare NCD 230.8's four-week definition… (claim 93unconfirmed) and A22 No United States clinical guideline… (claim 94unconfirmed), the Clinical chair's [C3] and [C4]; pillar B4 Prescribers will send a woman home with a…, graded fatal, whose settled by: names these two and claim 95unconfirmed by number; Regulatory through the label and REQ-0044 Can the intended_use drop the precondition…; and Commercial's [C4] at A16 This candidate's intended use, which holds the four-week precondition as a commercial choice to keep or drop. A22 No United States clinical guideline…'s "if it is true" clause — that the device asks clinicians to do something their own specialty society's non-surgical statement does not list — survives only in the narrow form. The front end corrects A22 No United States clinical guideline… rather than letting the chair's framing stand: what is established is an absence from the graded statements of one surgical guideline; what is established against it is a panel position of non-superiority on immature data. Those are not the same claim and this record has more than once spent a coverage fact and a recommendation fact as one.
• What it costs: B4 Prescribers will send a woman home with a… is the pillar that asks whether a prescriber will produce the documented four-week failure the indication requires, and its cheapest test was twenty structured calls after these guideline reads landed. They have landed and B4 Prescribers will send a woman home with a… is no better off than it was, for one reason: neither read tells a US prescriber anything. NG123 binds nobody in the launch jurisdiction, and the one US document read is scoped to surgery. If the four-week precondition really has no clinical trigger in US practice, the prescribing event the entire revenue chain begins with has none either, and that is structural rather than a matter of prescriber effort — but nothing read this pass establishes it in the country this device launches in. The cost of the gap is that B4 Prescribers will send a woman home with a…'s twenty calls would currently be spent without knowing what the callers' own societies tell them to do.
• Outcome: commission
• Because: both lines are unconfirmed with a named, unread, retrievable document set behind them, which is the case the protocol routes to a commission, and one answer settles both. Absorb is not available: A21 Medicare NCD 230.8's four-week definition… and A22 No United States clinical guideline… sit under a fatal pillar and claim 94unconfirmed's headline is load-bearing for the prescriber argument in exactly the form that the check found unsupported. Pivot is not available: neither came back refuted, and the correct response to "the chair's reading outran its source" is to read the rest of the source set, not to redesign the product. The commission is REQ-0062 Read the United States guideline corpus for…, front-end → research: read ACOG, AUGS and the AUA/SUFU non-neurogenic overactive bladder guideline and report, per document, (a) any stated minimum duration or supervision requirement for a first-line trial of pelvic floor muscle training, and (b) whether electrical stimulation is named as a non-surgical option for stress urinary incontinence, with its evidence grade — quoting the sentence, and saying "not stated" where the document is silent. That is the only thing that can turn claim 94unconfirmed's universal quantifier into either a finding or a correction, and it is the only thing that gives claim 93unconfirmed a comparator with US standing. It is deliberately not a connector sweep: no connector in this repo reaches specialty-society guidance and none was invoked on either line. What this commission does not buy: it does not tell anyone what a urogynecologist in front of a patient at week five actually does. That is still twenty calls.

• The statement: claim 95unconfirmed — that NICE NG123 recommends against routine pelvic floor EMG as biofeedback (1.4.6) and against routine electrical stimulation combined with PFMT (1.4.9), and confines electrical stimulation and biofeedback to women who cannot actively contract (1.4.10).
• What the check found: all three recommendation numbers are right and two of the three paraphrases are verbatim-accurate. Four findings sit on top of that, and two of them cut against each other. (i) "Confines" is not what 1.4.10 says. 1.4.10 is a purely positive recommendation — *"Electrical stimulation and/or biofeedback should be considered for women who cannot actively contract pelvic floor muscles **to aid motivation and adherence to therapy**"* — and contains no restrictive clause at all. The confinement is an inference from conjoining 1.4.6 and 1.4.9 with 1.4.10. It is a fair reading of §1.4 as a whole; it is still a reading and not a quotation, and the front end records it as the chair's synthesis. (ii) **1.4.10's stated purpose is narrower than "the modality's one positive indication" suggests** — it is a motivational adjunct to PFMT, not a continence-efficacy recommendation. (iii) 1.4.6 names perineometry as well as EMG — *"Do not use perineometry or pelvic floor electromyography as biofeedback as a routine part of pelvic floor muscle training"* — so the manometric substitution that Technology's [C8] proposes buys nothing against this guideline. (iv) All three carry [2006], against 1.4.4's [2019]: the 2019 update did not revisit them, so the negative stance rests on evidence predating the current home-device generation. That is context for a human to weigh, not a finding that the recommendations have changed — they are current as published. The tag is structural and permanent within this repo: UK guidance has no typed connector, recheck_claims.py cannot replay a Bookshelf accession fetch, and _shared/claim-grammar.md says a jurisdiction beyond US and EU stays unconfirmed for that reason. Upgrading this line needs a human attaching the primary NICE publication, not a different agent lookup.
• Who depended on it: register row A23 UK clinical guideline NICE NG123 recommends…, the Clinical chair's [C5]; pillar B4 Prescribers will send a woman home with a…, which names it by number in its settled by:; and claim 96unconfirmed at A24 Women who cannot actively contract the…, which borrows 1.4.10 as "the one population with a positive guideline recommendation for its modality". Those dependants survive, and A23 UK clinical guideline NICE NG123 recommends… says why in its own first sentence: **NG123 binds no US payer and no US prescriber.** What a UK "do not routinely use" costs this candidate is not coverage, not the 510(k) and not US clinician behaviour; it is the non-US expansion leg of a market model this file does not yet have. A24 Women who cannot actively contract the…'s borrowing survives too, but it must now carry 1.4.10's own rationale with it — the population claim 96unconfirmed says the architecture forfeits is one the guideline recommends the modality for **as a motivational adjunct**, which is a smaller prize than "its one positive indication" reads as.
• What it costs: the guideline-tailwind argument, in the UK, for a device that launches in the US. Priced honestly that is a line in a market model that has not been built and an expansion leg that starts behind. It is not zero and it is not on the critical path, and this review declines to inflate it into either.
• Outcome: absorb
• Because: the substance was retrieved, verbatim and complete, from the primary text; the only thing standing between this line and verified is that no instrument in this repo can reach UK guidance, which the claim's own residual calls permanent. Commissioning against that would be commissioning a lookup that cannot return — the honest version is the one already written into the claim, that a human attaches the NICE publication. And pivot is not available twice over: nothing failed, and the consequence is bounded to a jurisdiction outside this candidate's launch_jurisdiction:. The dependants are named above and each survives for a stated reason rather than by silence: A23 UK clinical guideline NICE NG123 recommends… survives because its own worst case is a non-US expansion leg; B4 Prescribers will send a woman home with a… survives because it is a US prescriber question and NG123 does not reach one, which is why REQ-0062 Read the United States guideline corpus for… in IR-14 claims 93 and 94 and not this review is what B4 Prescribers will send a woman home with a… is actually waiting on; A24 Women who cannot actively contract the… / claim 96unconfirmed survives but narrows, because 1.4.10 read in full is a motivation-and-adherence recommendation rather than an efficacy one. Two findings here bear on the shape options rather than on this claim, and they are recorded there with their evidence rather than acted on here: 1.4.10's "to aid motivation and adherence" strengthens Option 2 the adherence device rather than the…, and 1.4.6 naming perineometry weakens Option 3 the manometric trigger. Recording them under ## Pivot options on the table is not a pivot and pivot: stays at P1 commercially-insured first.

• The statement: claim 92unconfirmed — that between-day reliability of vaginal surface EMG is too low for it to be a between-day outcome measure of pelvic floor muscle function, so an EMG trend transmitted from a home device to a prescribing clinician is not a clinically interpretable measure of change.
• What the check found: both papers were re-retrieved cold and both say what the chair says they say. PMID 19539646 (J Neurosci Methods 2009) is exact — *"between-day reliability was generally poor for both vaginal probes (ICC((3,1))=0.08-0.84)"*. Three things qualify it, and the first is the one a reader has to carry. **(i) The quoted conclusion starts mid-sentence and drops a concessive clause that cuts the other way.** The full sentence opens *"The results suggest that **although it is acceptable to use PFM surface EMG as a biofeedback tool for training purposes**, it is not recommended for use to make between-subject comparisons or to use as an outcome measure between-days…"*. The same sentence the chair cites against the clinician-facing trend endorses surface EMG for the in-session training-biofeedback use, which is this candidate's other use of the same signal. (ii) The ICC range from PMID 31642114 is mislabelled in the chair's summary — 0.65 is a plateau mean, not a peak; the peak-derived parameters are the lower pair, 0.48 and 0.51, and confidence intervals reach 0.04–0.28. Neither paper's own conclusion is a prohibition: 31642114's authors treat interday reliability as a parameter-choice and menstrual-timing problem. (iii) **The load-bearing half was not measured by either paper.** Both used a clinic-positioned probe under one controlled protocol. Neither measured a self-inserted probe, unsupervised, at home, where placement depth and rotation, hydration, time of day and technique all vary in ways these designs hold constant. The step from "between-day ICC is poor in the lab" to "a home-device EMG trend is not clinically interpretable" is an inference no retrieved source reaches. The tag is also structurally capped — literature-detail returns a ScanResult — but that is not the whole of the gap here, which is why this review does not end the way IR-18 claims 98 and 99 came back exact and are… does.
• Who depended on it: register row A20 Between-day reliability of vaginal surface…, the Clinical chair's [C2]; Market Access through the RTM leg at A8 Pelvic floor muscle status qualifies as…; pillars B3 The discriminating trigger is a difference… and B7. A20 Between-day reliability of vaginal surface… is the row that has the recurring-revenue leg failing on content rather than on billing, and B7 is the pillar that calls the longitudinal EMG dataset a defensible asset. **The front end narrows A20 Between-day reliability of vaginal surface… against what the check found:** the failure mode this row describes is specific to the *between-day, clinician-facing trend*, and the same source endorses the in-session biofeedback function, so a reader must not carry this claim across to the device's biofeedback feature. That is a correction in the candidate's favour on one leg and no help at all on the other.
• What it costs: if it is true, what is billed under RTM is a clinician reading probe-repositioning variance, and the asset B7 calls defensible is a longitudinal record of probe placement. That is a second and independent way the recurring-revenue leg fails, next to the coding question at A8 Pelvic floor muscle status qualifies as… and A15 A money model that counts no… — and the two are independent, so a favourable coding judgement does not rescue it. It also reaches claim 108unconfirmed, the sixteen-days-in- thirty capture threshold, which counts days of use and says nothing about whether the captured signal means anything.
• Outcome: commission
• Because: the load-bearing half of this claim has never been measured by anybody, in either direction, and that is a gap a search can close or can report as empty — which is a commission, not an absorb. Absorb would require the dependants to survive the claim being true, and A20 Between-day reliability of vaginal surface… is explicit that they do not: the RTM leg fails on content independently of the coding judgement. Pivot is not available; nothing failed, and the option this finding bears on (the adherence device, Option 2 the adherence device rather than the…) is recorded under ## Pivot options on the table with its evidence rather than taken here. The commission is REQ-0061 Search for any published measurement of…, front-end → research: any published measurement of between-day reliability of pelvic floor surface EMG with a self-inserted probe outside a clinic — ICC or equivalent, with task, probe, interval and sample — and an explicit "none found" plus the nearest analogue if nothing exists. Direction for whoever reads the answer, recorded as reasoning and not as evidence, because the claim's own note records it the same way: the lab condition is the best case, so home variance is unlikely to be lower. Nobody has measured it, and this claim is not verified on the strength of two papers that did not.

• The statement: claim 97unconfirmed — that intravaginal electrical stimulation for urgency and mixed incontinence rests on a different proposed mechanism and a different stimulation frequency band from stimulation for stress incontinence — neuromodulatory inhibition of detrusor overactivity at low frequency versus motor recruitment at higher frequency — so broadening this candidate's indication within continence is a second therapy on the same hardware rather than a wider claim on the same therapy.
• What the check found: three different things, and the claim's own note splits them correctly. The mechanism half is supported by a typed connector and a primary source. clearance-summary K133826 --expect-indication "detrusor" --jurisdiction US → verified: FDA's cleared Indications for Use for an intravaginal stimulator in this candidate's own predicate pool names both mechanisms as two distinct things in one sentence — "strengthening of pelvic floor muscles" and *"inhibition of the detrusor muscle through reflexive mechanisms"*. ncd 230.8 → verified adds that *"the stimulus frequency and other parameters are chosen based on the patient's clinical diagnosis"*, which supports the shape of the claim. The frequency-band half is unsourced as retrieved. Neither source names a band, a number or a direction: NCD 230.8 says frequency varies by diagnosis and stops, and K133826's IFU carries no stimulation parameters at all. The only stimulation envelope anywhere in this file's evidence is the stress-side one (50 Hz, at claim 101unconfirmed); the urge-side band has never been retrieved, and two literature sweeps returned none and three off-point records. So the contrast that makes this claim informative is **one measured number short** — searched and unresolved, not refuted. **And the conclusion runs against the one regulatory fact found.** "A second therapy on the same hardware rather than a wider claim on the same therapy" is a characterisation, and K133826 is FDA clearing **one device under one indication statement that spans both mechanisms at once**. Whatever is true mechanistically, the regulatory system — which is the frame that actually governs whether broadening needs a new submission — has already treated the pair as one cleared indication.
• Who depended on it: register row A25 Intravaginal electrical stimulation for…, the Clinical chair's [C7]; pillar B1 The discriminating trigger works in the…; Commercial's broadening proposal; and Regulatory through REQ-0049 Can an intended use covering stress, which asks whether an indication covering stress, urgency and mixed can be supported against a KPI predicate whose own Indications for Use already name mixed urinary incontinence. **The front end corrects A25 Intravaginal electrical stimulation for… against what the check found.** A25 Intravaginal electrical stimulation for…'s "if it is true" branch says the broadening becomes a second protocol in a different frequency band with its own bench and clinical justification and, on this file's Elidah record, a second 510(k). The frequency-band premise of that branch is now known to be unsourced, and the regulatory premise has one primary document pointing the other way. What survives untouched is the part of A25 Intravaginal electrical stimulation for… that never depended on the band: an inhibitory protocol has nothing voluntary to gate on, so the EMG-gating differentiator has no role in it, and the shape that makes the market largest is still the shape that makes the differentiator smallest. That sentence is the reason this row exists and the check did not touch it.
• What it costs: if the claim is right, broadening buys a bigger market at the price of a second evidence package and a differentiator that does not apply to half of it. If it is wrong, the broadening Commercial has framed is close to free and the full scope of the one verified commercial policy in this file (Aetna CPB 0223, claims 74 and 75) is reachable on the hardware as designed. The pass has not moved that fork — it has established that this file cannot currently tell which side of it is true, and that the one document it holds leans towards the cheap side.
• Outcome: commission
• Because: the missing piece is a number in a published document, which is the definition of a commission rather than an absorb: absorb would require the dependants to survive the claim being false, and Commercial's broadening proposal and REQ-0049 Can an intended use covering stress are both waiting on exactly this. Pivot is not available — nothing failed, and choosing between the stress-only and the broadened shape is Commercial's and Regulatory's call rather than this department's, which is what A25 Intravaginal electrical stimulation for…'s last sentence already says. The commission is REQ-0064 Establish, front-end → research: establish from primary documents the stimulation frequency used for urge or overactive-bladder intravaginal stimulation against the frequency used for stress — the cleared labelling and Instructions for Use of the KPI pool (K133826, K150180, K121820) for programme parameters, and the per-trial protocol tables of PMID 29271482 and PMID 40120731 by indication, with an explicit "not stated" per document. REQ-0049 Can an intended use covering stress stays open and is the other half: whether two mechanisms imply two submissions is an FDA submission-strategy judgement settled in a pre-submission, not by any database read here, and REQ-0064 Establish does not reach it.

• The statement: two, grouped because they share a cause and an answer — both are ClinicalTrials.gov reads through trial-detail, both returned every figure exact, and both are capped at unconfirmed by the same property of the instrument. claim 98unconfirmed — that no published trial of home pelvic floor electrical stimulation for stress urinary incontinence reports its primary efficacy endpoint at four weeks or earlier, the registered sponsor trials of the two adjacent cleared devices setting primary endpoints at twelve weeks (NCT03782116) and six weeks (NCT04752709). claim 99unconfirmed — that 31 of 82 randomised women (37.8%) did not complete the six-week intervention period in NCT04752709, across two active arms with no sham.
• What the check found: **claim 99unconfirmed is the best-evidenced line in this set and is exact to the decimal.** The posted results module gives STARTED 26 + 56 = 82, COMPLETED 17 + 34 = 51, NOT COMPLETED 9 + 22 = 31; 31/82 = 37.80%. "Two active arms with no sham" is confirmed structurally — Arm 1 EXPERIMENTAL, Arm 2 ACTIVE_COMPARATOR, one DEVICE intervention, no PLACEBO_COMPARATOR or SHAM_COMPARATOR anywhere in the record — and masking is QUADRUPLE despite there being no sham, the blinding being between two active protocols. The residual is that why 31 women did not complete is not on the record: the module posts milestone counts and no withdrawal reasons. **claim 98unconfirmed's two named endpoints are exact, and the pass found two further home-use trials the chair did not cite, both pointing the same way.** NCT00762593 — transvaginal, home use, stress urinary incontinence, PHASE2, randomised, double-masked, n=150 — sets its primary endpoints at 8 weeks, and is the closest trial in the registry to this candidate. NCT02423005 (Atlantic Therapeutics, n=180) sets all thirteen outcomes at 12 weeks. **The earliest primary endpoint found anywhere is 8 weeks, twice the four-week precondition.** The nearest thing to a counterexample is recorded and dismissed for two independent reasons: NCT06419517 lists a registry primary at "four weeks", but it is supervised clinic PFMT plus stimulation rather than home stimulation, so it is outside this claim's scope as worded — and its publication reports the primary at 10 weeks, a registry-versus-publication discrepancy a human citing either document should know about. Three residuals stand and none of them reverses anything: the first half of claim 98unconfirmed is a universal negative answered with registry sweeps, and older literature predating ClinicalTrials.gov is outside every sweep run; registered endpoints are intentions — NCT00762593 and NCT02423005 both carry hasResults=false; and **"the two adjacent cleared devices" is one sponsor and one regulation**, both Elidah/ELITONE under product code QAJ / 21 CFR (Title 21 of the Code of Federal Regulations) 876.5330, which this candidate's own frontmatter says is not its regulation, with one of the two being an urge trial cited inside a stress claim.
• Who depended on it: register row A26 No published trial of home pelvic floor… (claim 98unconfirmed), the Clinical chair's [C8], feeding pillars B1 The discriminating trigger works in the… and B4 Prescribers will send a woman home with a…; and register row A27 In Elidah's registered trial of the ELITONE… (claim 99unconfirmed), [C9], feeding pillar B5 Women will start an intravaginal home… — graded fatal — and Commercial's broadening proposal. Both survive, and the reason is the point of this review: their dependants read the documents, not the tag. A26 No published trial of home pelvic floor… exists to say that the first clinical work has to run to twelve weeks to be comparable to the only posted result in the space, and that conclusion is now carried by four registered trials at 8, 12, 12 and 6 weeks instead of two — it got stronger, not weaker. A27 In Elidah's registered trial of the ELITONE… exists to say that 37.8% over six weeks is the third independent attrition measurement in this device category and the worst per unit time, next to 29.8–40.0% over twelve weeks (NCT03782116) and 15.2% over twelve (PMID 25074260); every one of those figures is on the record and REQ-0050 Reconcile the adherence evidence for… is already open to read them as one series with denominators stated. **One thing does not survive and is corrected rather than absorbed:** claim 98unconfirmed's "two adjacent cleared devices" framing. They are not adjacent in the sense A26 No published trial of home pelvic floor… has been reading them — they are one company under a different regulation from this candidate's, and half the citation is an urge trial. The support for A26 No published trial of home pelvic floor… should be read off NCT00762593, which is intravaginal, home and stress, and which the chair did not cite.
• What it costs: nothing that a lookup can recover, and that is the finding. The twelve-week duration A26 No published trial of home pelvic floor… implies is a real cost to the first clinical work and no plan in this file carries it; the 37.8% at A27 In Elidah's registered trial of the ELITONE… is a real number and it removes the last argument for broadening into urgency, which was that a different indication might hold patients better. Both of those costs are already priced into B1 The discriminating trigger works in the… and B5 Women will start an intravaginal home…. What the unconfirmed tags cost is **nothing at all**, because no decision anywhere in this file turns on either line's tag moving.
• Outcome: absorb
• Because: there is no gap in the world here, only a gap between a retrieved document and a vocabulary. trials and trial-detail return a ScanResult — found / none / unavailable — which by construction carries no confidence tag and has no to_claim_line() renderer, so there is no code path from these connectors to verified. Both lines were checked against the primary record, every figure came back exact, and claim 99unconfirmed in particular is exact to the decimal place across five separate figures. Commissioning here would be commissioning a re-run that cannot return a different tag, and dressing a structural ceiling as open work; claim 98unconfirmed's genuinely open residual — a universal negative over pre-registry published literature — is the one part that a hand-search could touch, and REQ-0051 Hand-search the included-study tables of… already sends a hand-search into the included-study tables of the two pooled reviews that would hold it. Pivot is not available: nothing failed, and claim 98unconfirmed came back better supported than it was written. This absorb is deliberately not a clean bill of health, and two things are stated so nobody reads it as one. First, A26 No published trial of home pelvic floor… and A27 In Elidah's registered trial of the ELITONE… stay live and unconfirmed, so they are neither verified nor commissioned and this record does not converge over them; closing them needs a human, or a connector this repo does not have, not another agent pass. Second, the correction above — that the "two adjacent cleared devices" are one sponsor under the wrong regulation — is absorbed as a citation error and not as a finding about the world, because the conclusion it was supporting is carried by trials the chair had not found.

