idea-006 · narrative
A synthesis across this candidate's claims, dossier and decks — every statement here traces to something that already exists elsewhere in the knowledge base. A narrative never invents a fact, upgrades a tag, or recommends anything.
How to read this. This is a synthesis, not a new source of truth: every claim it cites carries the tag the candidate file gives it right now, copied not paraphrased, and every line quoted from a deck is quoted verbatim. See docs/narrative-spec.md.
This is a proposed second entrant into a device category that did not exist before 2023: a home-use headphone-plus-transcutaneous-electrode device pairing acoustic and electrical stimulation to treat chronic subjective tinnitus, entering via 510(k) (premarket notification) against the device whose De Novo (De Novo classification request) grant created the category it would join. The regulatory container is real and checked against primary source — the classification exists, is Class II (FDA device class II), and matches the candidate's own wording claim 1verified claim 2verified — but almost everything the candidate's own strategy depends on sits outside that container: whether a follower can actually use the De Novo grantee as a predicate claim 3unconfirmed, whether the mechanism transfers off the predicate's tongue-stimulation site to the trigeminal/cervical skin site this candidate proposes claim 14unconfirmed, and whether anyone would prescribe a cheaper second device into a VA channel the incumbent appears to already occupy claim 41unconfirmed. Since the prior narrative, a business case appended nine further assumptions to the file — all unverified, none yet checked — covering the VA population overlap, an independent VA listing, the scope of the capital estimate, unit economics, and three alternative routes to market, discussed below. In the three and a half years the classification has existed, no follower 510(k) has cleared under it claim 4unconfirmed, and FTO (freedom to operate) — the rubric's one automatic-kill factor — has never been asked claim 28unconfirmed.
The regulatory premise is a bet on someone else's pathway. Neuromod Devices' Lenire device received a De Novo grant on 2023-03-06 claim 1verified, and that grant created product code QVN, 21 CFR (Title 21 of the Code of Federal Regulations) 874.3410, "combined acoustic and electrical external stimulation device for the relief of tinnitus" claim 2verified — a classification that, on primary source, did not exist before that date. The candidate proposes to enter that classification via 510(k) against Lenire as predicate, but the mechanism by which a De Novo grantee's device becomes available as a predicate for a follower is unconfirmed on this file's own reading of 21 CFR part 860, which confirms the classification order but not the downstream predicate consequence claim 3unconfirmed. The empirical record over the same period is silence: no 510(k) other than DEN210033 (an FDA De Novo decision number) itself has cleared under QVN in either a three-year or five-year openFDA window claim 4unconfirmed. The adjacent, older masker regulation (874.3400) is verified to exist as a fallback, but it trades the therapeutic-relief claim for a masking claim and does not support this candidate's stated indication claim 5verified claim 6verified.
The candidate's own design rationale names its central technical risk plainly: the genuine product change is substituting a transcutaneous trigeminal/cervical electrode for the predicate's intraoral tongue electrode, and whether bimodal-plasticity treatment transfers to that new site is an open scientific question, not an engineering detail claim 14unconfirmed. The class-level evidence behind the mechanism is substantial but sits entirely on the predicate's own configuration: a 326-adult randomized double-blinded pivotal trial reported a Cohen's d of -0.87 to -0.92 on the Tinnitus Handicap Inventory at 12 weeks, P<.001 claim 31unconfirmed; a follow-on settings trial reported 83.8% compliance with 70.3% of participants indicating benefit claim 32unconfirmed; and a 220-patient US real-world retrospective chart review reported a 91.5% responder rate claim 33unconfirmed. None of it is evidence about a non-tongue electrode. The nearest thing to that evidence is a completed, University of Michigan randomized crossover trial of non-intraoral auditory-somatosensory bimodal stimulation, 100 participants, with results already posted — located but not yet read claim 29unconfirmed. Against the standard of care this class is meant to displace, the guideline-recommended alternatives underperform on their own measures: a Cochrane review found CBT's effect on the THI fell below the review's own 7-point minimal clinically important difference claim 35unconfirmed, a Cochrane review of sound therapy found no evidence of superiority over waiting-list control or placebo claim 36unconfirmed, a large randomized trial of tinnitus retraining therapy found no meaningful separation from standard of care at 18 months claim 34unconfirmed, and the governing 2014 guideline contains no recommendation addressing bimodal neuromodulation at all claim 39unconfirmed.
