idea-001 · desirability deck
Every slide states a condition that would have to be true, then reports where it stands using the candidate file's own claim and its own confidence tag. A deck never upgrades a tag, invents a number, or recommends anything, and there is no ask slide.
The chair: a woman with stress urinary incontinence who has already tried the exercises, and the pelvic floor PT or urogynecologist deciding what to do next with her.
How to read this: every slide is a condition, not a conclusion. A bracketed claim reference points into knowledge-base/candidates/idea-001.md — the number is the claim's position in its ## Claims list, the tag is copied from it and never adjusted here. A [no claim] marker means nothing in the file speaks to the condition at all. See docs/deck-spec.md.
What changed since the 2026-08-24 build: the research lane added a body of named clinical literature (claims 39-45), and the CMS caches were populated on 2026-08-30, which resolved the reimbursement questions this deck had been carrying as "never searched". Two of those resolutions cut against the candidate. Rebuilt from the file, not patched.
Would have to be true: There would have to be a large population of women who have done a documented four-week pelvic muscle exercise trial, failed it, and currently fall out of care — because the intended use admits nobody else.
Where it stands: The denominator is now sourced rather than estimated, though still untagged as verified: NHANES 2015-March 2020 (n=6,446) puts symptomatic urinary incontinence at 22.3% of US women aged 20+, with stress-predominant incontinence 16.2% of incontinence cases [claim 44: unconfirmed] — a named PMID rather than the Generator's original round number [claim 20: unconfirmed]. The drop-out premise is now backed by a real cohort rather than by this deck's own assumption: in a 5-year follow-up of women who completed a guided PFMT programme, 56.8% (75 of 132 responders) had discontinued training entirely [claim 40: unconfirmed], and long-term PFMT adherence across 19 studies of 1,141 women ranges 10-70% [claim 39: unconfirmed]. The original assumption stands beside them [claim 26: unconfirmed]. The four-week prerequisite has lost the justification it was designed around. It was written into the intended use to mirror the coverage criterion the file attributed to Medicare NCD 230.8 — and that claim is now [claim 9: refuted]: no NCD or LCD in the cached coverage file names E0740. Read the claim's own caveat before treating that as settled; the Verifier found a 230.8,NCD,Non-Implantable Pelvic Floor Electrical Stimulator row present in the same cache with a blank codes column, one of 356 of 646 rows with that field empty.
What would settle it: For the funnel, a pelvic floor PT clinic's own drop-off data after a failed exercise trial — the cohort that matters is "failed and disengaged", which no published series measures directly. For the four-week gate, the actual NCD 230.8 policy text, which a targeted fetch could not retrieve (CMS returned HTTP 403) and which therefore needs a human.
If it's false: The addressable population is not the SUI population — it is the much smaller "already in care, already failed, still engaged" slice, and the TAM claims [claim 19: unconfirmed] are measuring the wrong thing. And if the NCD gate turns out not to exist, the intended use is carrying a documentation burden bought for nothing.
Would have to be true: She would have to insert an intravaginal probe at home, on a schedule, for the months a neuromuscular training effect takes — and keep doing it after the novelty ends.
Where it stands: Still no direct evidence, but the proxies are now named devices with named trials rather than nothing. Adherence to the prescribed 12-week course remains an assumption this deck put on record [claim 27: unconfirmed]; the closest checkable proxy is InControl Medical's own open-label pilot of the InTone intravaginal device, where 5 of 33 enrolled women (15.2%) withdrew before completing the 12-week course [claim 42: unconfirmed]. Acceptability against the external alternative is the sharper question [claim 28: unconfirmed], and here the proxy runs the other way: in the ELITONE transcutaneous pivotal trial, 17 of 57 (29.8%) treatment-arm and 12 of 30 (40.0%) sham-arm participants did not complete the intervention period [claim 41: unconfirmed] — the non-intravaginal comparator did not obviously retain women better. Neither figure is about this candidate's device. The alternative itself is cleared and on the market [claim 4: verified], so a woman offered an intravaginal device is being offered the more intrusive of two options for the same indication.