The 2026-09-29 Stage 3 batch — twenty-three claims, nineteen reviews, and why they are grouped that way

The Verifier ran claims 86–120 on 2026-09-29. Every register row resting on them came back unconfirmed. None came back refuted. check_loop.py failed this record on twenty-three identical lines: A13 Medicare Advantage rather than Medicare…, A15 A money model that counts no…, A17 A prescription home pelvic floor electrical… and A29 A closed loop in which stimulation is…–A49 No annual build volume has been stated… except A37 Automatic capture of at least 16 days of…. Nineteen reviews, IR-19 claim 86 to IR-37 claim 120, answer those twenty-three.

Three more rows came back unconfirmed in the same pass, and the checker did not name them. They are A14 The average Medicare fee-for-service… (claim 87unconfirmed), A16 This candidate's intended use (claim 89unconfirmed) and A37 Automatic capture of at least 16 days of… (claim 108unconfirmed). The checker counts a claim as reviewed if any review body cites it. IR-2 claims 16 and 17 came back refuted, IR-8 claim 56 came back unconfirmed and IR-16 claim 92 cite claims 87, 89 and 108 in passing, in the [claim N: unverified] form, because they were written when those claims had never been attempted. Those citations are dependencies noted before any check ran. They are not answers to the check, and a passing checker over them would be "note it and carry on" reached by accident. IR-38 claim 87 to IR-40 claim 108 answer them. The stale tags inside IR-2 claims 16 and 17 came back refuted, IR-8 claim 56 came back unconfirmed and IR-16 claim 92 are left as written: they record what those claims were on the day those reviews were written.

The groupings are stated so they can be argued with:

• IR-23 claims 101 and 109 groups claims 101 and 109. Both were read against the same abstract of PMID
  1. Both are held because the fact they assert sits in a full text that has returned HTTP 403 to three agents. Same document, same gap, same request.
• IR-29 claims 107 groups claims 107, 116 and 117. All three hit the same clearance-summary crash on K150180, K133826 and K121820. That crash is an environment fault in the Verifier's runtime, not a finding about any 510(k). This pass confirmed it is still present: python3 -c "import _cffi_backend" fails with ModuleNotFoundError.
• IR-36 claims 118 and 119 groups claims 118 and 119. Both are the same MAUDE (Manufacturer and User Facility Device Experience) call, returned in full (24 of 24, not truncated). Both are held by the same property of the instrument, a ScanResult, which has no path to verified.
• Everything else is answered alone. That includes 100, 103 and 104, which share a request (REQ-0153 Read the full specifications of US6240315…) but fail for three different reasons: an argument with no source, a crashed connector, and a disclosure no connector reads.

The routing rule this batch actually turned on. Twenty-three unconfirmed tags are twenty-three gaps in the check, so the default is commission. It is not applied blindly. Where a claim's dependants already run on its unfavourable branch, nothing waits on the tag moving, and a commission would be a heading over work already raised. There the honest outcome is absorb, and it says what carries the dependency instead. Seven of the twenty-two reviews take that route: IR-24 claim 102, IR-28 claim 106, IR-31 claim 111, IR-32 claim 112, IR-36 claims 118 and 119, IR-37 claim 120 and IR-38 claim 87. Fifteen commission, and nine new requests back them: REQ-0152 Retrieve and read the FDA 510 to REQ-0160 For app-connected home pelvic floor devices. No pivot, because nothing failed about the world. pivot: stays at P1 commercially-insured first.

One verified fact from the same pass is used below and is not reviewed, because it did not fail. claim 82verified: E0740 is a Medicare capped-rental item, category CR, with one RR line and no purchase line. Title passes after 13 continuous rental months (42 CFR 414.229). The 2024 book averages about 6.0 rental months per beneficiary. Web searches had suggested E0740 was "inexpensive or routinely purchased". That lead is contradicted on the face of the fee schedule. It bears on IR-32 claim 112 and IR-33 claim 113. It is also most of what REQ-0059 Which DMEPOS payment category does HCPCS… asks. Answering REQ-0059 Which DMEPOS payment category does HCPCS… is Market Access's act, not this department's.

• The statement: claim 86unconfirmed. Medicare Advantage, not fee-for-service, accounts for a majority of US Medicare beneficiaries. So the commercially insured share of the SUI population cannot be derived as the complement of the fee-for-service population the 689-beneficiary E0740 count comes from.
• What the check found: nothing about the premise. No connector in this repo reaches Medicare enrolment by program. procedures E0740 confirms only that the 689 figure is fee-for-service only: it excludes Medicare Advantage, commercial, Medicaid, VA and cash-pay. The Verifier calls the inference half a method point that holds only if the premise holds.
• Who depended on it: A13 Medicare Advantage rather than Medicare…, which is Commercial's [C1]. Behind it are REQ-0003 Payer mix for diagnosed stress urinary…'s eventual answer, A7 A material share of the addressable… claim 32unconfirmed (which the note under P1 commercially-insured first calls the whole of the pivot's forward-looking premise), and pillar B9 Payment reaches this device as billed. A13 Medicare Advantage rather than Medicare… is registered in A12 Medicare NCD 30.1.1 excludes home use of…'s direction: the expensive branch is the claim being true.
• What it costs: if the claim is true, a two-bucket payer-mix answer (commercial against Medicare) mis-states P1 commercially-insured first's addressable population by the entire Medicare Advantage enrolment. That population is Medicare-primary. It is outside the commercial DME benefit P1 commercially-insured first is built on. It is invisible in the fee-for-service book. A7 A material share of the addressable… would then come out of its own commission still open.
• Outcome: commission
• Because: this is a pure gap in the instrument, and the missing fact is one public CMS table away. Absorbing is not available: A13 Medicare Advantage rather than Medicare… decides the shape of REQ-0003 Payer mix for diagnosed stress urinary…'s answer, and that answer is still outstanding. A pivot is not available because nothing failed. REQ-0154 From CMS primary sources (front-end → research) asks for the Medicare Advantage against Original Medicare split from CMS's enrolment files, for 2024 and the latest month. That settles the premise. Once the premise is read, the method half is arithmetic. Whatever REQ-0154 From CMS primary sources returns, REQ-0003 Payer mix for diagnosed stress urinary… should come back in three buckets rather than two. That is stated here so the answer cannot arrive in the wrong shape.

• The statement: claim 88unconfirmed. A money model with no RTM revenue and no biofeedback service revenue, leaving only the E0740 (or commercial-equivalent) device line, supports a standalone hardware business at commercially insured unit volumes.
• What the check found: the Verifier routed the claim to market, which returned unverifiable. The claim has no free, authoritative, machine-queryable source, and it stays unconfirmed permanently unless a human attaches a licensed report. The Verifier named the registry facts that bear on it without settling it. The only Medicare line is a capped-rental monthly fee (claim 82verified). The 2024 fee-for-service floor is 689 beneficiaries.
• Who depended on it: A15 A money model that counts no…, which is Commercial's [C3], and A8 Pelvic floor muscle status qualifies as… through the RTM leg. Pillar B2 What the manufacturer collects per…, graded fatal, and the Commercial chair's abstention also depend on it.
• What it costs: if the claim is false, the money model is not a smaller business but a different one. It becomes a hardware company that does not close on hardware and needs a recurring line CMS has not named for this organ (A8 Pelvic floor muscle status qualifies as…), carrying a content problem it cannot bill around (A20 Between-day reliability of vaginal surface…).
• Outcome: commission
• Because: "permanently unconfirmed" is a property of the connector set, not of the question, which is IR-9 claim 57 came back unconfirmed's point in a new place. The claim is a model conclusion, and what settles a model conclusion is the model. REQ-0046 Build the market model this candidate has… (front-end → commercial) builds it, on a per-beneficiary unit. REQ-0060 When the market model REQ-0046 commissions… (operations-npi → commercial) makes it state a build volume by object and a price to the channel. Only with both can a device-line-only case be shown to close or not. Both are open and neither is re-scoped by this review. **What this review adds:** with claim 82verified, the device line is itself a stream of monthly rental lines ending at month 13 at the latest, averaging about six. The "one-time device sale" that A8 Pelvic floor muscle status qualifies as… described as the fallback does not exist under Medicare. REQ-0046 Build the market model this candidate has… has to model the device line as rental months, not units sold. Commercial should read this before re-stating.

• The statement: claim 90unconfirmed. A commercially insured patient is reached through the same DMEPOS supplier channel that bills Medicare. So the 28 suppliers and 553 referring providers in the 2024 fee-for-service E0740 row describe the national channel, not just its Medicare-billing subset.
• What the check found: procedures E0740 returned verified on the release itself. The 28 and 553 are in the cached national row. The release is fee-for-service only by construction, so those are the suppliers and referrers with Medicare fee-for-service E0740 claims in 2024. The release cannot show whether a commercial-only dispensing channel exists, so it neither establishes nor refutes the claim's reading. That reading rests entirely on the unchecked premise that commercial patients go through the same suppliers.
• Who depended on it: A17 A prescription home pelvic floor electrical…, which is Commercial's [C5]. Also the one-representative- territory go-to-market reading, the channel build named in claim 53unconfirmed, and pillar B9 Payment reaches this device as billed.
• What it costs: if the claim is false, the channel P1 commercially-insured first would sell through is unmeasured. The channel build stops being a line item and becomes the business, sized from nothing. A17 A prescription home pelvic floor electrical… was registered because the same 28 suppliers are also the strongest signal that nobody has built a business in this benefit. Both readings are load-bearing.
• Outcome: commission
• Because: the numbers were read, but the premise under them never was. That premise is a question about commercial DME network rules, which is Market Access's field, and no request in this file asked it. REQ-0045 Does a commercial medical policy finding… is adjacent and different: it asks whether a medical-policy finding pays under the DME benefit, not through whom. REQ-0155 Is a commercially insured patient's home… (front-end → market-access) asks whether national commercial plans (Aetna first) require Medicare DMEPOS enrollment of in-network DME suppliers or route home stimulators through a national vendor, and whether any of the 28 are in-network.

• The statement: claim 100unconfirmed. A closed loop triggered by a threshold crossing of voluntary surface EMG needs that EMG sensed before the burst, not during it. So concurrent same-electrode recovery is not required for this candidate's closed-loop control function.
• What the check found: no connector reaches control-architecture reasoning, and none was invoked. The Verifier recorded that claim 104unconfirmed's patent record is consistent with an alternating stimulate/measure architecture existing. It added in terms that this is not evidence for the "requires" half.
• Who depended on it: A29 A closed loop in which stimulation is…, which is Technology's [C1] and its answer to REQ-0047 Can the closed loop be implemented as…. Also A6 FTO — no live, enforceable US patent claims…, which has read "the mechanism" as one object since 2026-09-13, and pillar B1 The discriminating trigger works in the…, graded fatal, whose bench gate moved to A34 No published study reports the sensitivity… on the strength of this row. The register's standing decisions depend on it too: not registering claim 34unconfirmed, and not editing the mechanism: frontmatter. Both were argued on this claim, and both named its tag as what would reverse them.
• What it costs: if the claim is false, concurrent recovery is required after all. claim 34unconfirmed becomes load-bearing again, the IP exposure re-welds to the signal-recovery problem (A6 FTO — no live, enforceable US patent claims…'s two architectures collapse into one), and B1 The discriminating trigger works in the…'s bench half is a millisecond budget with no headroom. One boundary a reader must hold: the claim is scoped to the control function. The frontmatter says "concurrent surface-EMG sensing for closed-loop biofeedback". Whether a biofeedback display during stimulation needs concurrent sensing is a separate question this claim does not answer. It is recorded here and not resolved, because the front end does not decide engineering questions.
• Outcome: commission
• Because: the claim is an argument, and an argument is settled by the existence of what it predicts. It predicts devices that trigger stimulation on voluntary EMG sensed outside the burst. REQ-0153 Read the full specifications of US6240315… (front-end → research) reads the full specification of US6240315 and US6289245 for the alternating combination mode, and the MyoTrac Infiniti / MyOnyx manual for whether EMG is sensed during the burst or only outside it. REQ-0156 From the full text of PMID 42190320 reads PMID 42190320's methods for where its trigger was taken from. REQ-0048 Do the issued claims of US10307589 (clinical → legal-ip) carries the freedom-to-operate half. Nothing is reversed while these run. The mechanism: frontmatter stays as written, claim 34unconfirmed stays unregistered, and the conversion condition recorded in the register (claim 100unconfirmed verified plus REQ-0053 US6240315 and US6289245 or REQ-0054 Two of the three KPI predicates this file… back) is unchanged. unconfirmed is not refuted, and the register said in advance what refuted would do.

• The statement: two claims, grouped. claim 101unconfirmed: PMID 42190320's stimulation envelope is 50 Hz, 250 µs symmetrical biphasic, 2 s on and 4 s off with 1 s ramps, leaving an inter-burst interval of about 4,000 ms. claim 109unconfirmed: that trial, the only published RCT (randomised controlled trial) of this architecture in this indication, delivered the stimulation under clinic supervision three times a week for ten weeks alongside supervised PFMT. So it evidences the architecture, not unsupervised home use.
• What the check found: literature-detail 42190320 returned the abstract (a ScanResult, which carries no tag). **For claim 101unconfirmed the abstract contains none of the envelope figures**: no 50 Hz, no 250 µs, no duty cycle, no ramp. The figures are unconfirmed, not contradicted. The ~4,000 ms figure is derived, and it depends on something nobody has read: whether the 1 s ramps sit inside the 2 s on-time or the 4 s off-time. **For claim 109unconfirmed the abstract states**: "Both groups received supervised PFMT three times/week for 10 weeks plus a daily home program". The supervision and the schedule are therefore stated for the PFMT. Where the stimulation itself was delivered, and under whose supervision, is not stated. The "only published RCT" half is an absence, which a ScanResult cannot establish (claim 91unconfirmed's sweep history).
• Who depended on it: A30 The therapeutic stimulation envelope… (Technology [C2]). Every interval in the Technology position moves with this envelope, which carries A29 A closed loop in which stimulation is…'s headroom argument and A31 Real-time recovery of volitional surface…'s "about 500 times" comparison. A38 The only published randomised trial of… (Technology [C10]) and pillar B1 The discriminating trigger works in the…'s second half depend on 109, and so does the intended_use:, which names the home and no supervisor.
• What it costs: if 101 is wrong, the trigger detector's time budget moves by whatever the real duty cycle is. At 50 Hz any plausible off-time still leaves orders of magnitude more than a 20 ms inter-pulse period, so the direction of A29 A closed loop in which stimulation is… is unlikely to move on this alone. The number quoted for it would. If 109 is wrong in the favourable direction (stimulation delivered at home), the one trial of this architecture transfers to this product with less argument. If it is right, a positive trial under supervision is exactly what a deck would quote for a home device, and it does not reach one.
• Outcome: commission
• Because: both facts are in one document, and the only barrier is access. REQ-0156 From the full text of PMID 42190320 (front-end → research) asks for the envelope, including where the ramps sit, the site and supervision of the stimulation separately from the PFMT, and the trigger channel. It rides the same institutional retrieval as REQ-0063 Obtain the full text of PMID 42190320, so one copy answers both. The "only published RCT" half is not re-commissioned here. REQ-0051 Hand-search the included-study tables of…'s hand-search of the two pooled reviews already reaches it. What this review asks of every later reader: do not quote "about 500 times" until REQ-0156 From the full text of PMID 42190320 returns. It is a ratio over a number nobody here has read.

• The statement: claim 102unconfirmed. Concurrent recovery of volitional EMG during functional electrical stimulation on shared electrodes needs a blanking window of about 8 ms per pulse (7.6 ms in PMID 33836509). At 50 Hz that is about 38% of the 20 ms inter-pulse period. So concurrent works on a millisecond budget and inter-burst on one about 500 times larger.
• What the check found: the abstract confirms verbatim that this paper's detection front end prevents saturation "with a blanking time of 7.6 ms". It was tested in limb FES on eight able-bodied volunteers and three stroke patients, not intravaginally. The arithmetic 7.6/20 = 38% holds. Not verified: generalising one prototype's blanking into what concurrent recovery requires. The "about 500 times" rests on claim 101unconfirmed's 4,000 ms, which is itself unconfirmed (IR-23 claims 101 and 109).
• Who depended on it: A31 Real-time recovery of volitional surface…, which is Technology [C3]. The row's own words say "the specific thing that breaks is the comparison rather than the inter-burst design". A29 A closed loop in which stimulation is…'s headroom argument cites it as contrast. Pillar B1 The discriminating trigger works in the… lists it among the claims it rests on. claim 34unconfirmed would come back only as an option.
• What it costs: if the generalisation is false and concurrent recovery can be done with far less blanking, concurrent becomes affordable. That puts claim 34unconfirmed back on the table as a design option nobody has priced. It does not make the inter-burst design worse, remove the bench gate at A34 No published study reports the sensitivity…, or change what the four grants read on. The cost is an unpriced alternative, not a broken dependency.
• Outcome: absorb
• Because: nothing this opportunity has chosen rests on concurrent being expensive. A29 A closed loop in which stimulation is… rests on whether concurrent is required, which is claim 100unconfirmed (IR-22 claim 100, commissioned), not on how much it costs. B1 The discriminating trigger works in the…'s bench half moved to trigger specificity at A34 No published study reports the sensitivity… (IR-27 claim 105, commissioned), which is indifferent to blanking time. A31 Real-time recovery of volitional surface… survives as a statement of contrast. It stays live and unconfirmed, so this record does not converge over it, and its false branch is an option to be priced if anyone ever wants concurrent. It is not a premise to defend. Commissioning a literature survey of blanking windows would buy the number for a design this candidate is not building. The one correction carried forward is the one IR-23 claims 101 and 109 makes: the "500×" is not quotable yet.

