idea-017 · human · US
For prescription use in the home by adults aged 18 years and older with chronic; respiratory failure consequent to chronic obstructive pulmonary disease, to deliver; noninvasive bi-level positive airway pressure ventilation with a back-up respiratory rate; through a nasal or facial mask and to record overnight transcutaneous carbon dioxide tension; for review by the treating practitioner, who sets and approves every ventilation setting; not; for life support, not for use in acute respiratory failure requiring hospital care, and not; for automatic adjustment of ventilation settings without practitioner approval.
Still being built — 6 of 8 wrap criteria met; nothing for you to decide yet. No department has been asked. 10 of 45 claims verified, 1 refuted. Nobody has yet written down what it rests on.
A prescription home bi-level positive airway pressure device with a back-up respiratory rate, capable of the high-intensity settings (high inspiratory pressure, fixed back-up rate above the patient's spontaneous rate) used to control nocturnal hypoventilation in hypercapnic COPD, paired with a reusable transcutaneous CO2 (tcCO2 (transcutaneous carbon dioxide)) sensor the patient applies at the ear or chest for overnight recording at home. The device stores the overnight tcCO2 trace alongside its own ventilation data, computes a proposed settings change against a practitioner-set CO2 target, and presents that proposal to the treating practitioner for approval; no setting changes without practitioner approval. The clinical loop it is copying is the one used in the HOT-HMV trial, where home NIV (non-invasive ventilation) was titrated to overnight transcutaneous CO2 — but performed in the home instead of an attended laboratory.
In the mechanism, as the candidate file states it — the plain-language paragraph in ## Wrap has not been written yet. Domain: none linked
No business case yet, so nothing here says who writes the cheque. /business-case idea-017 builds one.
What this idea rests on that is not yet known, in the order this idea's own decision queue puts it: worst if false first, cheapest to settle first inside that. The first row is open; open any other for what would settle it.
This idea is unassessed, so nobody has yet named the few things it rests on. Until then, the gaps below are the rubric factors that are unscored or carry an open question. /assess idea-017 replaces this list with the real one.
Why it scored what it did: intended_use: is stated, so this is scorable. Scored on the launch route only (claims 17, 18): if the MNS / 21 CFR (Title 21 of the Code of Federal Regulations) 868.5895 pairing holds and the tcCO2 subsystem is Class II (FDA device class II) under special controls, a 510(k) (premarket notification) with marketed single-function…the rest of this reasoning
analogs is plausible. Held to 3, not higher, because claim 18verified was read off a search summary with no K-number, claim 17verified does not identify a CFR section or product code, and claim 19unconfirmed is explicit that no combined-function predicate was found — which is an absence-of-evidence result, not white space, so the De Novo (De Novo classification request) route in device_class: stays live.
What would settle it: openFDA classification and 510(k) records keyed on product code MNS, plus the product code behind the FDA (US Food and Drug Administration) PcCO2 special-controls guidance; cheap if openFDA answers, and the guidance itself may need a human to attach since fda.gov gates…the rest of this gap analysis
automated fetches.
Why it scored what it did: Claims 1-8 and 14. Strongest leg in the file if the claims hold: the premise is that CMS (Centers for Medicare & Medicaid Services) has already performed the site-of-care shift by issuing NCD 240.9 (a Medicare national coverage determination number) for exactly this population, with an aligned LCD (local coverage determination) (L33800 (a Medicare local coverage determination number)/A52517) and an existing…the rest of this reasoning
paying code (E0471 (a HCPCS code for durable medical equipment), capped rental) — that is an existing coverage instrument, not a path to one. Held to 4 because none of the three dollar amounts in claim 14unconfirmed was retrieved, no DMEPOS (durable medical equipment, prosthetics, orthotics and supplies) allowable exists in the file (claim 10unconfirmed), and claim 30unconfirmed — whether the coverage documents accept a home overnight tcCO2 recording in place of an attended study — is an unchecked assumption that would invert the economics. claim 5refuted's high-intensity envelope (IPAP > 20 cm H2O, back-up rate >= 14 bpm) may come from the proposed decision memo rather than the final NCD (national coverage determination).
What would settle it: Direct primary-source fetch of the NCD 240.9 text, decision memo CAG-00465N, transmittal R13808NCD, LCD L33800/A52517, and the DMEPOS fee schedule — no connector in this repo reads the CMS NCD or LCD databases, so claims 1-6 need a manual…the rest of this gap analysis
retrieval or a new connector.