Why the VA channel specifically: tinnitus is claimed to be the most prevalent service-connected disability among US veterans, but that claim has never been checked — no connector in this repo reads VA benefits data, and it is recorded as unchecked rather than weakly supported claim 8unconfirmed. What has moved since the candidate's first build is that the channel this idea was aimed at now looks occupied rather than open: per company press materials, Neuromod USA was awarded a VA Federal Supply Schedule 65 II A contract effective 2024-06-17, extending eligibility to the VA, DoD, Bureau of Prisons, Indian Health Services and Public Health Service claim 40unconfirmed — sourced to a press release, not a government record, against a schedule this file describes only in general terms claim 30unconfirmed. The domain dossier frames the same channel history from the supply side: it locates a University of Minnesota trial protocol (NCT05518682 (a ClinicalTrials.gov registration number)) that explicitly tests self-adjustment via virtual visits — the exact question this candidate's remote-fitting design rests on — but the protocol has posted no results, so it is a registered intention, not a finding, on that question [dossier: chronic-subjective-tinnitus-neuromodulation]. Whether a prescriber orders a second, cheaper device once the first is already on contract with a published pivotal trial behind it is the one question the file classifies as permanently unresolvable by any connector here — a stakeholder-future claim adoption refuses by design claim 41unconfirmed.
Reimbursement and market size are largely unbuilt rather than unfavorable. No CPT (Current Procedural Terminology) or HCPCS (Healthcare Common Procedure Coding System) code specific to a tinnitus device has been identified, and the claim is stated as an absence found by search rather than a confirmed absence in the code set claim 7unconfirmed. General population prevalence is now sourced — 9.6% 12-month tinnitus prevalence among US adults, of whom 7.2% rated it a big or very big problem claim 38unconfirmed — but price claim 11unconfirmed and addressable share claim 12unconfirmed remain Generator assumptions, and the severity-qualified population this device's intended_use: actually turns on has never been sized claim 10unconfirmed. The dossier is explicit that this is a gap in what exists to find, not a number anyone has looked up and failed to source: no severity-banded prevalence figure matching the indicated population was located anywhere [dossier: chronic-subjective-tinnitus-neuromodulation].
Most recently, on 2026-09-03, a business case was appended directly to the file's ## Claims as nine further assumptions, each recorded with the source line "business-case assumption (2026-09-03) — never checked; recorded so the Verifier can see it" and tagged unverified rather than unconfirmed — distinct from the earlier claims, and none of the three decks below (all built 2026-08-30) reference any of them. One narrows what claim 8unconfirmed's disability-compensation count would have to mean for it to bound the VA-channel population, rather than being two largely uncorrelated counts claim 42unverified. Two scope the capital estimate: that a follower device could obtain its own VA Federal Supply Schedule listing independent of whether it then gets prescribed claim 43unverified, and that the Generator's original low-single- digit-millions/18-30-month estimate covers regulatory and clinical activity only, excluding FSS contracting, private-audiology channel build, and the remote-fitting telemetry platform claim 44unverified. Two are unit-economics assumptions, that the assumed device price yields positive gross margin after cost of goods, platform amortization and clinical support claim 45unverified, and that a VA contract price would sit meaningfully below the self-pay range claim 46unverified. One assumes the remote-fitting and telemetry platform is buildable within the same capital and calendar envelope as the regulatory pathway rather than adding to it claim 47unverified. The remaining three name alternative routes to market not previously in the file: a sound-only fallback under the already-verified 874.3400 masker regulation reaching market at materially lower capital and shorter calendar than the QVN 510(k) route claim 48unverified; a partner already in the QVN classification paying a royalty or transfer price for a validated non-intraoral electrode-plus-telemetry module rather than building it in-house claim 49unverified; and bench validation plus regulatory scoping to license that module into an already-cleared QVN device being achievable at materially smaller capital and calendar than a full clearance-plus-channel route claim 50unverified. None of these nine resolves anything the candidate's own Scores table, Gate Check, or Wrap section already named as open — they are new assumptions layered on the same unresolved regulatory and mechanism questions, not answers to them.