• The statement: claim 103unconfirmed. EMG-triggered stimulation through a single intravaginal probe used for both stimulation and sensing is a capability of a device on the market today (Thought Technology MyoTrac Infiniti / MyOnyx). So the inter-burst architecture is a product-level capability in this anatomy, not an unbuilt one.
• What the check found: 510k "MyoTrac Infiniti" --applicant "Thought Technology" returned verified: K053266, decided 2005-12-29. The cleared scope (EMG-triggered stimulation, a vaginal sensor specified for STIM and EMG) is in the 510(k) summary. clearance-summary crashed at import on every K-number tried, so K053266 was not attempted. "Marketed today" is answered by no connector, since a clearance says nothing about current sale. The product-specification wording is manufacturer literature, which is not reached.
• Who depended on it: A32 EMG-triggered electrical stimulation…, which is Technology [C4]. Also A43 Measuring the trigger sensitivity and… and A34 No published study reports the sensitivity…, whose cheap branch (a purchase order plus an IRB rather than 15–25 prototype probes) rests on this device being a usable rig. REQ-0058 Candidate claim 103 says EMG-triggered…, B1 The discriminating trigger works in the…'s bench half, and the "the architecture is ordinary" reading all depend on it.
• What it costs: if the claim is false, the inter-burst architecture in this anatomy is unbuilt, the first same-probe build is ours, and A34 No published study reports the sensitivity… falls into its expensive branch: months, set by the probe and its safety file.
• Outcome: commission
• Because: a 2005 clearance was found. That is a real advance on a marketing page, and the gap between it and the claim is three documents, one of them behind a broken import. REQ-0152 Retrieve and read the FDA 510 (front-end → research) reads the K053266 summary for what was cleared. REQ-0153 Read the full specifications of US6240315… reads the manufacturer manual for sense/stimulate timing and current sale. REQ-0156 From the full text of PMID 42190320 reads PMID 42190320 for whether its trigger came off the stimulating probe; Technology's own position names that as the branch that would drop this row and the published precedent together. REQ-0058 Candidate claim 103 says EMG-triggered… (operations-npi → technology) stays the vehicle for whether the device exposes a raw channel and a settable threshold. Separately, the Verifier re-run owed once clearance-summary imports is recorded at IR-29 claims 107.

• The statement: claim 104unconfirmed. US6240315 (priority 1998, filed 2000-05-30, expired) discloses an apparatus that stimulates vaginal-wall muscles and measures EMG from them in a combination mode that alternates the two, with a user-settable switching interval. That is an inter-burst architecture in this exact anatomy, disclosed and now in the public domain.
• What the check found: only the dates were checked, and only partly. patent US6240315B1 via EPO (European Patent Office) OPS returned verified on the filing date, 2000-05-30, exactly. "Expired" matches only as a naive 20-year estimate (2020-05-30) that does not model term adjustments. **No priority date is returned**, so "priority 1998" is unchecked. No connector reads the disclosure content. The returned title, "Electrical apparatus for medical treatment using EMG envelope signal", is consistent but is not the disclosure. The Verifier marks "in the public domain" as a legal conclusion.
• Who depended on it: A33 US6240315, which is Technology [C5]. Legal & IP depends on it through REQ-0053 US6240315 and US6289245. Pillar B6 No live US patent stops P2 from shipping (fatal) depends on it as the only prior art found that narrows the four live grants from outside their own file histories. So does B3 The discriminating trigger is a difference…'s moat half, because the same disclosure is equally available to a fast-follower.
• What it costs: if the disclosure is not what the Technology chair read it as, B6 No live US patent stops P2 from shipping loses its only public-domain floor. REQ-0053 US6240315 and US6289245 would then be asking Legal & IP to weigh a document nobody here has read.
• Outcome: commission
• Because: a patent description is a public primary document. No connector here renders one, and that is the whole gap. REQ-0153 Read the full specifications of US6240315… (front-end → research) reads both specifications for the priority date and the combination mode. REQ-0053 US6240315 and US6289245 (technology → legal-ip) stays the vehicle for what the disclosure does to the live grants. Whether anything is "in the public domain" for this candidate's purposes is counsel's reading, not a fact a request can return. It is not commissioned as one.

• The statement: claim 105unconfirmed. No published study reports the sensitivity or specificity with which a vaginal surface-EMG channel separates a voluntary pelvic floor contraction from a Valsalva or abdominal brace. So the specificity of an unsupervised home EMG trigger in this anatomy has never been measured.
• What the check found: no sweep was run. The Verifier recorded that a term-matched sweep returning ScanResults cannot establish that no study exists (the structural limit on claim 91unconfirmed), so none could change the tag. The claim is untested, not tested and found wanting.
• Who depended on it: A34 No published study reports the sensitivity…, which is Technology [C6] and is B1 The discriminating trigger works in the…'s bench gate since REQ-0047 Can the closed loop be implemented as…. A43 Measuring the trigger sensitivity and… decides that study's cost band. B1 The discriminating trigger works in the… is graded fatal.
• What it costs: this row's unfavourable direction is true. If nobody has measured it, an unsupervised home device could fire on Valsalva and reward the wrong manoeuvre with nobody in the room. The next spend is then a 10–15 volunteer bench study, weeks or months depending on A43 Measuring the trigger sensitivity and….
• Outcome: commission
• Because: exactly as the register said in advance: "a true answer here is a commission, not a refutation". REQ-0055 Search for any reported sensitivity or… (technology → research) is the literature search that could break the absence, and it names electrode site, manoeuvres and metric, with "not reported" where a study compares without a metric. It is open and correctly scoped. If it returns nothing, the bench study follows, with its rig decided by REQ-0058 Candidate claim 103 says EMG-triggered… (IR-34 claim 114). This review does not re-band B1 The discriminating trigger works in the…. Exposure bands are rendered from tags, and no tag here moved toward verified.

• The statement: claim 106unconfirmed. Moving from a concurrent to an inter-burst architecture removes the stimulation-artefact problem. It does not change the between-day reliability of the vaginal surface EMG measure, so it does nothing to make a home EMG trend more interpretable.
• What the check found: this is reasoning over claim 92unconfirmed, the between-day reliability claim, and no connector reaches the inference.
• Who depended on it: A35 Moving from a concurrent to an inter-burst…, which is Technology [C7]. A20 Between-day reliability of vaginal surface… is the row it is about. A8 Pelvic floor muscle status qualifies as… and A15 A money model that counts no… carry the recurring-revenue leg that fails on content rather than billing. Pillar B3 The discriminating trigger is a difference… also rests on it, and it lists this claim by name.
• What it costs: if the claim is false, the architecture change partly repairs A20 Between-day reliability of vaginal surface…. That would be good news for the data leg. If it is true, A20 Between-day reliability of vaginal surface… survives REQ-0047 Can the closed loop be implemented as… untouched and the data leg still fails on content.
• Outcome: absorb
• Because: everything that depends on this row already runs on its unfavourable branch, and none of it waits on the tag. A20 Between-day reliability of vaginal surface… is live, reads the content problem as unrepaired, and is commissioned through REQ-0061 Search for any published measurement of…: a search for home, self-inserted-probe reliability. That search settles the question this inference is about. A8 Pelvic floor muscle status qualifies as…, A15 A money model that counts no… and B3 The discriminating trigger is a difference… already carry the data leg as failing on content independently of billing. No deck, row or pillar in this file claims the REQ-0047 Can the closed loop be implemented as… answer helped reliability, which is the only thing a false answer here would license. A commission would buy a second reading of the same question REQ-0061 Search for any published measurement of… asks directly. A35 Moving from a concurrent to an inter-burst… stays live and unconfirmed, so this record does not converge over it, and it is the row to re-read if REQ-0061 Search for any published measurement of… comes back favourable, because a favourable A20 Between-day reliability of vaginal surface… makes this inference moot rather than true.

• The statement: three claims, grouped because they share a cause and an answer. claim 107unconfirmed: closed-loop control contingent on a voluntary contraction can be built on an intravaginal pressure (manometry) sensor instead of surface EMG. That is the modality K150180's cleared Indications for Use name. claim 116unconfirmed: replacing the EMG trigger with a manometric one adds to the disposable probe rather than removing from it. Either a barrier filter is needed on an air-coupled path, or a transducer and a per-unit calibration step move into the probe. The InTone insertion unit is the precedent. claim 117unconfirmed: KPI-pool probes are supplied non-sterile and cleaned by the user, so terminal sterilisation validation is not on this device's critical path.
• What the check found: nothing about any of the three documents. clearance-summary K150180 crashed at import before any request was made: pypdf → cryptography binding, "No module named '_cffi_backend'", PanicException. It crashed the same way on K133826 and K121820. The Verifier recorded this as "an environment fault in the Verifier's runtime, not a finding about any 510(k) summary". The design halves of 107 and 116 are answered by no source. This pass confirmed the fault is still present on 2026-09-29.
• Who depended on it: A36 Closed-loop control contingent on a… (Technology [C8]), pillar B6 No live US patent stops P2 from shipping, REQ-0054 Two of the three KPI predicates this file… (technology → regulatory), and Option 3 the manometric trigger under ## Pivot options on the table. A45 Replacing the surface-EMG trigger with an… (Operations [C6]) and pillar B2 What the manufacturer collects per… depend on 116. A46 Intravaginal pelvic floor electrical… (Operations [C7]), B2 What the manufacturer collects per… and the transfer schedule depend on 117. claim 83unconfirmed (B8 The route is one 510) is not a register row, but it is held by the same crash.
• What it costs: if 107 is false, surface EMG is the only way to build the differentiator, and claim 85unconfirmed's design escape from the four grants costs the closed loop itself. If 116 is false, Option 3 the manometric trigger is a cost reduction as well as a regulatory move. If 117 is false, terminal sterilisation, a sterile barrier and residual and package-integrity validation land on a cost floor and schedule that have never carried them. A46 Intravaginal pelvic floor electrical… was registered for exactly that reason: so the good news cannot be spent before a label is read.
• Outcome: commission
• Because: this is the textbook case: a missing module in our runtime, not a fact about the world. Two things are commissioned, and both are named. **First, a Verifier re-run of claims 107 and 117, and of claim 83unconfirmed alongside them**, once clearance-summary imports. That is an environment repair, not a department's work, and it is recorded here so it is not mistaken for one. Second, REQ-0152 Retrieve and read the FDA 510 (front-end → research) retrieves and reads the three KPI summaries and labelling by another route. It covers the manometric-biofeedback language, sterile or non-sterile supply, the insertion-unit transducer architecture and controller reprocessing. That way the commission does not wait on an environment fix no department owns. REQ-0054 Two of the three KPI predicates this file… stays the vehicle for 107's regulatory half. The design halves of 107 and 116 remain the chairs' engineering statements, and the protocol has them written as propositions for exactly this reason. Nobody here closes them by reasoning.

• The statement: claim 110unconfirmed. A closed-loop EMG-gating module can be an accessory to an already-cleared stimulator only where the host exposes a cleared external trigger input. Otherwise gating means controlling the host's energy delivery, which is a change to the host, not the addition of a sensor.
• What the check found: no connector was invoked. This is a question of how FDA treats accessories and device modifications, which is guidance and practice. regulation serves CFR text only and cannot settle interpretation (regulatory.md scope limit).
• Who depended on it: A39 A closed-loop EMG-gating module can be an…, which is Technology [C11]. Also pillar B3 The discriminating trigger is a difference…, where the claim sits, and the OEM-module shape the assessment names as a neighbouring bet. Regulatory depends on it through claim 54unconfirmed's residual on 21 CFR 807.81(a)(3)(i).
• What it costs: if the claim is true, the OEM route is a host-device modification dressed as an accessory, and the buyer inherits a premarket notification. That makes it expensive for a fast-follower too, which is half of B3 The discriminating trigger is a difference…'s moat argument. If false, the OEM shape is a light accessory play for everyone.
• Outcome: commission
• Because: A39 A closed-loop EMG-gating module can be an… recorded on 2026-09-19 that "no request in this file asks it in those words yet". That was a gap the front end left open, and it is closed now. REQ-0157 Where a closed-loop EMG-gating module is… (front-end → regulatory) asks the question against FDA's accessory guidance and the 510(k)-for-a-change guidance. It goes to Regulatory because the answer is Regulatory's field, and that chair has not re-stated since 2026-09-13.

• The statement: claim 111unconfirmed. This candidate is at least four objects with four lifetimes: a single-patient probe, a returnable controller, an app and cloud pathway, and a connectivity subscription. So a single per-unit cost of goods, as claim 48unconfirmed is worded, cannot be stated without naming which object and over what life.
• What the check found: "Operations/product-architecture assertion; no connector reaches it". Nothing was found for or against.
• Who depended on it: A40 This candidate is at least four…, which is Operations [C1], and the Operations chair's abstention. Commercial's [C3] (A15 A money model that counts no…), pillar B2 What the manufacturer collects per…, claim 48unconfirmed, REQ-0046 Build the market model this candidate has… and REQ-0060 When the market model REQ-0046 commissions… all depend on it.
• What it costs: if the claim is false, the abstention was over-careful and a single cost of goods can be stated in claim 48unconfirmed's unit. That costs nothing: it removes work. If it is true, the cost floor has no unit until someone names the object and the life.
• Outcome: absorb
• Because: the claim is a description of this candidate's own design, not a fact about the world. Its halves are already in the file as design choices: claim 84unconfirmed's single-use probe on a returnable system, the app and connectivity in mechanism:. Every dependant already runs on the true branch. REQ-0060 When the market model REQ-0046 commissions… asks for build volume *separated into returnable controllers and single-patient-use probes*. B2 What the manufacturer collects per…'s settled-by names A40 This candidate is at least four… as the reason the denominator does not exist. The register declined to register claim 48unconfirmed on exactly this diagnosis. No source could move the tag, and no decision waits on it moving: a bill of materials delivered by object closes the dependency whatever the tag says. A commission would be a request to confirm a design the file has already adopted. A40 This candidate is at least four… stays live and unconfirmed. Named so it is not lost: if a later pass collapses the architecture (a purchased integral unit, say), this absorb is void and the row is owed again.

• The statement: claim 112unconfirmed. On a rental-flagged code the comparator a manufacturer needs is fleet cost per beneficiary-year against allowed amount per beneficiary-year. So claim 48unconfirmed's comparison of per-unit cost of goods against $74.52 compares a per-object-built figure with a per-service payment, and cannot be closed in its unit.
• What the check found: the premise came back registry-supported. payment E0740 returned verified: $74.52 is E0740's RR monthly rental fee, category CR (capped rental), and the code's only line. --modifier NU returned no purchase line. So $74.52 is a per-rental-month payment, not a per-device price. That is the same fact claim 82verified now carries, with title passing at 13 months and about 6.0 rental months per beneficiary in 2024. What is not settled, and no source settles it, is the method half: that fleet cost per beneficiary-year is the comparator.
• Who depended on it: A41 On a rental-flagged HCPCS code the cost…, which is Operations [C2]. Also A14 The average Medicare fee-for-service… claim 87unconfirmed, the same reframing from the revenue side, and pillar B2 What the manufacturer collects per… and claim 48unconfirmed. REQ-0059 Which DMEPOS payment category does HCPCS… depends on it too, and its question the verified claim above now largely answers.
• What it costs: if the method half is false, a per-unit cost of goods against $74.52 is a fair comparison. With $74.52 now read as one rental month, that branch survives only if "per unit" is redefined as per rental month. So the false branch is less a live possibility than a wording.
• Outcome: absorb
• Because: the part of the claim that a source could reach came back the way the row was written. The rest is a costing convention, and the file adopted it before this pass. REQ-0046 Build the market model this candidate has… was raised asking for a per-beneficiary comparator, and B2 What the manufacturer collects per…'s own proposition is written "in the unit a rental-flagged code actually pays in". No source will ever tag a method choice, so commissioning would only produce a permanently unconfirmed line and another request. What survives, and why: A14 The average Medicare fee-for-service… and A41 On a rental-flagged HCPCS code the cost… still have to land together, as the register says. The market model under REQ-0046 Build the market model this candidate has… is where they land. B2 What the manufacturer collects per… is unchanged because its standing is rendered, not written. **One consequence recorded for the chairs rather than acted on:** REQ-0059 Which DMEPOS payment category does HCPCS…'s web lead ("inexpensive or routinely purchased") is contradicted by the fee-schedule category field, CR. Market Access can answer REQ-0059 Which DMEPOS payment category does HCPCS… against claim 82verified, and it is theirs to answer.

• The statement: claim 113unconfirmed. A returnable rental system leaves a reprocessing and asset-management burden on the reusable controller even where claim 84unconfirmed's single-use probe removes it from the probe. That means cleaning and disinfection, incoming functional and electrical-safety test, battery assessment, data de-provisioning, and a non-return and loss rate.
• What the check found: "Operations/reprocessing assertion; no connector reaches it". Nothing for or against.
• Who depended on it: A42 A returnable rental system leaves a…, which is Operations [C3]. Pillar B2 What the manufacturer collects per… and claim 84unconfirmed depend on it, and so does the working capital a manufacturer funds before payment. A42 A returnable rental system leaves a… was registered in A12 Medicare NCD 30.1.1 excludes home use of…'s direction: the expensive branch is true.
• What it costs: if true, a per-object cost of goods is not the cost of this business. A recurring per-return operating cost and a pre-funded fleet sit outside it, and no figure in this file carries either. **This pass moved one precondition of that cost from open to read.** A42 A returnable rental system leaves a… said the size of the burden "cannot be computed until REQ-0059 Which DMEPOS payment category does HCPCS… says which payment category E0740 sits in". claim 82verified says capped rental, title at 13 continuous months, about 6.0 rental months per beneficiary. So under Medicare the typical controller does come back, well before title passes. The burden is live for most units and absent only for the minority who rent to month 13.
• Outcome: commission
• Because: the return is established, but the burden per return is not. What a returned controller needs is set by its labelling and by what the KPI pool already specifies for reusable parts. REQ-0152 Retrieve and read the FDA 510 item (d) asks exactly that of K150180, K133826 and K121820. The size of the fleet is REQ-0060 When the market model REQ-0046 commissions… behind REQ-0046 Build the market model this candidate has…, controllers separated from probes. REQ-0059 Which DMEPOS payment category does HCPCS… is not the vehicle any more for the category question, which is read. It stays open for Market Access to answer or re-scope.