Why it scored what it did: Claims 9-12. Structure is sound and deliberately written as retrievable components rather than guesses, but no figure of any kind is asserted: volume (claim 9verified) names a CMS public use file without a count, price (claim 10unconfirmed) names the fee…the rest of this reasoning
schedule without an allowable, addressable share (claim 11unconfirmed) is an explicit unsourced Generator estimate, and the 16M COPD population base (claim 12unconfirmed) is recollection. Scored 3 on what the claims would imply if all four held; a Medicare-only volume would in any case be a floor, not a market.
What would settle it: The CMS DME (durable medical equipment) by Referring Provider public use file for E0471 allowed services and the DMEPOS fee schedule allowable — both free downloads. Hypercapnic-CRF prevalence has no connector reach here and likely needs a human-attached primary…the rest of this gap analysis
table.
Why it scored what it did: Claims 25 and 26. No patent search was run and no patent number is asserted anywhere in this file — the field is unexamined, not clear. Scored low rather than blank because claim 25unconfirmed names two specific live risk surfaces…the rest of this reasoning
(volume-assured/target-ventilation control algorithms of the iVAPS/AVAPS kind, and tcCO2 sensor head construction, heating and calibration) that sit on two well-capitalised incumbents' active products. claim 26unconfirmed's expectation that the ~1990 bi-level-with-back-up-rate foundational patents have expired is inference from the modality's age only. This carries the rubric's only automatic kill.
What would settle it: A patent search on the two named surfaces against ResMed, Philips Respironics and SenTec assignee portfolios, via PatentsView or EPO (European Patent Office) OPS; low cost once an API key is confirmed available in this environment, which it has not been.
Why it scored what it did: Claims 19, 24, 25. The proposed moat is the combined-function integration, but there is no IP (intellectual property) behind it (claim 25unconfirmed) and the two incumbents named in claim 24unconfirmed already ship both halves separately — adding a tcCO2 channel to an existing…the rest of this reasoning
bi-level platform is a plausible fast-follow. claim 19unconfirmed's failure to find a combined device is an absence-of-evidence result and is not credited here as open space.
What would settle it: Same patent search as FTO (freedom to operate), plus a product-code-keyed clearance scan to establish whether any incumbent has already cleared a combined channel; cheap, and it is the same work item as the two rows above.
Why it scored what it did: Claims 15 and 16. HOT-HMV (Murphy et al., JAMA 2017) is a randomised trial reporting extended admission-free survival, and claim 16unconfirmed states the titration target was overnight transcutaneous CO2 — which is the single load-bearing mechanism…the rest of this reasoning
claim, since the whole device is that loop moved into the home. Scored 4 rather than 5 because claim 16unconfirmed was read from a search summary and not the paper's methods, the PMID (PubMed identifier) is itself unverified, the trial was UK-conducted, and claim 29unconfirmed (adherence to nightly IPAP > 20 cm H2O and correct at-home sensor placement outside trial supervision) is an unsourced assumption.
What would settle it: Open the JAMA paper's methods section via the PubMed record for PMID 28528348 to confirm tcCO2-guided titration; free and fast.
Why it scored what it did: Claims 7, 8, 17, 18, 28. If the 510(k) route and existing E0471 capped-rental payment hold, first revenue does not wait on a new code or a new coverage decision, which is unusually fast. Offset by a two-subsystem hardware build (ventilator…the rest of this reasoning
plus reusable sensor with a consumable), and by claim 28unconfirmed's unresearched assumption that DME suppliers will stock and service the extra sensor inside the existing rental allowable.
What would settle it: Supplier-economics primary work (DME channel margin against the E0471 allowable) — a licensed report or human-run supplier interviews; the most expensive item in this column.
Why it scored what it did: Claims 19, 22, 23, 24. Scored as a crowded space: claim 24unconfirmed names two entrenched incumbents with marketed volume-assured bi-level devices plus SenTec in tcCO2. claim 19unconfirmed is explicitly an absence-of-evidence keyword-search result and is…the rest of this reasoning
not scored as white space. The only opening is claim 23unconfirmed's 2024 Philips consent decree restricting US sales, which the Generator did not check for whether it has since been lifted — a temporary dislocation is not a structural gap.
What would settle it: Check the current status of the Philips consent decree (court docket or Philips investor filings) and pull openFDA recall records for Philips Respironics; both cheap, and the decree status is the one that changes the read.
Why it scored what it did: Left unscored deliberately. No claim in this file speaks to the portfolio this candidate would sit in. The nearest claims, 20 and 21, establish the opposite of a fit argument: the Generator's own design position is that RAPID gives…the rest of this reasoning
this file no coverage-timing tailwind and none should be credited. Guessing a number here would not trace to a claim.