intended_use: claim 2verified
adoption connector, is unverifiable by design absent primary research: dispensing
and self-pay willingness claim 17unconfirmed, daily-schedule adherence
claim 18unconfirmed, VA dispensing as routine rather than case-by-case
claim 19unconfirmed, acceptance of remote fitting claim 20unconfirmed, audiology as
the point of presentation claim 21unconfirmed, a follower winning VA FSS placement
claim 22unconfirmed, price and remote fitting as the only moat
claim 23unconfirmed, and private practices adding a second device line
claim 24unconfirmed
unverified and explicitly recorded
as never checked: that the veteran compensation count and the clinically-indicated population
overlap enough to bound the VA-channel population claim 42unverified; that a follower
device can obtain its own VA FSS listing within an absorbable cost and calendar, independent
of prescribing behaviour claim 43unverified; that the Generator's capital/calendar
estimate excludes go-to-market spend on FSS contracting, channel build, and the telemetry
platform claim 44unverified; that the assumed price point yields positive gross margin
claim 45unverified; that a VA contract price would sit meaningfully below the self-pay
range claim 46unverified; that the remote-fitting/telemetry platform is buildable within
the regulatory pathway's own capital and calendar envelope claim 47unverified; that a
sound-only fallback under 874.3400 reaches market at materially lower cost and shorter
calendar than the QVN route claim 48unverified; that a partner already in the QVN
classification would pay to license a validated non-intraoral module rather than build one
claim 49unverified; and that bench-validating and licensing that module is achievable at
materially smaller capital and calendar than a full clearance-plus-channel route
claim 50unverified
All three decks were built 2026-08-30 and, per python3 engine/scripts/check_decks.py --candidate idea-006, none is STALE: their source_updated stamps were restamped to match the candidate's current updated: 2026-09-03 because nothing substantive in their referenced claims had moved. Said plainly, though, because it matters for how to read them: none of the nine business-case claims added 2026-09-03 (claims 42-50) appears in any of the three decks — they were written before that addition and have not been rebuilt against it.
The desirability deck turns on the same channel-occupancy question the story above traces from the supply side: the channel this candidate was aimed at is reported open, and the incumbent is reported already standing in it.
"whether VA audiologists would prescribe a cheaper follower bimodal device instead of, or alongside, the incumbent already available through the VA Federal Supply Schedule" — the load-bearing condition of the desirability deck (built 2026-08-30).
The viability deck turns on the legal and empirical mechanics of the follower strategy itself — whether the pathway this candidate's whole business case rests on actually exists, independent of whether anyone would buy the result.
"whether a De Novo grantee's device is actually available as a predicate for a follower 510(k) into QVN" — the load-bearing condition of the viability deck (built 2026-08-30).
The feasibility deck turns on the mechanism-transfer question the candidate's own design rationale names as its central risk — the one condition that, if it fails, invalidates the cheap regulatory branch, the class evidence, and the price-only differentiation argument all at once.
"whether bimodal plasticity transfers from the predicate's tongue stimulation to a trigeminal or cervical cutaneous electrode" — the load-bearing condition of the feasibility deck (built 2026-08-30).
Where the three agree: each lens, from its own chair, arrives at a version of the same structural fact — this candidate's strategy depends on an incumbent's regulatory, clinical, and channel position being beatable by a follower, and none of the three conditions that would establish that has been checked. Where they diverge is what each treats as decisive if it resolves favorably: desirability treats prescriber behavior as the ceiling even if the legal and technical questions clear; viability and feasibility both treat their own condition as upstream of desirability's, since neither a legally unavailable predicate nor a mechanism that fails to transfer leaves anything for a prescriber to decide about. The business case's three alternative routes (a standalone VA listing, a sound-only fallback, a licensing deal) sit outside all three decks' load-bearing conditions as written — none of the decks has been asked whether those routes change what each lens treats as decisive.
If the follower route into QVN turns out to be legally available and the bimodal mechanism turns out to transfer to a trigeminal or cervical electrode, does a VA audiologist who can already prescribe the incumbent — on contract, with a published pivotal trial and a real-world series behind it — have any reason to order a cheaper second device instead, or alongside it? The business case's alternative routes (claims 43 and 48-50) do not answer this question; they are unchecked assumptions about ways to reach revenue that sidestep parts of it without yet establishing that any of them works.
_shared/plain-language.md