• The statement: claim 114unconfirmed. Measuring trigger sensitivity and specificity needs a functionally representative probe on human volunteers, because discrimination depends on electrode geometry and seating, not on the detector alone. So unless a cleared clinic device can serve as the rig, the schedule is set by a prototype probe build plus biocompatibility and electrical-safety justification.
• What the check found: "study-design and schedule assertion; no connector reaches it".
• Who depended on it: A43 Measuring the trigger sensitivity and…, which is Operations [C4]. Also A34 No published study reports the sensitivity…, whose cost band it decides; A32 EMG-triggered electrical stimulation… claim 103unconfirmed, on which its cheap branch rests entirely; and pillar B1 The discriminating trigger works in the… (fatal).
• What it costs: if true and the clinic device is not usable, B1 The discriminating trigger works in the…'s bench gate is months set by a probe and its safety file, not weeks set by a study. If false, the probe drops out of the rig and the cheap costing holds.
• Outcome: commission
• Because: the claim is conditional on one fact, and the fact has a vehicle. REQ-0058 Candidate claim 103 says EMG-triggered… (operations-npi → technology) asks whether the marketed device exposes a raw sEMG channel and a settable threshold. The Operations chair calls it the cheapest high-value unknown on its page. REQ-0152 Retrieve and read the FDA 510 (K053266 summary) and REQ-0153 Read the full specifications of US6240315… (the manual) now feed it from the document side (IR-25 claim 103). The "discrimination depends on the probe" premise is a bench-physics statement the Operations and Technology chairs share, and nothing here settles it by argument.

• The statement: claim 115unconfirmed. ISO 10993-1 evaluation of a probe the patient cleans between sessions must use test articles conditioned through the labelled cleaning regimen and aged to shelf life. So claim 36unconfirmed cannot close before a cleaning regimen is specified, and cleaning validation sits ahead of the battery.
• What the check found: ISO 10993-1 is a licensed standard and FDA's biocompatibility guidance is guidance. No connector reaches either (regulatory.md scope limit). Nothing was found for or against.
• Who depended on it: A44 ISO 10993-1 biological evaluation of a…, which is Operations [C5]. Also pillar B1 The discriminating trigger works in the…, the closing route for claim 36unconfirmed, and the Technology/Operations sequencing disagreement recorded at A44 ISO 10993-1 biological evaluation of a… at its stated narrow scope (sequencing, not the battery).
• What it costs: if true, a cleaning regimen is a design input on the critical path that nobody in this file owns. If false, biocompatibility runs beside everything else and Technology's "readiness, not possibility" reading holds.
• Outcome: commission
• Because: the governing text is public in its FDA form, and it is Regulatory's field. REQ-0158 For a single-patient-use intravaginal probe… (front-end → regulatory) asks whether FDA's ISO 10993-1 guidance requires final-finished, cleaned and aged test articles. It asks nothing about the battery, keeping the disagreement at the scope both chairs stated. **This is not a ruling between Technology and Operations.** It sends their disagreement to the one chair whose field the answer is in. Registering A44 ISO 10993-1 biological evaluation of a… was the same restraint.

• The statement: two claims, grouped because they share a call, a cause and an answer. claim 118unconfirmed: in MAUDE reports coded KPI (2010-01-01 to 2026-09-19), the recurring device complaint for non-implanted pelvic floor stimulators is burning or pain at the stimulating electrode. No report attributes an event to a sensing function, so production controls belong on the electrode, not the sensing chain. This is a statement about which failure mode recurs, never a rate. claim 119unconfirmed: the same KPI set includes implanted sacral neuromodulators, an electromagnetic chair, an I.V. bag and an orthopaedic plating system, so KPI is not a usable post-market surveillance filter.
• What the check found: `maude --product-code KPI --since 2010-01-01 --until 2026-09-19 --limit 100 returned a ScanResult: found`, **24 reports, total_matches 24, truncated false. The pull is the whole population, not a sample. **For 118, burning at the electrode recurs among the non-implanted reports: MW5110267, MW5110278 ("PATHWAY VAGINAL EMG/STIMULATION"), and 1723686-2013-00002 (Zynex, "burning sensation … at 23mA"). The one EMG-capable device, MW5110278, describes a stimulation burn, not a sensing fault. For 119, every category the claim names is present: InterStim (e.g. 20300107, 7120554, MW5093294, MW5073208), Emsella (MW5118108, MW5149580), an I.V. bag (MW5071908) and an orthopaedic plate (0009613350-2017-01398). The Verifier marks both conclusions as inference, and both are capped because MAUDE returns a ScanResult, which carries no tag.
• Who depended on it: A47 In MAUDE reports coded to product code KPI… (Operations [C8]), pillar B1 The discriminating trigger works in the…'s sensing-chain half, and the production-control plan, which has no owner. A48 MAUDE reports returned under product code… (Operations [C9]), pillar B3 The discriminating trigger is a difference…, and the post-market surveillance plan, which does not exist yet.
• What it costs: nothing a lookup can recover. The two things the rows exist to protect are already protected by what was read. MAUDE must never be read as a rate (A47 In MAUDE reports coded to product code KPI…), and a contaminated MAUDE product code says nothing about FDA's 510(k) product codes or the predicate pool (A48 MAUDE reports returned under product code…). Both limits are written into the rows and stand unchanged.
• Outcome: absorb
• Because: there is no gap in the world here, only a gap between a complete document pull and a vocabulary. This is IR-18 claims 98 and 99 came back exact and are…'s shape: maude returns found/none/unavailable and has no code path to verified. A commission would be a re-run that cannot return a different tag. What survives, and why: A47 In MAUDE reports coded to product code KPI…'s dependants read the narratives, which point where the row points. A48 MAUDE reports returned under product code…'s dependants read the device names, which contain every contaminant the row names. The conclusions ("production controls sit on the electrode", "KPI is not usable for PMS") are Operations' design judgements drawn from those reads. They are not facts a Verifier could ever reach. B3 The discriminating trigger is a difference…'s settled-by names the one read that would add anything, a re-pull by device name rather than product code, and that is an improvement to a future PMS plan, not an open statement. Deliberately not a clean bill of health: A47 In MAUDE reports coded to product code KPI… and A48 MAUDE reports returned under product code… stay live and unconfirmed, and this record does not converge over them. Twenty-four reports in sixteen years is itself a small number, and nothing here should be read as a tolerability finding.

• The statement: claim 120unconfirmed. No annual build volume is stated anywhere in this candidate file. The only unit figure, 689 fee-for-service beneficiaries, is a payer floor, not a build plan. Cost of goods at hundreds of probes a year and at tens of thousands are different numbers, so claims 48 and 49 are functions of a volume the file does not have.
• What the check found: the first half is about the file, not the world, and no connector answers it. The Verifier searched the file anyway on 2026-09-29 for "build volume", "units per year" and "annual volume" and found no build-volume figure. The "annual volume" hits are Medicare utilisation, claim 80unconfirmed, and Stage 4 notes. The cost-structure half is reasoning that no source reaches.
• Who depended on it: A49 No annual build volume has been stated…, which is Operations [C10]. Also A18 The number of commercially insured women… claim 67unconfirmed, the units leg this row says does not exist, claim 48unconfirmed, pillar B2 What the manufacturer collects per…, and the REQ-0046 Build the market model this candidate has… → REQ-0060 When the market model REQ-0046 commissions… chain.
• What it costs: if the claim is false, a volume exists somewhere and a cost of goods is arithmetic. The search says it does not. If true, claims 48 and 49 cannot be closed by lookup at all, only by producing a volume.
• Outcome: absorb
• Because: the checkable half was checked by the only method that reaches it, a read of the file, and it came back as written. The other half is standard cost-structure reasoning that no source will ever tag. Its dependants already act on it: REQ-0060 When the market model REQ-0046 commissions… exists because this is true, and it asks for the volume by object with a band. A commission would re-raise REQ-0060 When the market model REQ-0046 commissions…. The claim is self-limiting: it stops being true the day REQ-0060 When the market model REQ-0046 commissions… answers, and at that point A49 No annual build volume has been stated… is superseded by the answer rather than reviewed. What survives, and why: claims 48 and 49 stay open for the reason this row gives, B2 What the manufacturer collects per… carries that reason in its settled-by, and A18 The number of commercially insured women…'s units leg stays with REQ-0046 Build the market model this candidate has…. A49 No annual build volume has been stated… stays live and unconfirmed, and this record does not converge over it.

IR-38 claim 87 to IR-40 claim 108 answer three rows check_loop.py did not name. See the note at the head of this batch for why they are here.

• The statement: claim 87unconfirmed. The average Medicare fee-for-service allowed amount per E0740 beneficiary in 2024 is about $347: 4,130 services at $57.89, divided by 689 beneficiaries. So the revenue unit for a rental-flagged code is allowed amount per beneficiary per year, not the per-service average or the $74.52 schedule amount.
• What the check found: the arithmetic is exact: 4,130 × 57.889305085 ÷ 689 = 346.9998. The inputs were re-verified on 2026-09-29 by procedures E0740. The per-service average is the same 2024 national row's Avg_Suplr_Mdcr_Alowd_Amt (claim 71verified). Not verified: that per-beneficiary-per-year is the revenue unit, which is a modelling choice. The Verifier added two registry facts that bear on it. $74.52 is a monthly rental fee (claim 82verified). And 4,130 ÷ 689 = 6.0 rental months per beneficiary. "Per year" is a data-year aggregate, and a rental that spans two years splits across them.
• Who depended on it: A14 The average Medicare fee-for-service…, which is Commercial [C2], and A41 On a rental-flagged HCPCS code the cost… claim 112unconfirmed, the same correction from the cost side (IR-32 claim 112). Pillar B2 What the manufacturer collects per… depends on it. So does the only argument anyone has made for reopening P1 commercially-insured first, Option 1 unwind P1 under ## Pivot options on the table: that the same 689 women are worth several times what this file assumed.
• What it costs: if the method half is false, the per-service average stands as the unit, claim 18unconfirmed's comparison stands, and Option 1 unwind P1 loses its argument.
• Outcome: absorb
• Because: the number is exact, and the unit it argues for is already this file's working unit. REQ-0046 Build the market model this candidate has… asks for "a per-beneficiary-per-year figure", and B2 What the manufacturer collects per… is written in "the unit a rental-flagged code actually pays in". A method choice will never be tagged by any source, so a commission would only buy another permanently unconfirmed line. What survives, and why: A14 The average Medicare fee-for-service… and A41 On a rental-flagged HCPCS code the cost… land together, as the register requires, because both are absorbed on the same ground in the same pass. Option 1 unwind P1 keeps its argument at the strength the arithmetic gives it. The one caution the Verifier's note adds: "per year" is a data-year artefact. Anything that multiplies $347 by a number of women must say whether it means a data year or a rental course. The honest per-course figure is $74.52 times the number of rental months, capped at 13, which is claim 121unconfirmed's question.

• The statement: claim 89unconfirmed. By requiring a documented failed four-week exercise trial, this intended use restricts the population the device may be sold to below what its best-evidenced commercial policy would cover, because that policy conditions coverage on indication and sex alone.
• What the check found: both document halves hold. The intended_use: is SUI-only and requires the four-week trial. Aetna CPB 0223 covers "stress, urgency or mixed urinary incontinence" in women with no prior-therapy precondition (claim 74verified). So the intended-use population is a subset of the policy's. Not verified: (a) whether a cleared indication restricts sale or prescription, as opposed to labelling and promotion, which is an FDA-interpretation question; (b) "best-evidenced", which is a ranking, not a fact.
• Who depended on it: A16 This candidate's intended use, which is Commercial [C4]. Regulatory depends on it through REQ-0044 Can the intended_use drop the precondition…. So do pillars B4 Prescribers will send a woman home with a… and B9 Payment reaches this device as billed, and A21 Medicare NCD 230.8's four-week definition… claim 93unconfirmed from the other side, which says keeping the precondition may cost the prescription itself.
• What it costs: if (a) is false, the precondition is a labelling constraint and not a population constraint. The narrow label then costs promotion rather than patients, and A16 This candidate's intended use's "live design decision" is cheaper than written.
• Outcome: commission
• Because: the half that decides the cost has never been asked in those words. REQ-0044 Can the intended_use drop the precondition… asks whether the precondition can be dropped and keep the 510(k) route. It does not ask what keeping it restricts. REQ-0159 Does a cleared 510 (front-end → regulatory) asks that: whether a cleared prior-therapy precondition limits who may be prescribed and dispensed the device, or only what may be labelled and promoted.

• The statement: claim 108unconfirmed. Capture of 16 days in every 30 is not limited by the device or the app, because session logging and a periodic BLE sync are ordinary engineering. It is limited by whether the woman uses the device on 16 days. So claim 35unconfirmed is an adherence claim, not an engineering claim.
• What the check found: "Engineering-effort and adherence assertion; no connector reaches either half." Nothing for or against.
• Who depended on it: A37 Automatic capture of at least 16 days of…, which is Technology [C9]. Pillar B3 The discriminating trigger is a difference… reads it by name (its settled-by makes B3 The discriminating trigger is a difference… "hostage to B5 Women will start an intravaginal home…'s attrition series" on this claim). Pillar B5 Women will start an intravaginal home… depends on it through REQ-0050 Reconcile the adherence evidence for…. The RTM device-supply leg at A8 Pelvic floor muscle status qualifies as… and A15 A money model that counts no… also depends on it.
• What it costs: both branches cost something, in different places. If the claim is true, the RTM capture leg fails when and because B5 Women will start an intravaginal home… fails, and is not independent of it. If it is false, a technical risk sits under the RTM leg that no chair has costed. The file currently spends the engineering half as if established: nobody has costed capture and sync.
• Outcome: commission
• Because: the file relies on one half and nobody has checked it. The engineering half is Technology's own statement, and a chair may not settle its own condition, so this does not go back to Technology. REQ-0160 For app-connected home pelvic floor devices (front-end → research) asks for any published or labelled capture-rate figure for app-connected home pelvic floor devices and RTM-billed devices, with shortfalls split between non-use and capture or sync failure. That is the one measurement that separates the two branches. REQ-0050 Reconcile the adherence evidence for… stays the vehicle for the behavioural series.

The 2026-10-01 Stage 3 batch — twenty-one rows, fifteen reviews, and why they are grouped that way

The Verifier ran claims 140–167 on 2026-10-01. Five came back verified (153, 154, 156, 158 and 165). Every other register row resting on the batch came back unconfirmed. None came back refuted. check_loop.py failed this record on twenty-one identical lines: A50 In a multicentre double-blind randomised…–A60 The cleared stimulation envelopes read from…, A63 A closed-loop intravaginal stimulator with…, A65 Under 21 CFR 801.4 a cleared indication's…, A67 No KPI / 21 CFR 876.5320 510–A72 The software and cybersecurity…, A74 A sensing-only study that records vaginal… and A75 No prescription device offered for sale in…. Fifteen reviews, IR-41 claims 142 and 144 to IR-55 claim 167, answer all twenty-one. Every review below was written from the Verifier's note on the claim line, not from the tag.

The routing rule, and how often it was overridden. Twenty-one unconfirmed tags are twenty-one gaps in the check, so the default is commission. It is not applied blindly, on the same rule the 2026-09-29 batch used: where the Verifier re-retrieved the source cold, found the content as written, and held the tag only because the instrument has no path to verified (a ScanResult, a class-2 finding with no approved corroborated tag), a commission would buy a re-run that cannot return a different tag. Where, in addition, the dependants already run on the branch the record shows, nothing waits on the tag. Nine rows are absorbed on that ground (IR-41 claims 142 and 144, IR-43 claim 143, IR-45 claim 148, IR-48 claims 151 and 159, IR-49 claim 152, IR-53 claim 163, IR-54 claim 166) and twelve are commissioned (IR-42 claims 145 and 146, IR-44 claim 147, IR-46 claim 149, IR-47 claim 150, IR-50 claim 155, IR-51 claim 157, IR-52 claims 160, IR-55 claim 167). Three requests were raised: REQ-0166 Read the manufacturer labelling (labelling reads), REQ-0167 REQ-0163 asks whether K243079 (the two ETS products REQ-0163 Two cleared KPI / 21 CFR 876.5320 devices… does not name) and REQ-0168 Retrieve from fda.gov and the US Code (guidance and statute text). Every other commission names a request that was already open. No pivot. pivot: stays at P1 commercially-insured first.

The groupings are stated so they can be argued with:

• IR-41 claims 142 and 144 groups claims 142 and 144; IR-42 claims 145 and 146 groups 145 and 146. All four are about one trial, PMID 42190320 / NCT06419517, and were suggested as one group. They are split because they take different outcomes: 142 and 144 were read cold and match, and nothing a lookup can do moves them; 145 and 146 turn on whether the publication followed its registration, which the full text REQ-0161 From the full text of PMID 42190320 asks for answers.
• IR-48 claims 151 and 159 groups claims 151 and 159. Different findings (a published study and nine 510(k) summaries), one consequence: both are Option 3 the manometric trigger's case, engineering and regulatory, and Option 3 the manometric trigger is not the current shape.
• IR-52 claims 160 groups claims 160, 161, 162 and 164. Same finding: the CFR operands that could be checked were checked, and the proposition rests on an FDA guidance or statute no connector serves. Same outcome: one retrieval request. Their costs differ and are stated separately. claim 157unconfirmed has the same kind of finding and rides on the same request, but it is reviewed alone (IR-51 claim 157) because its consequence lands on B4 Prescribers will send a woman home with a…, a fatal pillar, where the others do not. claim 163unconfirmed has the same kind of finding and is not in the group, because its dependants hold no estimate the answer could move (IR-53 claim 163).