What would settle it: A preference study between external and intravaginal delivery in the target cohort — the one question the proxies cannot answer, since a completion rate inside a trial with study visits is not a home-use adherence rate.
If it's false: The EMG dataset that carries the moat [claim 22: unconfirmed] and the RTM billing that carries part of the revenue [claim 13: unconfirmed] both dry up, because both are generated by continued use rather than by purchase.
Would have to be true: A pelvic floor PT or urogynecologist would have to hand a patient a home device instead of keeping her in clinic — and document a failed four-week trial to do it.
Where it stands: Unmoved, and it is the one condition on this deck the last six days of work did not touch. The candidate names it itself, and names it as an assumption rather than a fact: the idea depends on clinicians being willing to prescribe a home device and bill remote monitoring alongside in-clinic sessions "rather than treating the device as revenue cannibalization" [claim 24: unconfirmed]. The documentation half is a second, separate ask [claim 29: unconfirmed], and the Verifier's disposition on it is explicit that no checkable proxy was identified at all. Whether the RTM codes that are supposed to make the trade worthwhile even apply here still cannot be checked in this repo: CPT is AMA-licensed and the connector cannot return verified or refuted under any circumstance [claim 13: unconfirmed], [claim 15: unconfirmed].
What would settle it: Interviews with pelvic floor PTs and urogynecologists on the in-clinic-versus-home economics as they experience them; a licensed AMA CPT source to establish whether the RTM offset is real.
If it's false: The prescriber is being asked to trade billable visits for documentation work, and the channel closes at the point of prescription — the single most likely quiet death for this candidate.
Would have to be true: The burden — an intravaginal device in the house, a four-week failed trial before she qualifies, an app reporting her adherence to her clinician, and whatever she pays — would have to stay under what the incontinence itself costs her.
Where it stands: This slide is the one the CMS caches actually answered, and the answer is mixed. The code is real and its descriptor matches the device as written, including "complete system" [claim 7: verified], and a published DMEPOS allowed amount now exists [claim 10: verified]. The candidate records that amount as $74.52 for 2026 and, in the claim's own words, "a dollar figure now exists, and it is a small one" — a national unadjusted rate that the claim explicitly declines to treat as covering this device's bill of materials, since an app-linked EMG-sensing intravaginal probe is a higher-BOM device than the bare stimulators E0740 was historically priced against. PDAC guidance reads E0740 as intended for stimulators with integrated monitoring capability [claim 45: unconfirmed], which is this device's shape rather than an obstacle to it. What sits against all of that is coverage: no NCD or LCD in the cached file names E0740 [claim 9: refuted], which leaves claims to case-by-case contractor adjudication — a real reimbursement risk signal, and, in the claim's own words, "NOT the same as a denial". The monitoring asymmetry is worth naming plainly, because it is a real cost to her and a benefit to someone else: the adherence and EMG stream that becomes the manufacturer's moat [claim 22: unconfirmed] is her data.
What would settle it: The NCD 230.8 text read directly — the one retrieval that failed (HTTP 403) and the one that decides whether a covered pathway exists. Then a patient-facing read of what a beneficiary actually pays for a DME item under this benefit, which the fee-schedule amount alone does not give.
If it's false: She declines, or starts and stops — and neither shows up in any factor the rubric scores.
Would have to be true: She would have to see something change — a biofeedback trace improving, fewer episodes — early enough to keep going, and her clinician would have to be able to point at evidence that closed-loop stimulation beats what she already failed at.