• The statement: two claims, grouped because they share a source, a finding and an answer. claim 142unconfirmed: in a multicentre double-blind RCT of 42 women with SUI and very weak pelvic floor muscles, sEMG-triggered intravaginal stimulation added to supervised PFMT outperformed PFMT plus sham on leakage episodes at 12 months. claim 144unconfirmed: that trial enrolled only women aged 18–50, six to eighteen months postpartum, Modified Oxford 0–1, so it does not establish an effect outside that band, which this intended use does not exclude.
• What the check found: for 142, a cold re-retrieval of PubMed PMID 42190320 (EJOGRB 2026): "multicentre, randomized, double-blind, parallel-group trial, 42 women with SUI … PFMT
  • sEMG-triggered electrical stimulation, or … PFMT + placebo stimulation"; "At 12 months, experimental group showed larger reductions in weekly leakage (-1.51 episodes/week; p = 0.001)". Every element matches. The Verifier adds three cautions: leakage was a secondary outcome (primary was PFM strength); 36 of 42 completed 12-month follow-up; the country is not stated and the journal is European, so jurisdiction: US must not be read as a US population. For 144, the registry NCT06419517 eligibility reads "Adult women (18 years old - 50 years old), with a Modified Oxford Score … of 0 - 1" and "Between six and eighteen months after childbirth". Both are held only because literature-detail and trial-detail return ScanResults, which carry no tag by construction, and because "the only randomised trial" is an absence no sweep establishes.
• Who depended on it: A50 In a multicentre double-blind randomised… (142): pillar B1 The discriminating trigger works in the…'s outcome half, which is fatal; Clinical & Scientific, whose gated-against-open-loop trial is priced on the gated modality having an effect; A19 No randomised comparison of closed-loop… (claim 91unconfirmed) and A24 Women who cannot actively contract the… (claim 96unconfirmed). A52 The only randomised trial of sEMG-triggered… (144): Clinical & Scientific [C10]; the Clinical chair's Study 1, whose population is chosen because of this row; A24 Women who cannot actively contract the…; B1 The discriminating trigger works in the….
• What it costs: nothing a lookup can recover. The branch the record shows is the one each row was written to survive. A50 In a multicentre double-blind randomised…'s own "if true" branch reads: an effect over sham, not a margin over open-loop, so B1 The discriminating trigger works in the…'s outcome half narrows to exactly the margin. A52 The only randomised trial of sEMG-triggered… is registered in A12 Medicare NCD 30.1.1 excludes home use of…'s direction, and the record shows its costly branch. The cost there is a citation: claim 142unconfirmed cannot be quoted for this intended use.
• Outcome: absorb
• Because: this is IR-18 claims 98 and 99 came back exact and are…'s and IR-36 claims 118 and 119's shape. The content was re-retrieved cold and matches, and the tag is held by the class of the instrument. Moving it waits on the corroborated tag proposal, which is a Tier 3 rubric decision, not on any work this file can commission. What did depend on it, and why that survives. B1 The discriminating trigger works in the…'s outcome half turns on the margin, claim 33unconfirmed at A28 Closed-loop EMG-gated stimulation produces…, which IR-11 claim 33 is unconfirmed commissioned and which stays live. claim 142unconfirmed cannot carry that margin on its registered arms; IR-42 claims 145 and 146 says why and keeps that question open. The Clinical chair's Study 1 already runs on A52 The only randomised trial of sEMG-triggered…'s costly branch: its population is the intended use as written, with no postpartum or Oxford restriction, because the one RCT does not reach it. The absence half of 144 is REQ-0051 Hand-search the included-study tables of…'s question (claim 91unconfirmed), and nothing in this file spends the branch where a second trial exists. **What survives as a rule for every artifact:** claim 142unconfirmed is cited as a secondary outcome, at abstract depth, in 36 of 42 early-postpartum Oxford 0–1 women outside the US, against sham. It is never cited as evidence for this intended use and never as evidence about the trigger. A50 In a multicentre double-blind randomised… and A52 The only randomised trial of sEMG-triggered… close as absorbed by IR-41 claims 142 and 144. That is not a finding that the trial supports this device.

• The statement: claim 145unconfirmed: in NCT06419517 the comparator got 2 Hz, 10 µs stimulation to sensory response against 50 Hz, 250 µs to motor response in the triggered arm, with supervised PFMT in both. So PMID 42190320 compares therapeutic-dose triggered stimulation against a sub-therapeutic sham and does not isolate the trigger. claim 146unconfirmed: the registry and the publication disagree on enrolment (50 vs 42 allocated), primary time point (four weeks vs 10 weeks) and schedule (two sessions a week for ten weeks, and "daily … for four weeks" on a portable unit, vs supervised PFMT three times a week); the registry has no posted results.
• What the check found: every registry and abstract operand matches. Arm B "Supervised PFMT + sham EMG-triggered ES", "Frequency: 2 Hz. - Pulse width 10 microseconds Intensity: to sensory response"; arm A "Frequency: 50 Hz. - Pulse width 250 microseconds Intensity: to motor response" (ClinicalTrials.gov API, armsInterventionsModule, 2026-10-01). Enrolment 50 ACTUAL; time frame "Baseline, four weeks, 6 and 12 months"; "Twenty sessions … 2 per week for ten weeks"; "daily endovaginal electrostimulation sessions for four weeks", "portable unit EVOSTIM"; hasResults=false. Abstract: "42 women … allocated", primary "Change in PFM strength at 10 weeks". The one thing the Verifier could not establish is the inference in 145: that the published comparison used the registered arm parameters. The abstract does not say, and 146 records that the two documents disagree on how the trial was run.
• Who depended on it: A53 In NCT06419517 the comparator arm received… (145): Clinical & Scientific [C11]; A28 Closed-loop EMG-gated stimulation produces… (claim 33unconfirmed), the margin; A19 No randomised comparison of closed-loop… (claim 91unconfirmed); A50 In a multicentre double-blind randomised… (claim 142unconfirmed); the Clinical chair's Study 2, specified with no sham arm on the strength of this row; B1 The discriminating trigger works in the…. A54 The NCT06419517 registry record and its… (146): Clinical & Scientific [C12]; Technology & R&D [C10] through A38 The only published randomised trial of… (claim 109unconfirmed), which the registry's portable-unit sentence could move in the favourable direction; A50 In a multicentre double-blind randomised…; REQ-0161 From the full text of PMID 42190320; B1 The discriminating trigger works in the….
• What it costs: if 145 holds as published, nothing in this file shows the trigger does anything, and A28 Closed-loop EMG-gated stimulation produces…'s margin is unmeasured in any population. If the published sham differs from the registration, for instance a triggered sham at therapeutic dose, claim 142unconfirmed is the first in-domain estimate of what the trigger adds, and Study 2 would be replicating a result rather than producing the first. Until 146's discrepancies are resolved, **no artifact may state the trial's setting, its schedule or its primary endpoint.**
• Outcome: commission
• Because: both gaps sit in one document this repo has not read, and that document is already asked for. REQ-0161 From the full text of PMID 42190320 (clinical → research) asks for the full text's (a) setting and supervision, (b) sham parameters and whether the sham was triggered, (c) flow from 50 to 42 to 36, and (d) prespecified primary time point. That is exactly 145's unresolved inference and every one of 146's discrepancies. REQ-0063 Obtain the full text of PMID 42190320 and REQ-0156 From the full text of PMID 42190320 ride on the same institutional fetch and are named so that one retrieval answers all three. No request is raised: a fourth ask for the same paywalled article would be a duplicate. A53 In NCT06419517 the comparator arm received… and A54 The NCT06419517 registry record and its… stay live.

• The statement: claim 143unconfirmed. Among women referred to pelvic floor PT (physical therapist) for urinary incontinence in 2009–2010, 33% of privately insured women did not participate, and insurance noncoverage was a risk factor for nonparticipation.
• What the check found: a cold re-retrieval of PMID 21620356 (Am J Obstet Gynecol 2011): "A cross-sectional study of women referred to PFPT for urinary incontinence between January 2009 and June 2010"; "Thirty-three percent of women with private insurance and 17% with other insurance were PFPT nonparticipants"; "Risk factors for nonparticipation included insurance noncoverage (OR, 103.85; 95% CI, 6.21-infinity)". The content matches. The Verifier notes the window and that the odds ratio's interval runs to infinity, which means very sparse cells. The tag is held on structural grounds: a ScanResult, class 2.
• Who depended on it: A51 Among women referred to pelvic floor…: pillar B4 Prescribers will send a woman home with a…'s documentation half, which is fatal; P1 commercially-insured first, because the finding is about privately insured women; Regulatory via REQ-0159 Does a cleared 510, now answered as claim 157unconfirmed at A65 Under 21 CFR 801.4 a cleared indication's…; Commercial via claim 89unconfirmed.
• What it costs: nothing a lookup can recover. The figure is what the row says. What it bears on, whether the women who never complete PT are cut out by the FDA label or by payer policy, is not this claim's question. It is A65 Under 21 CFR 801.4 a cleared indication's…'s.
• Outcome: absorb
• Because: the content was re-read cold and holds, and the tag is held by the class of the instrument, as in IR-41 claims 142 and 144. What did depend on it, and why that survives. B4 Prescribers will send a woman home with a…'s documentation half is settled by its own settled by: line, twenty prescriber calls behind REQ-0062 Read the United States guideline corpus for…, and none of that waits on this tag. Whether the cut A51 Among women referred to pelvic floor… describes bites at the prescription or only at promotion and payment is A65 Under 21 CFR 801.4 a cleared indication's…, which IR-51 claim 157 commissions. A51 Among women referred to pelvic floor…'s own text already bounds its reach to cases where the ordered plan is a PT referral. What survives as a rule: the 33% is cited with its 2009–2010 window and its referral population. The noncoverage odds ratio is cited as a direction only and never as a magnitude, because an interval to infinity is a statement about cell counts, not about effect size. A51 Among women referred to pelvic floor… closes as absorbed by IR-43 claim 143.

• The statement: claim 147unconfirmed. The MyoTrac Infiniti User Guide (2005) describes EMG-triggered sessions on the same electrodes that trigger on EMG held above a threshold for a set delay during a work step, then stimulate for the rest of that step, with trigger and stimulation off in the rest step. So a commercial same-electrode triggered stimulator senses before the burst, not during it.
• What the check found: no connector reaches a manufacturer user guide, and the guide is hosted on a distributor site no connector output points to, so it was not fetched. The FDA summary for the product family does not describe trigger timing: K053266 lists only a "threshold for trigger by the patient and physician" and work/rest periods. The whole inference rests on the unread guide.
• Who depended on it: A55 The MyoTrac Infiniti User Guide: Technology & R&D [C12]; A29 A closed loop in which stimulation is… (claim 100unconfirmed), whose existence evidence this is, which IR-22 claim 100 asked for; Regulatory [C4] at A63 A closed-loop intravaginal stimulator with… (claim 155unconfirmed), whose equivalence argument maps the architecture onto K243079; B1 The discriminating trigger works in the…'s sensing half.
• What it costs: if false, the inter-burst architecture claim 100unconfirmed argues for has no shipped instance, IR-22 claim 100's question reopens, and A63 A closed-loop intravaginal stimulator with… has no described architecture to compare.
• Outcome: commission
• Because: a gap in the instrument, with the document named and public. REQ-0153 Read the full specifications of US6240315… (front-end → research) already asks, at item (b), for the MyoTrac Infiniti / MyOnyx specification: whether EMG-triggered stimulation is listed, which sensor is used for vaginal STIM and EMG, and whether EMG is sensed during the burst or only outside it. That is this claim. No request is raised. REQ-0166 Read the manufacturer labelling reads the K243079 detector, which is A63 A closed-loop intravaginal stimulator with…'s half, and is not a substitute for this one.

• The statement: claim 148unconfirmed. The MyoTrac Infiniti User Guide (2005) states "EMG triggered STIM is not available for incontinence application" and that EMG-Stim sessions record no data. So as labelled the device delivers neither triggered stimulation in its incontinence mode nor a log of trigger events.
• What the check found: the guide quotation is unchecked; no connector reaches it. The Verifier flagged a conflict for a human. The K201014 510(k) summary, read in full, describes its predicate, the MyoTrac Infiniti System (K053434), in its comparison table with "EMG-Triggered Stimulation (ETS) using NMES specifications for incontinence" and a preloaded "Stress incontinence ETS" programme (200 µs, 45 Hz, work 6 s, rest 10 s). So the claim's conclusion is contradicted for at least the later labelling by an FDA primary source. A phrase reading cannot return refuted, and the 2005 guide may describe earlier firmware, so the tag is unconfirmed. The Verifier asks that the claim not be read as saying the device cannot do ETS for incontinence, and points to claim 165verified.
• Who depended on it: A56 The MyoTrac Infiniti User Guide: Technology & R&D [C13]; A32 EMG-triggered electrical stimulation… (claim 103unconfirmed), whose "in this anatomy" half it runs against; Regulatory [C14] (claim 165verified), which answers it; Operations & NPI [C4] through REQ-0058 Candidate claim 103 says EMG-triggered…, the rig choice; B1 The discriminating trigger works in the…. A56 The MyoTrac Infiniti User Guide is registered in A12 Medicare NCD 30.1.1 excludes home use of…'s direction: the costly branch is the claim being true.
• What it costs: A56 The MyoTrac Infiniti User Guide's costly branch was two things. First, the exclusion as an engineering signal: the one manufacturer experienced in same-electrode triggering chose not to offer it in this anatomy. Second, the rig: a purchased unit buys only the sensing half, and the representative-probe half needs a prototype build.
• Outcome: absorb
• Because: neither half of the costly branch is waiting on this tag. **The engineering signal is already settled by a verified claim.** claim 165verified traces the exclusion to the cleared scope of K053266, which has no incontinence indication. It also shows the same manufacturer later cleared triggered stimulation for incontinence through vaginal probes (claim 154verified). A73 Thought Technology holds a KPI / 21 CFR… records the consequence: the exclusion carries no engineering information. The FDA record the Verifier found points the same way: K053434's later configuration carries a stress-incontinence ETS programme. **The rig consequence is already the plan.** B1 The discriminating trigger works in the…'s settled by: says the sensing half starts with a purchase order and an IRB, and that "the representative-probe half still needs 15-25 purpose-built probes". Nothing in this file spends the favourable branch, in which a bought unit runs the whole triggered study. The one half of 148 that might still bite is trigger-event logging. A verified line already says the successor device does not log stimulation: K201014 records RMS EMG and pressure, and "Stimulation data are not recorded" (claim 154verified). So the rig plan should assume no trigger log from a purchased unit whatever the 2005 guide says. Which device to buy, and whether the MyoTrac is still sold, is REQ-0153 Read the full specifications of US6240315… item (b). What survives as a rule: no artifact may cite claim 148unconfirmed as evidence that EMG-triggered stimulation for incontinence is unavailable, unbuildable or uncleared. claim 148unconfirmed describes one 2005 document whose relationship to current labelling is contradicted on the FDA record. A56 The MyoTrac Infiniti User Guide closes as absorbed by IR-45 claim 148.

• The statement: claim 149unconfirmed. The MyoTrac Infiniti records two EMG channels at 2048 samples/s, 14-bit, 20–500 Hz, saves raw EMG to compact flash or streams it by USB (User Guide, 2005), and has a programmable trigger threshold (K053266). So the sensing half of a trigger-specificity study can run on a purchased cleared device without a prototype probe.
• What the check found: the FDA-reachable operands hold. K053266 lists a programmable "threshold for trigger by the patient and physician". The K201014 summary's predicate column gives the MyoTrac Infiniti (K053434) EMG as "14-bit ADC, Bipolar, 2048 samples/s", "20Hz - 500Hz (Band-Pass Filter)", 2 channels. Unchecked: the compact-flash and USB raw-data claims come only from the user guide. "Without building a prototype probe" is a feasibility inference. The Verifier also notes K053266 is IPF / 890.5850; the incontinence indication is K053434.
• Who depended on it: A57 The MyoTrac Infiniti records two channels…: Technology & R&D [C14], its answer to REQ-0058 Candidate claim 103 says EMG-triggered…; Operations & NPI [C4] through A43 Measuring the trigger sensitivity and… (claim 114unconfirmed), whose cheap branch this is; Regulatory [C15] (claim 166unconfirmed), which sets the study's device status; B1 The discriminating trigger works in the…'s settled by:.
• What it costs: if the raw-data half is false, the sensing study cannot run on a purchased device. The first go/no-go becomes Operations' prototype build of 15–25 probes, months later and an order of magnitude dearer. B1 The discriminating trigger works in the…'s cheapest stopping test is the thing at stake. A verified line sharpens the risk: the successor MyOnyx stores RMS EMG at 20 samples/s (claim 154verified), which is not raw EMG and cannot carry an offline detector study.
• Outcome: commission
• Because: the missing fact is a sentence in a public manual, and no request asked for it. REQ-0153 Read the full specifications of US6240315… reads the same manufacturer's specification for trigger timing and current sale, not raw-data export. REQ-0166 Read the manufacturer labelling (front-end → research), item (d), asks whether the MyoTrac Infiniti saves or streams raw EMG, at what rate, with a settable threshold. Item (c) asks the same of K243079 and K201014, so that if the 2005 unit is not sold the alternative rig is already read. It goes to Research, not Technology, because a chair may not settle its own condition. A57 The MyoTrac Infiniti records two channels… stays live.

• The statement: claim 150unconfirmed. Vaginal-probe pelvic floor EMG rises during Valsalva and bracing as well as voluntary contraction (PMID 19730763), and symptomatic women show more pelvic floor EMG during Valsalva (PMID 16496395). So an amplitude threshold on one vaginal channel cannot be assumed to discriminate, and a discriminating trigger needs a reference signal.
• What the check found: both abstracts say what the claim says. PMID 19730763 (nine continent women): "PFM EMG and IAP increased during all tasks", including bracing and Valsalva. PMID 16496395 (13 symptomatic against 13 continent): "During Valsalva, the PFM and EO were more active in the symptomatic group." The tag is held because both are ScanResults. Both studies are small and outside the US. The design conclusion is an engineering inference, and claim 105unconfirmed records that nobody has measured single-channel discrimination at all.
• Who depended on it: A58 Pelvic floor surface EMG recorded with a…: Technology & R&D [C15]; A34 No published study reports the sensitivity… (claim 105unconfirmed), the bench gate; A24 Women who cannot actively contract the… (claim 96unconfirmed); B1 The discriminating trigger works in the…; the frontmatter mechanism:. Registered in A12 Medicare NCD 30.1.1 excludes home use of…'s direction.
• What it costs: if true, the closed loop as framed has no discriminating trigger. The device becomes a two-site system that the woman places correctly herself each session, which is a change to mechanism: and a shape question (open question 3).
• Outcome: commission
• Because: the record supports the premise, that the signal rises on a strain. It does not touch the conclusion, that a single channel cannot discriminate, because the conclusion is a burden of proof and not a measurement. Pivoting mechanism: to two sites on "cannot be assumed" would change the product on an untested inference. REQ-0055 Search for any reported sensitivity or… (technology → research) asks for any reported sensitivity or specificity of a vaginal sEMG detector separating a voluntary contraction from Valsalva or bracing. That is the measurement that turns this burden into a finding in either direction. Behind it, the sensing-only study that A57 The MyoTrac Infiniti records two channels… and A74 A sensing-only study that records vaginal… describe records a vaginal and an abdominal channel together. It therefore tests the one-probe and two-site detectors for the same money. **Recorded now so the next turn cannot meet it fresh:** if REQ-0055 Search for any reported sensitivity or… and that study show single-channel discrimination fails, the review that fires is a pivot review on mechanism:, and open question 3 is its agenda. A58 Pelvic floor surface EMG recorded with a… stays live.

• The statement: two claims, grouped by consequence: both are Option 3 the manometric trigger's case. claim 151unconfirmed: in 13 continent women, vaginal pressure change did not differ between a correct contraction and Valsalva (P = 0.971, PMID 16302270), so a manometric trigger carries the same bench gate as an sEMG trigger. claim 159unconfirmed: no KPI 510(k) summary read in this file describes stimulation triggered from intravaginal pressure. Pressure is cleared as biofeedback only, while EMG triggering is cleared, so a manometric loop sits further from the predicate pool.
• What the check found: for 151, the abstract matches: "there was no difference in the change in vaginal pressure between PFM contraction and Valsalva (P = 0.971)"; "The change in IAP was greater during Valsalva (P = 0.001)". The tag is held because the source is a ScanResult; the step to a device is a design inference. For 159, across nine summaries (K110179, K121820, K133826, K134020, K150180, K201014, K220161, K243079, K053434), pressure appears only as biofeedback, and every trigger is EMG. The negative is bounded by what extracted, because some pages are scanned. "Further from the pool" is a substantial-equivalence characterisation only FDA makes.
• Who depended on it: A59 In 13 continent women the change in vaginal… (151): Technology & R&D [C16]; A36 Closed-loop control contingent on a… (claim 107unconfirmed) and Option 3 the manometric trigger; A45 Replacing the surface-EMG trigger with an… (claim 116unconfirmed); Regulatory [C8] (claim 159unconfirmed); B1 The discriminating trigger works in the…, as the escape from a failed sEMG trigger. A67 No KPI / 21 CFR 876.5320 510 (159): Regulatory [C8], its answer to REQ-0054 Two of the three KPI predicates this file…; A36 Closed-loop control contingent on a… and Option 3 the manometric trigger; A45 Replacing the surface-EMG trigger with an…; Technology [C16]; B8 The route is one 510.
• What it costs: both rows' costly branch is the one the record shows. Option 3 the manometric trigger carries the same bench gate as the EMG trigger and moves away from the cleared pool. What remains of its case is the freedom-to-operate escape at claim 107unconfirmed (pillar B6 No live US patent stops P2 from shipping).
• Outcome: absorb
• Because: Option 3 the manometric trigger sits under ## Pivot options on the table and is not the current shape. Nothing P1 commercially-insured first rests on spends the favourable branch of either row. **What did depend on them, and why that survives.** Option 3 the manometric trigger's argument, read from today, is the FTO argument alone. That argument is B6 No live US patent stops P2 from shipping's and stays live on claim 107unconfirmed, with REQ-0152 Retrieve and read the FDA 510 reading the K150180 indication behind it. A36 Closed-loop control contingent on a… and A45 Replacing the surface-EMG trigger with an… stay live on their own claims, which these rows bound but do not answer. B1 The discriminating trigger works in the…'s escape route through a manometric trigger is gone as an engineering shortcut, and B1 The discriminating trigger works in the… already has no line that spends it. **The rule for any later turn that tables Option 3 the manometric trigger:** it must state that it carries the same bench gate and an un-precedented trigger input, with FTO as its only advantage. A59 In 13 continent women the change in vaginal… and A67 No KPI / 21 CFR 876.5320 510 close as absorbed by IR-48 claims 151 and 159.

• The statement: claim 152unconfirmed. The cleared envelopes of K150180, K110179 and K053266 leave at least 2 s off between bursts, so an inter-burst detector has at least 2,000 ms, about 260 times the 7.6 ms blanking window of PMID 33836509.
• What the check found: the FDA operands hold, read in full: K150180 "1 second at 50 Hz, 2 seconds no stimulation, 2 seconds at 13 Hz, 2 second no stimulation (repeating)"; K110179 "On 20 seconds, Off 10 seconds"; K053266 "Rest Period (sec) 2 - 50 seconds". 2,000 / 7.6 = 263. The 7.6 ms figure is a ScanResult. The Verifier's caution: these three are not the nearest EMG-triggered products. K243079 and K201014 are, and K201014 lists stimulation delivery "Continuous (no rest phase) for EMG-triggered stimulation". So "nearest products" and "any of these envelopes" must be read as these three only.
• Who depended on it: A60 The cleared stimulation envelopes read from…: Technology & R&D [C17]; A29 A closed loop in which stimulation is… (claim 100unconfirmed), whose headroom argument stands on this row; A30 The therapeutic stimulation envelope… (claim 101unconfirmed), which it replaces as the timing base; A31 Real-time recovery of volitional surface… (claim 102unconfirmed); B1 The discriminating trigger works in the….
• What it costs: if the off-periods were shorter, post-burst settling would bind, and an envelope stretched to make room would move away from cleared envelopes.
• Outcome: absorb
• Because: the half A29 A closed loop in which stimulation is… spends was read off three FDA summaries by the Verifier, and holds. The ≥2 s off-period is inside the cleared KPI pool, so an envelope this device chooses on K150180's pattern needs no stretch. The other operand, the 7.6 ms figure, was absorbed at IR-24 claim 102 on the ground that nothing chosen depends on it, and a 260-fold margin survives an order-of-magnitude error in it. What did depend on it, and why that survives. A29 A closed loop in which stimulation is…'s headroom argument survives at the strength of three verified envelopes. A30 The therapeutic stimulation envelope… and A31 Real-time recovery of volitional surface… stay live on their own claims. The caution goes to B8 The route is one 510, not to B1 The discriminating trigger works in the…: if this device's 510(k) leads with K243079 or K201014 (see the predicate_or_analog: correction under ## Open work), its envelope is compared against an ETS predicate whose triggered mode may have no rest phase. That is a predicate-comparison question carried by claim 83unconfirmed, and REQ-0166 Read the manufacturer labelling item (b) reads the two ETS envelopes. No artifact may describe these three as "the nearest products". A60 The cleared stimulation envelopes read from… closes as absorbed by IR-49 claim 152.

• The statement: claim 155unconfirmed. A closed-loop stimulator with this intended use, triggered by a threshold crossing of pelvic floor sEMG sensed before the burst, is substantially equivalent to K243079 on intended use and on the ETS characteristic. So the loop raises no different question and does not by itself need clinical data.
• What the check found: an SE finding is an FDA judgment on a future submission, and no connector answers it. On record: K243079's IFU encompasses this narrower indication (claim 153verified), and K243079 was itself cleared on nonclinical testing alone. Neither summary describes the trigger detector, so equivalence on "the EMG-triggered stimulation characteristic" as this claim specifies it is not shown by the predicate record.
• Who depended on it: A63 A closed-loop intravaginal stimulator with…: Regulatory [C4]; B8 The route is one 510, as the direct counter to claim 130unconfirmed; A2 A device with this intended use is eligible… (claim 6unconfirmed); the Clinical chair's Study 2, which "a KPI 510(k) will not require"; Technology & R&D, because only a pre-burst architecture is covered.
• What it costs: both branches cost something, in different places. If false, the 510(k) needs clinical data or the route becomes De Novo, which is B8 The route is one 510's expensive end and A2 A device with this intended use is eligible…'s 12–18 months. If true, nothing regulatory will ever pay for B1 The discriminating trigger works in the…'s margin study. Verified claim 153verified already points toward true: an ETS KPI device was cleared with no clinical data.
• Outcome: commission
• Because: the claim has a document half that can still be settled, and an FDA half that only a pre-submission settles. The document half is whether K243079's trigger works the way this device's would: what it thresholds, with what hold time, sensed before or during the burst. If K243079 triggers on something else, equivalence on "the characteristic" is a weaker argument than the IFU match suggests. REQ-0166 Read the manufacturer labelling (front-end → research), items (a) and (b), reads the K243079 and K201014 labelling for the detector and the ETS envelope. **The FDA half is not commissioned, because it cannot be.** A pre-submission discloses this candidate outside the organisation, and that is Tier 3 (_shared/one-rule.md). It is recorded here as a human decision the packet must name, as IR-5 claim 6 is unconfirmed and A2 rests on it did for claim 6unconfirmed, not raised as a request. A63 A closed-loop intravaginal stimulator with… stays live.

• The statement: claim 157unconfirmed. Under 21 CFR 801.4, a cleared indication's prior-therapy precondition bounds the manufacturer's labelling and promotion, not who a licensed practitioner may prescribe the device to. 801.109(a)(2) conditions sale on a practitioner's order without reference to prior treatment.
• What the check found: both regulation texts exist and say what is quoted. 801.4: "a firm would not be regarded as intending an unapproved new use … based solely on that firm's knowledge that such device was being prescribed or used by health care providers for such use". The connector's exact-phrase test missed only on wording. 801.109(a)(2): "Is to be sold only to or on the prescription or other order of such practitioner". Not resolved: the head proposition is an interpretation of how FDA applies these sections and of the practice-of-medicine boundary, and regulation text does not settle it.
• Who depended on it: A65 Under 21 CFR 801.4 a cleared indication's…: Regulatory [C6], its answer to REQ-0159 Does a cleared 510; A16 This candidate's intended use (claim 89unconfirmed); A51 Among women referred to pelvic floor… (claim 143unconfirmed), routed here through REQ-0159 Does a cleared 510 and absorbed at IR-43 claim 143 on the ground that this row carries its consequence; B4 Prescribers will send a woman home with a…, which is fatal; Commercial.
• What it costs: if false, the women who never complete PT (claim 143unconfirmed) are outside the market at the point of sale, by regulation, and B4 Prescribers will send a woman home with a…'s documentation half bites at the prescription. If true, the cut is a promotion and payer constraint (B9 Payment reaches this device as billed, Market Access), and a prescriber may prescribe without the documented failure.
• Outcome: commission
• Because: the governing text is public and has not been read. The practice-of-medicine provision is a statute, FD&C Act section 1006 (21 U.S.C. 396). No connector reaches the US Code, and nobody has quoted it in this file. REQ-0168 Retrieve from fda.gov and the US Code (front-end → research), item (e), asks for it and for any FDA text applying it or 801.4 to a prior-therapy precondition. It goes to Research and not back to Regulatory, because the claim is that chair's own condition. A65 Under 21 CFR 801.4 a cleared indication's… stays live, and it is the row IR-43 claim 143 handed A51 Among women referred to pelvic floor…'s consequence to.

• The statement: four claims, all Regulatory's answers to requests, grouped because they share a finding and an outcome. claim 160unconfirmed (C9): under FDA's 2017 510(k)-for-a-change guidance, an EMG-gating module that directs a host with no cleared trigger input is a control-mechanism change to the host. claim 161unconfirmed (C10): FDA's 2023 ISO 10993-1 guidance recommends testing in the condition of use and says nothing about ageing to shelf life. claim 162unconfirmed (C11): repeated home use over twelve weeks or more puts the probe above the under-24-hour mucosal category K150180 used. `claim 164unconfirmed` (C13): premarket software and cybersecurity content is set by FDA guidance and section 524B, not by Part 820, and K243079 documents against those guidances.
• What the check found: every operand a connector reaches was checked, and held. 21 CFR 807.81(a)(3)(i) contains "significantly affect the safety or effectiveness" (160). The K150180 summary says "tissue contacting for a duration of less than 24 hours" with "Cytotoxicity, Sensitization, Irritation" (162). The K243079 summary documents software V&V and cybersecurity against the named FDA guidances (164). **What carries each proposition was not reached.** That is the 2017 change guidance (160), the 2023 ISO 10993-1 guidance including a negative about what it omits (161), the same guidance's treatment of repeated contact (162), and the software guidances, section 524B and whether QMSR touches premarket content (164). No typed connector serves FDA guidance or the FD&C Act.
• Who depended on it: A68 FDA's guidance 'Deciding When to Submit a… (160): Regulatory [C9]; A39 A closed-loop EMG-gating module can be an… (claim 110unconfirmed); B3 The discriminating trigger is a difference…'s OEM-module shape; Commercial through claim 51unconfirmed; and B9 Payment reaches this device as billed's class argument, which names the OEM module as one of three shapes that keep B9 Payment reaches this device as billed reshaping. A69 FDA's guidance 'Use of International… (161): Regulatory [C10]; A44 ISO 10993-1 biological evaluation of a… (claim 115unconfirmed); claim 36unconfirmed; Operations & NPI [C5]; B1 The discriminating trigger works in the…. A70 Because this candidate's intravaginal probe… (162): Regulatory [C11]; claim 36unconfirmed; A44 ISO 10993-1 biological evaluation of a…; Operations [C5]; Technology's "known list" reading; B1 The discriminating trigger works in the…; registered in A12 Medicare NCD 30.1.1 excludes home use of…'s direction. A72 The software and cybersecurity… (164): Regulatory [C13]; Technology & R&D, whose estimates REQ-0056 Candidate claim 64's re-verification of… says all predate QMSR; claim 64unconfirmed; the clinician data path; B8 The route is one 510.
• What it costs: stated separately, because the costs differ. 160: with claims 153 and 154 now verified, hosts that already have a cleared trigger need no module. If 160 holds, hosts without one each need a new 510(k). The OEM shape then has close to no buyer on either branch, and B9 Payment reaches this device as billed loses a leg of the argument that keeps it reshaping (open question 4). 161: if false, an accelerated-ageing leg sits ahead of the biocompatibility battery, and no schedule carries it. 162: if true, the probe's battery carries endpoints beyond the three no predicate's battery can lend, and Technology's "known list" is wrong in the expensive direction. 164: if false, software estimates are rebuilt rather than re-framed. If true, 524B turns on whether the app reaches the internet, a design fact nobody has fixed.
• Outcome: commission
• Because: every one of these is a gap in the instrument with a public text behind it, so none is a finding about the world. The texts are on fda.gov and in the US Code. REQ-0168 Retrieve from fda.gov and the US Code (front-end → research) reads them: items (a) for 160, (b) for 161, (c) for 162 and (d) for 164, each quoting the sentence or reporting "not stated" after a full-text search. It goes to Research, not back to Regulatory, because these are Regulatory's own conditions. The request asks that interpretive residuals FDA alone can settle (for 160, whether a module is a change to the host or an accessory) be reported as residuals, not resolved. **The B9 Payment reaches this device as billed class question is not regraded here.** It waits on 160, and it is the next assessment pass's ## Regrades row whatever REQ-0168 Retrieve from fda.gov and the US Code returns. A68 FDA's guidance 'Deciding When to Submit a…, A69 FDA's guidance 'Use of International…, A70 Because this candidate's intravaginal probe… and A72 The software and cybersecurity… stay live.

• The statement: claim 163unconfirmed. The QMSR amendment (89 FR 7496, FR doc 2024-01709) took effect 2026-02-02. Design transfer is documented under amended Part 820 incorporating ISO 13485:2016. FDA states the two are substantially similar in totality, so existing activity lists need re-mapping, not new activities.
• What the check found: the documentary operands verify. 2024-01709 is 89 FR 7496, a Rule, "This rule is effective February 2, 2026". It states "the requirements in ISO 13485 are, when taken in totality, substantially similar to the requirements of the QS regulation". 21 CFR 820.7 incorporates ISO 13485:2016 by reference. Not resolved: whether existing lists need only re-mapping is a reading of what the amended part requires in practice.
• Who depended on it: A71 The QMSR amendment to 21 CFR 820: Regulatory [C12], its answer to REQ-0057 The production half of REQ-0056; Operations & NPI, whose design-transfer, process-validation and supplier-control activity lists are at stake; B8 The route is one 510's schedule. Like A57 The MyoTrac Infiniti records two channels…, the row is about the cost of a step, not whether the route exists.
• What it costs: if the residual is false, Operations' lists need new activities, and "every transfer estimate built on the old lists is low".
• Outcome: absorb
• Because: the costly branch has nothing to cost. Operations' conditions C1–C10 state activities without schedules, and REQ-0057 The production half of REQ-0056, the request this claim answers, says so: they were written that way for exactly this reason. No transfer estimate in this file was built on the old lists, so none is low. Everything the residual would let an estimate rest on came back verified: the date, the incorporation and FDA's own "substantially similar" sentence. The clause-by-clause re-map is a gap analysis that design transfer performs when a quality system exists. It is work, not a statement any source will tag, and commissioning it as a lookup would buy a permanently unconfirmed line. What survives: Operations costs its activities against ISO 13485 as incorporated, using FDA's similarity statement as the basis for re-use. The first schedule anyone writes must carry the gap analysis as a line item, not assume it away. A71 The QMSR amendment to 21 CFR 820 closes as absorbed by IR-53 claim 163.

• The statement: claim 166unconfirmed. A sensing-only study recording vaginal and abdominal sEMG on a cleared KPI/HCC device in women under incontinence treatment, with detectors applied offline, uses the device within its labelling (K053434) and is exempt from 21 CFR 812 under 812.2(c)(2). The same study in continent volunteers needs an IRB determination under 812.2(b).
• What the check found: the labelling phrase is in the K053434 IFU: "It is also indicated during incontinence treatment for assessing EMG activity of the pelvic floor and accessory muscles such as the abdominal or gluteal muscles". 812.2(c) exempts investigations of SE devices used per labelled indications. Not resolved: whether an offline-detector study "uses the device within its cleared labelling", and what a continent-volunteer study needs, are FDA and IRB determinations. The Verifier also notes the claim's "IRB significant-risk determination" wording is loose; 812.2(b) turns on a non-significant-risk finding.
• Who depended on it: A74 A sensing-only study that records vaginal…: Regulatory [C15]; Technology & R&D [C14] (A57 The MyoTrac Infiniti records two channels…, claim 149unconfirmed); the Clinical chair's Study 1, which wants Oxford 0–2 women; Operations & NPI through REQ-0058 Candidate claim 103 says EMG-triggered…; B1 The discriminating trigger works in the…'s settled by:. Like A57 The MyoTrac Infiniti records two channels…, it is about how B1 The discriminating trigger works in the…'s cheapest test is run.
• What it costs: if false, even the sensing-only study needs an IRB risk determination or more, which adds weeks to months in front of the bench gate.
• Outcome: absorb
• Because: the determination is made by the IRB that reviews the protocol. B1 The discriminating trigger works in the…'s `settled by:` already puts that IRB in front of the study ("a purchase order and an IRB"), so the step happens on either branch. The study population already chosen, women under incontinence treatment, is the one inside the labelled use. The continent-volunteer branch is in no plan. No document lookup can pre-empt an IRB's finding, so a commission would buy a re-run that cannot move the tag. What did depend on it, and why that survives. The sensing study's design survives unchanged. Its schedule carries an IRB determination either way, and the row's only cost is how long that takes. The labelling argument depends on the device bought being one whose IFU carries the EMG-assessment use. K053434 does; whether it is still sold is REQ-0153 Read the full specifications of US6240315… item (b). K243079's IFU also names "assessing EMG activity of the pelvic floor" (claim 153verified). What survives as a rule: no artifact describes the study as "exempt". It is a study whose population was chosen to sit inside a cleared labelled use, pending the IRB's determination. A74 A sensing-only study that records vaginal… closes as absorbed by IR-54 claim 166.