Where it stands: The clinical evidence arrived since the last build, and it is the worst news on this deck. A Cochrane systematic review of 56 randomised trials (n=3,781) found electrical stimulation with non-implanted devices probably more effective than no active treatment for subjective cure of stress urinary incontinence (RR 2.31, 95% CI 1.06-5.02), but low-quality evidence of no difference between electrical stimulation and pelvic floor muscle training (RR 0.85, 95% CI 0.70-1.03) [claim 43: unconfirmed]. Read against this candidate's intended use, that is the comparison that matters and the wrong side of it: the device is indicated precisely for women who have already failed pelvic muscle exercise training, and the pooled evidence does not show stimulation beating the therapy they failed. The candidate's own differentiating premise — that EMG-gated closed-loop stimulation outperforms open-loop stimulation or exercise alone by a margin a clinician would act on — remains this deck's own write-back and is unverifiable on the "would act on" half [claim 33: unconfirmed]. What is verified is regulatory, not clinical: devices in this class are cleared for this indication [claim 4: verified], which establishes permission to market and nothing about comparative benefit.
What would settle it: A head-to-head of closed-loop EMG-gated against open-loop stimulation in the post-PFMT-failure cohort specifically. Nothing in the Cochrane pool addresses the gating mechanism this candidate is built on, so the existing literature can neither support nor kill the differentiator — it only removes the fallback argument that stimulation as a class is enough.
If it's false: The product is a more expensive way to deliver a therapy the evidence does not distinguish from the exercises she already failed [claim 43: unconfirmed], available in a less intrusive external form [claim 4: verified], and the differentiation argument is entirely a data-and-billing argument with no clinical leg.
Would have to be true: Ten conversations at pelvic floor PT clinics would have to return the same story: a real drop-off cohort, an appetite to prescribe rather than retain, and patients who would accept an intravaginal home device.
Where it stands: No primary research of any kind sits behind this candidate. The connector that would be asked reports the whole class of question as out of scope, and did so again explicitly this pass on claims 26, 27, 28 and 29 — adoption "asserts what clinicians, payers, or patients would do... needs primary research... Do NOT substitute a model's own estimate of stakeholder behaviour", and produced no opinion [claim 24: unconfirmed]. The literature added since has narrowed what the clinic visits still have to answer — the drop-out cohort is now partly evidenced [claim 40: unconfirmed] — without touching the prescriber question at all.
What would settle it: The clinic visits. Slides 1, 2, 3 and 6 all resolve in the same room, which makes this the highest-yield unspent hour on the candidate.
If it's false: The idea is a coding strategy — E0740 plus RTM — wrapped around a device, and the user was inferred from the reimbursement structure rather than observed.
Every condition above with nothing verified behind it.
Three structural notes. First, the file's verified claims are still almost entirely about permission and payment plumbing — [claim 1: verified], [claim 4: verified], [claim 5: verified], [claim 7: verified], [claim 10: verified] — a code, a descriptor, a rate, a classification. Not one verified claim in this file is about anyone wanting the thing. Second, the strongest desirability signal in the file still arrived as a warning: a separate clearance was required to extend from stress to urge incontinence [claim 5: verified], the regulator confirming this product may only address the narrower of two overlapping user problems. Third, and new since the last build: the two TAM volume components were checked against real CMS utilisation and both came back [claim 16: refuted] and [claim 17: refuted] — 28,093 allowed services for CPT 90912 and 4,130 allowed units for E0740 against claimed thresholds of 200,000 and 10,000. Those are Medicare FFS floors, not market sizes, and the claims' own notes say so; but the direction of the correction is not in this candidate's favour.
If only one thing from this chair could be checked: whether pelvic floor PTs and urogynecologists will prescribe a home device instead of retaining the patient in clinic [claim 24: unconfirmed].
It survives as the load-bearing condition through a week in which almost everything else on this deck moved. The candidate itself flags it; the adoption connector refused it again this pass; and everything here runs through it — no prescription means no adherence question, no EMG dataset, no RTM billing, and no route to the patient at all, since the intended use is prescription-only by construction. It is also the condition the product's own revenue model works hardest against: the device competes with the visit it is supposed to be prescribed at.
Naming it is not a recommendation, a gate, or a kill.