• The statement: claim 167refuted. No prescription device offered for sale in the US as of October 2026 delivers intravaginal stimulation for incontinence triggered by the patient's own pelvic floor sEMG. In particular, neither the NeuroTrac MyoPlus Pro (K243079) nor the MyOnyx (K201014) is marketed in the US with that mode. So closed-loop triggering is a feature no US-marketed incumbent prescription device sells.
• What the check found: no connector reaches commercial availability. A clearance says a device may be marketed, never that it is offered for sale or which modes ship, and a negative across the US market is not answerable from any registry here. Not refuted. Four cleared products carry an ETS mode, all read in full this pass: K243079 (Rx, home use, ETS through a vaginal probe; claim 153verified); K201014 (Rx, ETS for incontinence; claim 154verified); K201290 (Medline DeNovo 4Pro, primary code IPF, IFU includes an "EMG Triggered Stimulation (ETS) mode (nonimplanted electrical continence device only)" for stress, urge or mixed UI); K220161 (Shenzhen Konmed, KPI, ETS mode, labelled OTC and so outside the claim's "prescription"). In the Verifier's words: if any of the prescription ones is sold in the US with the mode, the claim is false.
• Who depended on it: A75 No prescription device offered for sale in…: the front end, which drafted it; pillar B3 The discriminating trigger is a difference…, restated on 2026-10-01 to carry it; A28 Closed-loop EMG-gated stimulation produces… (claim 33unconfirmed) and A19 No randomised comparison of closed-loop… (claim 91unconfirmed), whose wording treats the closed loop as what separates this device from the cleared pool; Commercial through REQ-0163 Two cleared KPI / 21 CFR 876.5320 devices…; Regulatory, whose claims 153 and 154 raised it; Legal & IP through REQ-0165 FDA-posted 510. Registered in A12 Medicare NCD 30.1.1 excludes home use of…'s direction.
• What it costs: if false, the bet's comparison ("enough better than the existing open-loop devices") is against the wrong incumbents. The device is a me-too on its central feature, and A28 Closed-loop EMG-gated stimulation produces…'s grading reason for B1 The discriminating trigger works in the… ("a covered, cleared, already-marketed therapy with an app on it") is true even if B1 The discriminating trigger works in the… succeeds.
• Outcome: commission
• Because: taken on the merits, the case for a pivot today was weighed and not taken, for three reasons.
  1. The tag is a gap in the check, not a finding about the world. Nobody has looked for a US listing. Pivoting on an unconfirmed would put the front end's expectation where the Verifier's tag belongs.
  2. The answer decides which shape the pivot would be. If an incumbent sells ETS through a prescription home channel, the pivot has to move toward whatever that incumbent does not offer: the data leg (B3 The discriminating trigger is a difference…), the rental channel and coding (B2 What the manufacturer collects per…, B9 Payment reaches this device as billed), a different user, or Option 2 the adherence device rather than the…'s adherence device. Choosing among those without knowing the incumbent's feature set, price and channel would be a redesign that cites a review. REQ-0163 Two cleared KPI / 21 CFR 876.5320 devices… asks that exactly: "what does this candidate offer a prescriber or payer that they do not". So this commission is the input to the shape decision, not a deferral of it. It is also the cheapest test on B3 The discriminating trigger is a difference…, a sales-channel read in the cheap band.
  3. What the verified claims already settle does not need this tag, and is recorded now. Claims 153 and 154 are verified. The closed loop is therefore not a regulatory differentiator: two manufacturers hold clearances for it through a vaginal probe, one for home use on prescription. Those holders can market the mode without a new submission. So even on 167's favourable branch, the closed loop is a first-to-market lead at best, and no clearance protects it. B3 The discriminating trigger is a difference…'s "defensible" half cannot rest on the closed loop, and B3 The discriminating trigger is a difference…'s proposition already says so ("rather than … an incumbent already ships or a fast-follower can replicate"). From today, no artifact may describe the closed loop as new to the cleared pool or as what distinguishes this device from what is cleared. A19 No randomised comparison of closed-loop… and A28 Closed-loop EMG-gated stimulation produces…'s wording predates claims 153 and 154 and is read with this sentence beside it. The requests. REQ-0163 Two cleared KPI / 21 CFR 876.5320 devices… (regulatory → commercial) settles the two prescription devices the claim names. REQ-0167 REQ-0163 asks whether K243079 (front-end → commercial) covers the two the Verifier added and REQ-0163 Two cleared KPI / 21 CFR 876.5320 devices… does not name: K201290 and K220161. For K220161 it also covers the OTC case the claim's wording leaves outside it. An OTC ETS device on sale in the US would hit B3 The discriminating trigger is a difference…'s incumbent half whatever claim 167refuted says, and if one is found the front end drafts that proposition through the appender. **Decided now, so this is the last commission on this row and not the first of a series.** If either request returns a US listing of a prescription ETS device with the mode, the Verifier can tag 167 refuted, and the review that fires then is a pivot review on differentiation, not a second commission. If both return "not found", 167 stays unconfirmed, because absence across a market cannot be verified, and the next review absorbs or pivots on point 3 above. It does not re-commission. A75 No prescription device offered for sale in… stays live.

• The statement: claim 167refuted. No prescription device offered for sale in the US as of October 2026 delivers intravaginal stimulation for incontinence triggered by the patient's own pelvic floor sEMG, and in particular neither the NeuroTrac MyoPlus Pro (K243079) nor the MyOnyx (K201014) is marketed in the US with that mode. Register row A75 No prescription device offered for sale in…, pillar B3 The discriminating trigger is a difference…'s incumbent half.
• What the check found: The Verifier re-attempted the claim cold on 2026-10-01 and refuted it on one counterexample. The claim is a universal negative, so one is enough.
  • Listed for US commercial distribution. The FDA Establishment Registration & Device Listing record (21 CFR Part 807) for K201014 shows Thought Technology Ltd with an active registration and a US agent. It lists KPI and HCC from 2021-11-05 under "SA9020 - MyOnyx", "SA9030 - MyOnyx Mobile App" and the vaginal pressure probe.
  • Offered for sale in the US, prescription only. The manufacturer's Incontinence Expert System page says "the MyOnyx unit is only available for sale in the USA" and carries the federal Rx legend. It quotes $3,645.00 with a cart.
  • The mode is the one the claim names. The app runs "1 or 2 channels of surface EMG biofeedback or EMG triggered STIM". The page offers an "SA9572 - Vaginal Sensor - STIM & EMG", used "to stimulate or sense the pelvic floor muscles".
  • For incontinence. The K201014 IFU covers stress, urge or mixed incontinence. The weakest link, as the Verifier stated it: no single sentence on the page says the triggered mode runs through the vaginal sensor. That is read from a dual STIM & EMG vaginal sensor sold in the same incontinence bundle as the ETS-capable unit. The other three ETS products are leads and not part of the refutation. K243079 has an active registration with KPI listed 2026-07-10. Verity's own page reads "Single channel ETS (EMG triggered stimulation)", with vaginal probes and "prescribed home users", but shows no US offer of sale. K201290 is listed by Medline, with no product page read. K220161 is listed and is OTC.
• **Whether the weakest link leaves room to absorb or commission. It does not, for three reasons taken in order.**
  1. The gap is already filled by a verified claim. claim 154verified reads the K201014 summary cold. It lists "EMG-Triggered Stimulation (ETS) using NMES specifications for incontinence" and names the Life-care vaginal probe as both the EMG electrode and the stimulation probe. The configuration the page does not spell out is the cleared configuration of the device the page sells. What is left open is a MyOnyx that ships ETS but runs it only through skin electrodes. That is the branch an absorb would have to rest on, and nothing in this file points to it.
  2. This is a finding about the world, not about our lookup. The two sources are a federal listing and the manufacturer's own US sale page, and both say what the Verifier quoted. The listing was read straight from the API because no typed connector covers it (backlog 1.6), so the result cannot be replayed by machine. That is a reproducibility caveat on the instrument. It is not a reason to think a second look would reverse the result. The one exception the protocol allows, a refutation that is really about our lookup, does not apply.
  3. Even the most favourable reading leaves nothing load-bearing standing. claim 167refuted was never worded about the home. This file does not establish whether the MyOnyx is a clinic or a home system, so one could argue the home triggered niche is still unsold: K243079 is cleared for prescription home use claim 153verified and listed, but its US offer is not shown. An absorb built on that would rest on the front end's guess that Verity is not shipping a listed device into the US, which is a probability judgement and not this department's to make. On that branch too, IR-55 claim 167 point 3 already holds and does not need this tag: any of the clearance holders can market the mode without a new submission. So a home first-mover lead would be unprotected and temporary.
• Who depended on it: A75 No prescription device offered for sale in…: the front end, which drafted it; pillar B3 The discriminating trigger is a difference…, restated on 2026-10-01 to carry it and now rendered holed / blocking with posture repair-pending. A28 Closed-loop EMG-gated stimulation produces… (claim 33unconfirmed) and A19 No randomised comparison of closed-loop… (claim 91unconfirmed), whose consequence columns treat the closed loop as "the one feature that distinguishes this device" from the cleared pool. **The bet's comparison** in the assessment, "enough better than the existing open-loop devices". **The viability deck's load-bearing condition** (Slide 9), which named claim 167refuted as the one thing to check first because "it decides what kind of company this is". Behind them, by chair: Commercial through REQ-0163 Two cleared KPI / 21 CFR 876.5320 devices… and REQ-0167 REQ-0163 asks whether K243079; Regulatory, whose claims 153 and 154 raised the question; Legal & IP through REQ-0165 FDA-posted 510; and Clinical & Scientific, whose Study 2 (gated against open-loop, n≈120, $1.5-3M, IR-11 claim 33 is unconfirmed) prices the comparison the bet makes.
• What it costs: specifically, six things.
  • The bet's comparison is now against the wrong incumbents. A prescriber who wants EMG-triggered intravaginal stimulation for incontinence on prescription can already buy it. "Better than open-loop" no longer describes a reason to choose this device.
  • A28 Closed-loop EMG-gated stimulation produces…'s grading reason for B1 The discriminating trigger works in the… is now true even if B1 The discriminating trigger works in the… succeeds. "What is left is a covered, cleared, already-marketed therapy with an app on it." That is the device as framed at P1 commercially-insured first on B1 The discriminating trigger works in the…'s best day.
  • B3 The discriminating trigger is a difference…'s data leg does not stand in for the lost half. The MyOnyx already records RMS EMG and pressure on a tablet through its app claim 154verified. Collecting the EMG stream is something a sold incumbent already does, and B3 The discriminating trigger is a difference…'s data leg was always about whether anyone pays for the stream, which is still claim 14unconfirmed, claim 133unconfirmed.
  • **The viability deck's moat (its Slide 5) and share (its Slide 3) were argued against an absent incumbent.** The deck's own Slide 9 says both "have to be re-argued against a shipped product". The deck is not rebuilt here.
  • Clinical's Study 2 is priced against the comparison that no longer decides. A trial of gated against open-loop stimulation measures the category the MyOnyx already sells. It does not measure whether this device beats the MyOnyx.
  • The OEM-module neighbour shape loses its buyer. claim 51unconfirmed assumed incumbents lacking gating would license it. Two of them hold clearances for it, and one sells it (see also A68 FDA's guidance 'Deciding When to Submit a…).
• Outcome: pivot
• Because: taken on the merits, and not because IR-55 claim 167 said so. IR-55 claim 167 decided in advance that a US listing of a prescription ETS device would make the next review a pivot review. That advance decision was right to make, and it is the reason this review cannot drift into a third commission. It does not make the pivot correct. What does is the three findings above. The tag is a finding about the world, so commission is the wrong route. The one gap the Verifier names is filled by a verified claim. And absorb needs a dependant that survives, but the bet's comparison, B3 The discriminating trigger is a difference…'s incumbent half and A28 Closed-loop EMG-gated stimulation produces…'s grading reason all fail on this tag. Absorb would also need a probability judgement about Verity's US shipping, which this department may not make. What the pivot answers is the specific failure: the closed loop is no longer new. A shape that answers it has to name something the sold triggered devices do not do. A shape that only moves the business somewhere else (a channel, a payer, a setting) leaves the failure where it is. The shape chosen, the runners-up and the reasons are recorded at P2 the discriminating trigger under ## Pivots. Register row A75 No prescription device offered for sale in… is superseded by P2 the discriminating trigger. claim 167refuted itself is not superseded. It is a refuted premise, and that is the most decision-relevant line in the candidate file. It stays exactly as it is, as the record of why P2 the discriminating trigger exists.

The 2026-10-01 P2 the discriminating trigger batch — thirty-seven rows, twenty reviews, and why they are grouped that way

The Verifier ran claims 168–208 on 2026-10-01, plus claim 85unconfirmed again. These are the P2 the discriminating trigger restatement conditions and the front end's own P2 the discriminating trigger premises. Three came back verified: 184, 193 and 202. One came back refuted: claim 191refuted. Every other register row resting on the batch came back unconfirmed. check_loop.py failed this record on thirty-seven identical lines: A76 A prescribing urogynecologist or pelvic…–A79 A prescribing urogynecologist or pelvic…, A81 A trigger detector that discriminates a…–A90 Under the P2 mechanism the abdominal…, A92 Pre-gelled self-adhesive cutaneous…–A99 If a detached, A101–A108, A110–A115 and A116 (claim 85unconfirmed, registered after its answer and said so in the row). Twenty reviews, IR-57 claim 191 came back refuted to IR-76 claim 85, answer all thirty-seven. Each was written from the Verifier's note on the claim line, not from the tag.

The routing rule, and how often it was overridden. The one refuted tag is a finding about the world, and it lands on the branch its own register row called favourable, so it is absorbed (IR-57 claim 191 came back refuted). For the thirty-six unconfirmed tags the default is commission, and that default is not applied blindly. Where the Verifier re-read the source cold, found the content as written, and held the tag only because the instrument has no path to verified, a commission would buy a re-run that cannot come back different. Where the dependants also already run on the branch the record shows, nothing waits on the tag. Seven rows are absorbed on that ground or on IR-57 claim 191 came back refuted's (IR-57 claim 191 came back refuted, IR-61 claim 175, IR-62 claims 177 and 203, IR-71 claim 182, IR-72 claims 185 and 186). Thirty are commissioned (IR-58 claim 173–IR-60 claim 207, IR-63 claims 176–IR-70 claims 181, IR-73 claims 195–IR-76 claim 85). Nine requests were raised: REQ-0190 Read device labelling no connector reaches… to REQ-0198 Retrieve and quote. Every other commission names a request that was already open. No pivot. pivot: stays at P2 the discriminating trigger.

The groupings are stated so they can be argued with:

• claim 173unconfirmed (IR-58 claim 173) and claim 187unconfirmed (IR-59 claim 187) are not grouped, although both read the same four summaries. 173 is a finding that the summaries are silent. 187 is a lookup fault: a clearance-summary --expect-indication match on K201014 hit the predicate's IFU quoted in the comparison table. They are settled by different reads and owed different numbers of commissions.
• IR-62 claims 177 and 203 groups claims 177 and 203. They have the same finding: the physiology was read cold and matches, and the decision-rule inference is no source's. They have the same consequence: a relational detector, which Technology already specifies. **claim 175unconfirmed (IR-61 claim 175) has the same kind of finding and is kept apart.** Its consequence is the size of P2 the discriminating trigger's population, not the shape of its detector, and one operand outran its source.
• IR-63 claims 176 groups four absence claims about the evidence: 176, 178, 179 and 204. They have the same finding: a non-existence claim no connector can verify, with lead sweeps that found no counter-example. They have the same outcome: a search request. Their costs differ and are stated separately.
• IR-64 claims 168 groups three stakeholder-behaviour claims: 168, 174 and 170. All three are structurally unconfirmed: "no primary research attached". Two requests answer them, one per population asked.
• IR-66 claims 188 groups four Regulatory conditions: 188, 189, 190 and 192. Each had its CFR operand checked, and each rests on FDA guidance or an FDA judgment no connector serves. The costs differ and are stated separately.
• IR-70 claims 181 groups claims 181, 183 and 208. They are the three operands of one number: the per-session electrode's cost under a fixed $74.52 monthly rental.

• The statement: claim 191refuted. Regulatory [C20] wrote it. No cleared home-use ETS summary read in this file shows a patient-placed skin electrode as an input to the stimulation decision. K243079 is "the only one labelled for home use", lists only the vaginal probe and lists no usability testing. So self-placement of the abdominal reference is a use task this 510(k) cannot borrow from a predicate.
• What the check found: the claim fails on a stated fact, not on its conclusion.
  • **K220161 (Shenzhen Konmed, KPI/HCC, ETS for stress, urge and mixed UI, cleared 2022-07-20) is cleared for Over-The-Counter use on its Form 3881.** Its summary lists IEC 60601-1-11 testing ("used in the home healthcare environment") and a "Usability Study according to IEC 60601-1-6 … and IEC 62366-1".
  • K243079's own comparison table records its predicate K201290 as "Home Use: Yes", although K201290's own summary describes clinic use under supervision.
  • K220161 states identical technology to K202648. K202648's summary says it is "supplied with biofeedback reference lead wire with skin electrodes", connected with the vaginal electrode for dual-channel EMG. That is a skin electrode in the EMG input chain of an ETS device sold OTC. It reads as an EMG reference electrode, not an abdominal decision channel.
  • The other operands hold: K243079's only applied part is the vaginal probe and it lists no usability testing; K201014 and K201290 list IEC 60601-1-6 and IEC 62366. In the Verifier's words, the "cannot borrow" conclusion is "weakened, not reversed".
• Who depended on it: A98 No cleared home-use EMG-triggered…: Regulatory [C20], and the usability engineering file and its cost. A81 A trigger detector that discriminates a… (claim 172unconfirmed), self-placement. A112 (claim 205unconfirmed), misplacement as the residual risk. A99 If a detached (claim 192unconfirmed), whether reference placement is a critical task. Clinical and pillar B5 Women will start an intravaginal home….
• What it costs: nothing P2 the discriminating trigger was spending. A98 No cleared home-use EMG-triggered…'s consequence column reads both ways. If the claim is true, the usability file is "new work on the critical path" with "no precedent to borrow". If false, "P2 the discriminating trigger can lean on a predicate's usability work". The record now shows a cleared OTC home ETS device with a 62366-1 usability study and a lay-placed skin electrode on its EMG lead. That is a precedent for the kind of task, not for the abdominal decision task itself. One thing it does cost, and that thing is carried elsewhere: home triggered stimulation is broader than this file assumed. It includes an OTC device, which is a competitor and not only a precedent. That edge bears on P2 the discriminating trigger's differentiation (claim 173unconfirmed, IR-58 claim 173) and on whether the OTC device is sold here (REQ-0184 For each of K243079).
• Outcome: absorb
• Because: this is a finding about the world. It came from FDA summaries read in full, not from our lookup, so commissioning would re-litigate it. Nothing about P2 the discriminating trigger's shape changes, because it lands on the branch the row called favourable. **What depended on it, and why each survives.**
  • A81 A trigger detector that discriminates a… asks whether a woman can place the reference. 191 was only about whether a predicate's paperwork can be borrowed, so A81 A trigger detector that discriminates a… is untouched and stays commissioned (IR-65 claim 172).
  • A99 If a detached asks whether placement is a critical task needing summative validation. That is a guidance question (REQ-0183 Retrieve from fda.gov and quote), and 191's tag does not move it.
  • A112's fail-safe argument is engineering, unaffected.
  • The usability file's cost can only fall. Every 510(k) carries its own usability file, so "cannot borrow" was always partly true. What changed is that the task now has a cleared analog. The rule for every artifact from today: K243079 is not the only home-labelled ETS device. K220161 is cleared OTC with a home-environment and usability record, so any sentence describing P2 the discriminating trigger's home reference as a use task without precedent is withdrawn. claim 191refuted stays exactly where it is, as the refuted line it is. A98 No cleared home-use EMG-triggered… closes as absorbed by IR-57 claim 191 came back refuted. Regulatory is not stale, but its [C20] reads against a refuted claim and is worth the chair re-stating.

• The statement: claim 173unconfirmed. None of the four cleared ETS products (K243079, K201014, K201290, K220161) uses an abdominal, IAP or other reference channel in deciding when to fire. Each fires on the pelvic floor EMG alone, including the MyOnyx even though its app runs one or two channels. This is P2 the discriminating trigger's differentiation premise, drafted the day P2 the discriminating trigger was taken.
• What the check found: the Verifier read all four summaries and K202648. **None describes a reference channel in the fire decision.**
  • Only two describe the decision at all. K220161 fires "when the level of active contraction of the pelvic floor muscles reaches a threshold". K201290 fires "As soon as the EMG activity reaches a target threshold". claim 202verified carries both.
  • K243079 and K201014 are silent on their trigger logic, and a silent summary cannot show what a device does not do.
  • Two findings cut against reading silence as absence. K201014 has two EMG channels, both usable for ETS. The K201014 table credits its predicate K053434 with an IFU for assessing abdominal and gluteal EMG.
  • A third cuts closer. K220161 is identical to K202648, which ships a "biofeedback reference lead wire with skin electrodes" for dual-channel EMG. That reads as a standard EMG reference (common) electrode, not a decision channel, but it is a skin electrode in the input chain of an OTC ETS device. "Not refuted: nothing read describes an abdominal or IAP veto. To settle it, read the labelling or user manuals."
• Who depended on it: A78 None of the four cleared products whose 510: the front end, and P2 the discriminating trigger itself, whose mechanism: names the reference channel as what separates this device from the sold triggered devices. Pillar B3 The discriminating trigger is a difference…'s incumbent half. Technology & R&D, whose two-channel shape this is. Commercial, through REQ-0163 Two cleared KPI / 21 CFR 876.5320 devices…'s "what does this candidate offer that they do not". A106 and A94 The abdominal surface-EMG sensing that…, which fail with it if an incumbent already does it.
• What it costs: if false, the discriminating trigger is a second feature an incumbent already sells. P2 the discriminating trigger is then a me-too for the second time, and nothing on the table differentiates on the loop (A78 None of the four cleared products whose 510's column). That is IR-56 claim 167 came back refuted again, one feature further along.
• Outcome: commission
• Because: the tag is a gap in the check: two summaries are silent, and labelling is the document that is not. Pivoting on silence would put the front end's expectation where the Verifier's tag belongs. REQ-0166 Read the manufacturer labelling item (a), already open, reads how the K243079 and K201014 triggers decide to fire, and this includes the MyOnyx second-channel question. REQ-0190 Read device labelling no connector reaches… item (a), raised today, reads the two products REQ-0166 Read the manufacturer labelling does not name. For K220161/K202648 it asks what the reference lead actually does, and it also covers K201290's manual. Decided now, as IR-55 claim 167 decided for claim 167refuted, so this is the last commission on this row.
  • If either request returns labelling showing any of the four using a second channel's signal to withhold or permit stimulation, the Verifier can tag 173 refuted. The review that fires then is a pivot review on differentiation, not a second commission.
  • If the labelling shows K202648's skin lead is an EMG reference electrode only, that device stays outside the refutation.
  • If labelling is silent, 173 stays unconfirmed, because an absence across labelling cannot be verified. The next review then absorbs or pivots on what is on the record: summary-level single-threshold logic, claim 202verified. It does not re-commission. A78 None of the four cleared products whose 510 stays live.

• The statement: claim 187unconfirmed. Regulatory [C16] wrote it. Abdominal surface-EMG sensing is already a labelled function of cleared KPI/HCC devices: the IFUs of K053434, K201014 and K201290 each include assessing EMG of "pelvic floor and accessory muscles such as the abdominal or gluteal muscles". K201014 lists Uni-Gel surface EMG skin electrodes and two EMG channels, and K243079 lists two EMG channels. None of the four summaries describes an abdominal channel that withholds or permits ETS.
• What the check found: most operands hold. K053434 and K201290 carry the accessory-muscle sentence. K201014 lists T3425 and two channels, and K243079 lists "For EMG: 2". **None of the four describes an abdominal channel gating ETS**, so the sentence that could break P2 the discriminating trigger holds at summary level.
  • The K201014 operand does not hold as written. The MyOnyx IFU in its summary has no accessory-muscle sentence. The sentence appears only in the comparison table, quoted as the predicate's IFU.
  • The connector result that "verified" it was a false positive. `clearance-summary --expect-indication` ran its IFU extraction into the comparison table. The formal Form 3881 page has no text layer, so the subject IFU was never read.
• Who depended on it: A94 The abdominal surface-EMG sensing that…: Regulatory [C16]. A78 None of the four cleared products whose 510 (claim 173unconfirmed), which this bears on directly. A106 (claim 199unconfirmed), because a summary describing a reference-gated trigger is prior art. A109 (claim 202verified). Commercial's reason to abstain.
• What it costs: if the K201014 operand is simply false, the reference channel is slightly less "ordinary" than Regulatory said. That helps novelty and weakens one of three precedents for the regulatory argument. The two other IFUs carry that argument.
• Outcome: commission
• Because: this is the protocol's named exception: a failure about our lookup, not about the world. The Verifier said so ("Do not cite that result"). One commission: REQ-0190 Read device labelling no connector reaches… item (b) reads the K201014 Form 3881 IFU by OCR or from posted labelling. It is decided now that this is the only one. If the IFU carries the sentence, the operand holds. If it does not, the Verifier can tag accordingly, and the next review absorbs: the regulatory argument stands on K053434 and K201290, and nothing in P2 the discriminating trigger rests on the MyOnyx IFU wording. **The connector fault is recorded for the toolsmith, not fixed here.** clearance-summary --expect-indication can match a predicate's IFU inside a comparison table, so any past verified on a 510(k) whose IFU page has no text layer is suspect. A fix is an engine/ change routed through map/effects/CONTEXT.md.

• The statement: claim 207unconfirmed. Technology [C23] wrote it. K201014 lists ETS delivered "Continuous (no rest phase)", and a stress ETS program of 45 Hz, 200 µs, 6 s work and 10 s rest. So once a burst fires it runs its work time with no off-interval, and a pre-burst detector cannot withhold on a Valsalva that starts mid-burst.
• What the check found: the quoted figures match K201014. The summary does not say what "Continuous (no rest phase)" means for an ETS burst, and the reading "no off-interval inside the work time" sits awkwardly beside the same program's 10 s rest. The 45–50 Hz range and the link to claim 102unconfirmed's blanking problem are not in K201014.
• Who depended on it: A114: Technology [C23], and the mechanism: wording, which was narrowed to a pre-burst decision partly because of it. A31 Real-time recovery of volitional surface… (claim 102unconfirmed). A29 A closed loop in which stimulation is… (claim 100unconfirmed). Clinical through REQ-0181 Is an intravaginal NMES burst delivered…, and Regulatory [C21] (A99 If a detached).
• What it costs: if K201014's bursts do have an off-interval or are short, the mid-burst gap is smaller than stated and the narrowing was more than needed, at no cost. The expensive branch is not this tag: it is REQ-0181 Is an intravaginal NMES burst delivered… saying a burst during a strain is a harm. That would make a mid-burst abort a requirement and a change of shape (Open work, item 3).
• Outcome: commission
• Because: the gap is one labelling sentence. REQ-0166 Read the manufacturer labelling item (b), already open, asks for exactly this: the ETS work and rest times, ramps, and "whether any stimulation-off period exists inside a triggered cycle (K201014's summary says 'Continuous (no rest phase)…')". No new request. As A114's row records, the mechanism: correction does not rest on this tag, because P2 the discriminating trigger decides before each burst whatever K201014 does. The decision that would rest on something is REQ-0181 Is an intravaginal NMES burst delivered…'s, and it is open with Clinical.

• The statement: claim 175unconfirmed. Clinical [C13] wrote it. Incorrect technique is documented among untaught incontinent women at a size relevant to a trigger. After brief verbal instruction, 12 of 47 women (25%) made a Kegel effort that could promote incontinence (PMID 1872333). On initial instruction, 30% of 60 incontinent women depressed the bladder neck on an attempted lift, against 17% of 60 continent women (PMID 16532264).
• What the check found: a cold re-retrieval matches every number. One discrepancy: the PMID 1872333 abstract does not say the 47 were incontinent, and the claim calls them that. PMID 16532264's 60 incontinent women are 30 stress and 30 urge. The tag is held on structural grounds: literature-detail is a ScanResult, and a class-2 finding is Corroborator territory.
• Who depended on it: A82 Incorrect pelvic floor contraction…: Clinical [C13], and P2 the discriminating trigger's mechanism:. This is the claim that the problem the reference channel solves exists in women. A80 Pelvic floor surface EMG recorded with a… (claim 150unconfirmed). A79 A prescribing urogynecologist or pelvic… (claim 174unconfirmed). A83 No published study reports the proportion… (claim 176unconfirmed).
• What it costs: one citation, not a premise. The 25% figure cannot be quoted as a rate among incontinent women.
• Outcome: absorb
• Because: this is IR-41 claims 142 and 144's shape. The content was re-read cold and matches, and the tag is held by the instrument's class. Moving it waits on the corroborated tag proposal (Tier 3), not on any request.
  • What depended on it, and why that survives. P2 the discriminating trigger's "the problem exists in women" premise stands on PMID 16532264 alone: 30% of incontinent women against 17% of continent ones, on initial instruction. That is the branch A82 Incorrect pelvic floor contraction… needs.
  • The decisive version is commissioned. Whether the wrong manoeuvre *persists after the failed four-week trial this label requires* is A83 No published study reports the proportion… (claim 176unconfirmed), and IR-63 claims 176 commissions it through REQ-0170 Search the literature. 175's tag would not move that answer in either direction.
  • The rule for every artifact: PMID 1872333 is cited as women after brief verbal instruction, never as incontinent women. PMID 16532264's figure is cited as a mixed stress and urge group on initial instruction, never as a post-exercise or stress-only rate. A82 Incorrect pelvic floor contraction… closes as absorbed by IR-61 claim 175.

• The statement: two claims, one finding and one consequence.
  • claim 177unconfirmed (Clinical [C15]): abdominal EMG rises during a voluntary pelvic floor contraction in asymptomatic women (PMID 11135380). So a plain abdominal veto would withhold correct contractions too, and the false-withhold rate is a dose parameter to be measured in women.
  • claim 203unconfirmed (Technology [C19]): a maximal contraction in SUI women raises IAP by about 17 against 9 cmH2O and descends the bladder neck in 28% (PMID 30069729). Pelvic floor EMG and IAP rise in correct contractions as well as in Valsalva and bracing (PMID 19730763). So the rule must be relational, a ratio or onset order gated on a pelvic floor threshold.
• What the check found: each source was re-retrieved cold and matches. PMID 11135380: "In all subjects, the EMG activity of the abdominal muscles was increased above the baseline level during contractions of the PF muscles in at least one of the spinal positions." PMID 30069729: 17 vs 9 cmH2O in the incontinent group, and 28% bladder-neck descent. PMID 19730763: "PFM EMG and IAP increased during all tasks". In both claims the decision-rule conclusion is a design inference that no source states. Both tags are held on structural grounds.
• Who depended on it: A84 Abdominal muscle EMG rises above baseline… and A110. Behind them: Clinical [C15] and Technology [C19]; the detector design and its bench test; Study 1's false-withhold rate; REQ-0169 For P2's two-channel trigger(a), answered by Technology; A86 A two-channel trigger that withholds… (claim 179unconfirmed), the dose consequence for co-activators.
• What it costs: nothing a lookup can recover. Both rows say "both branches are survivable; the true one is the one that costs a calibration study". The record shows the costly branch, so a simple veto is off the table.
• Outcome: absorb
• Because: the physiology is read and matches, the instrument cannot say verified, and the dependants already run on the branch shown. Technology's detector in [C19] is relational and gated, not a veto. Clinical's Study 1 already measures the false-withhold rate. **What depended on it, and why that survives.**
  • The design inference is settled only by data from a detector run on women. That is REQ-0187 Search for any reported sensitivity (any published two-channel discrimination figure), REQ-0189 State the acceptance criterion the P2… (the pass line) and Study 1. Each of them is open, and none waits on these tags.
  • The dose consequence for co-activators is A86 A two-channel trigger that withholds…, commissioned at IR-63 claims 176.
  • The rule for every artifact: P2 the discriminating trigger's detector is described as relational and gated, never as an abdominal veto. Its false-withhold rate is an unmeasured parameter, never a design constant. A84 Abdominal muscle EMG rises above baseline… and A110 close as absorbed by IR-62 claims 177 and 203.

• The statement: four non-existence claims. Each says P2 the discriminating trigger rests on a quantity nobody has measured.
  • claim 176unconfirmed (Clinical [C14]): no study reports the wrong-manoeuvre proportion after a failed four-week exercise trial. So P2 the discriminating trigger's target stratum is unmeasured for this label.
  • claim 178unconfirmed (Clinical [C16]): no study or trial compares two-channel against pelvic-floor-only triggered stimulation on any continence outcome.
  • claim 179unconfirmed (Clinical [C17]): a two-channel trigger delivers fewer stimulated contractions to women who co-activate (PMID 16496395's pattern). No study reports whether they learn, within a twelve-week home course, a contraction the trigger accepts.
  • claim 204unconfirmed (Technology [C20]): no study reports the sensitivity or specificity of a two-channel pelvic-floor-plus-abdominal detector, or its tolerance to electrode displacement.
• What the check found: a claim that a study does not exist cannot be verified by a connector. Each sweep was run as leads only.
  • For 176, PubMed found two records, neither on point. Its second half (the cited figures are from initial or brief instruction) matches the abstracts.
  • For 178, PubMed "EMG-triggered AND (pelvic floor OR incontinence)" returned three records, none two-channel. ClinicalTrials.gov returned one, NCT06419517, not a two-channel comparison.
  • For 179, the cited pattern matches PMID 16496395 ("the abdominal and chest wall muscles were more active in the symptomatic group"). The prediction and the learning absence are not checkable.
  • For 204, the sweep found nothing on point (PMID 32119158 and 12853805 surfaced and are not).
• Who depended on it:
  • A83 No published study reports the proportion… (176): P2 the discriminating trigger's target stratum; Commercial [C6], whose share-only revenue needs that stratum's size; Study 1 Part A; A3 Medicare NCD 230.8 covers pelvic floor… (claim 9verified).
  • A85 No published study or registered trial… (178): P2 the discriminating trigger's clinical case, which now stacks two unmeasured margins on A28 Closed-loop EMG-gated stimulation produces… and A19 No randomised comparison of closed-loop…; Study 2; A79 A prescribing urogynecologist or pelvic…; A96 A labelling or promotional statement that….
  • A86 A two-channel trigger that withholds… (179): A24 Women who cannot actively contract the… (claim 96unconfirmed); A50 In a multicentre double-blind randomised… and A52 The only randomised trial of sEMG-triggered…, the Oxford 0–1 population of the one RCT; Technology's fallback states; A99 If a detached.
  • A111 (204): A34 No published study reports the sensitivity… (claim 105unconfirmed); the readiness of P2 the discriminating trigger's one new part; A98 No cleared home-use EMG-triggered… and A99 If a detached; pillar B1 The discriminating trigger works in the…'s sensing half.
• What it costs, separately:
  • 176: if a measured post-exercise proportion exists, it could show that supervised teaching corrects most wrong-manoeuvre women before they qualify. **That would leave P2 the discriminating trigger aimed at a population the coverage precondition has already treated.**
  • 178: if a comparison exists, a null result leaves the reference channel a mechanism with a measured absence of benefit. This is the one row whose counter-example would decide P2 the discriminating trigger most directly.
  • 179: if co-activators do not learn, P2 the discriminating trigger delivers less of the tested therapy to the RCT's own population, and a fallback makes it behave like K243079 for exactly those women.
  • 204: if a figure exists, it either clears P2 the discriminating trigger's bench gate in advance or ends its premise on a document.
• Outcome: commission
• Because: all four are gaps in the check. Each has a counter-example that would move P2 the discriminating trigger, and a proper search, not a term sweep, is what finds one.
  • 176 → REQ-0170 Search the literature, already open (Clinical): the post-programme wrong-manoeuvre proportion.
  • 204 → REQ-0187 Search for any reported sensitivity, already open (Technology): two-channel discrimination and displacement tolerance.
  • 178 and 179's absence half → REQ-0193 Search PubMed, raised today. Neither existing request asks for a continence-outcome comparison or a learning rate. REQ-0187 Search for any reported sensitivity asks for detector metrics, and REQ-0051 Hand-search the included-study tables of… asks whether a trial's arm was EMG-triggered, not two-channel.
  • 179's prediction half has no cheaper test than Study 1 Part B, Clinical's withhold fraction in women, which is funded and unsponsored. That is said here so the commission is not read as covering it. If every search returns "none found", all four stay unconfirmed, because absence cannot be verified. The next review absorbs them as measured gaps P2 the discriminating trigger is built across. It does not re-commission.

• The statement:
  • claim 168unconfirmed: a prescriber who has decided on a home stimulator chooses an EMG-triggered one over a stimulation-only one.
  • claim 174unconfirmed: that prescriber chooses a self-placed abdominal-reference discriminating trigger over a cleared single-channel ETS device such as K243079, despite the extra electrode. This is P2 the discriminating trigger's demand premise.
  • claim 170unconfirmed: a DME supplier dispensing under E0740 adds this device to rental inventory beside or instead of K243079.
• What the check found: stakeholder behaviour with "no connector, no primary research attached". The Verifier called 168 "permanently unconfirmed absent a human-attached preference study". That is a statement about what was attached, not about what prescribers or suppliers do.
• Who depended on it:
  • A76 A prescribing urogynecologist or pelvic… (168): pillar B1 The discriminating trigger works in the…'s "a clinician would switch for"; A28 Closed-loop EMG-gated stimulation produces… and A19 No randomised comparison of closed-loop…, whose margins it turns into a purchase; B4 Prescribers will send a woman home with a…'s calls.
  • A79 A prescribing urogynecologist or pelvic… (174): P2 the discriminating trigger's demand premise; B1 The discriminating trigger works in the…, one comparator further along.
  • A77 A DME supplier that dispenses home pelvic… (170): pillar B9 Payment reaches this device as billed; A17 A prescription home pelvic floor electrical… (claim 90unconfirmed); claim 82verified (the supplier bills); REQ-0155 Is a commercially insured patient's home…; B2 What the manufacturer collects per…'s numerator.
• What it costs:
  • 168: the closed loop can work and still not be bought.
  • 174: P2 the discriminating trigger has a better trigger nobody prescribes. A76 A prescribing urogynecologist or pelvic… stays a precondition but is no longer enough.
  • 170: no E0740 payment reaches this device, whatever the policies say. What survives is a manufacturer acting as its own DMEPOS supplier, which is a different channel and cost base.
• Outcome: commission
• Because: these are the clearest case of an unconfirmed that is a gap in the check. The check needs a human-attached artifact, and none has ever been commissioned. B4 Prescribers will send a woman home with a…'s "twenty structured calls" has been the assessment's settling test since 2026-09-14 without a request behind it.
  • REQ-0194 Run the twenty structured prescriber calls… (front-end → commercial) runs those calls and asks 168 and 174 in them, with the comparators named. It asks what evidence would move a prescriber to the second device.
  • REQ-0195 Call DME suppliers that dispense home… (front-end → commercial) calls at least eight of the 28 E0740 suppliers about stocking, the in-rental consumable and the skin electrode. The artifact either returns is a count of stated choices, with no extrapolation to a share. Whether the Verifier can lift a tag on it is the Verifier's call. A76 A prescribing urogynecologist or pelvic…, A77 A DME supplier that dispenses home pelvic… and A79 A prescribing urogynecologist or pelvic… stay live.