What would settle it: A portfolio-level statement from the requesting department, or a Stage 5/round-table position — not a primary-source lookup, and not Verifier work.
Seven chairs, each answering from its own field. The verdict, and the first condition it set.
Not convened. No department has stated a position on this idea yet — a normal state, not a defect. /round-table idea-017 asks all seven.
Only verified counts toward a score. A low count is usually a statement about the tooling — an absent data file, an unset key — rather than about the idea.
1 claim(s) the Verifier found to be false — a refuted claim is a finding, and the loop owes each one an answer.
<the single piece of evidence that would move this candidate most, what it would cost, and who would have to do it. Take it from the Scores table's "What would settle it" column — this is a selection from that gap analysis, not a new judgement. One sentence.>
Nothing is waiting on a decision. The next move above is work, not a choice.
Only verified counts toward a score. A low count is usually a statement about the tooling (an absent data file, an unset key) rather than about the idea.
No gate check recorded yet.
To finish it:
domains: — run /research**Cheapest next test:** is still the unfilled _template.md placeholder — name the test, or the criterion checks nothingNo human review is recorded. Shortlisting, legal and clinical sign-off and capital allocation are Tier 3: no agent in this repo makes or simulates one.
Runs only after a human shortlists the idea.
No loop record: nobody has written down what this idea rests on or convened a department on it. That is a normal state, not a defect. /front-end-loop idea-017 opens one.
Nobody has yet written down what this idea rests on and what it would cost to be wrong. Unassessed is a state, not a low score. /assess idea-017 adds it.
Click any step to see what it holds for this idea. donecurrentloopinghuman only
## Wrap section, as writtenThe packet a human reads. See docs/wrap-spec.md. This section says nothing about whether the idea is good — it says the file is finished enough to be read, and names what to do next. Filling it in advances no candidate and approves nothing.
The one-paragraph version. <not yet written — Stage 5/wrap>
What would have to be true. One line per lens, quoted verbatim from that deck's load-bearing condition. Do not summarise or re-rank them.
Cheapest next test: <the single piece of evidence that would move this candidate most, what it would cost, and who would have to do it. Take it from the Scores table's "What would settle it" column — this is a selection from that gap analysis, not a new judgement. One sentence.>
Where the evidence actually stands. State the coverage plainly, including what is blocked on tooling rather than on a finding — a missing CMS cache file and an unset EPO_OPS_KEY are not evidence that a code or a patent does not exist. Run python3 engine/scripts/wrap_status.py --candidate idea-017 for the numbers.
| Factor | Score | Why | What would settle it |
|---|---|---|---|
| Regulatory pathway | 3 | intended_use: is stated, so this is scorable. Scored on the launch route only (claims 17, 18): if the MNS / 21 CFR 868.5895 pairing holds and the tcCO2 subsystem is Class II under special controls, a 510(k) with marketed single-function…the rest of this reasoninganalogs is plausible. Held to 3, not higher, because claim 18verified was read off a search summary with no K-number, claim 17verified does not identify a CFR section or product code, and claim 19unconfirmed is explicit that no combined-function predicate was found — which is an absence-of-evidence result, not white space, so the De Novo route indevice_class: stays live. | openFDA classification and 510(k) records keyed on product code MNS, plus the product code behind the FDA PcCO2 special-controls guidance; cheap if openFDA answers, and the guidance itself may need a human to attach since fda.gov gates…the rest of this gap analysisautomated fetches. |
| Reimbursement pathway | 4 | Claims 1-8 and 14. Strongest leg in the file if the claims hold: the premise is that CMS has already performed the site-of-care shift by issuing NCD 240.9 for exactly this population, with an aligned LCD (L33800/A52517) and an existing…the rest of this reasoningpaying code (E0471, capped rental) — that is an existing coverage instrument, not a path to one. Held to 4 because none of the three dollar amounts in claim 14unconfirmed was retrieved, no DMEPOS allowable exists in the file (claim 10unconfirmed), and claim 30unconfirmed — whether the coverage documents accept a home overnight tcCO2 recording in place of an attended study — is an unchecked assumption that would invert the economics. claim 5refuted's high-intensity envelope (IPAP > 20 cm H2O, back-up rate >= 14 bpm) may come from the proposed decision memo rather than the final NCD. | Direct primary-source fetch of the NCD 240.9 text, decision memo CAG-00465N, transmittal R13808NCD, LCD L33800/A52517, and the DMEPOS fee schedule — no connector in this repo reads the CMS NCD or LCD databases, so claims 1-6 need a manual…the rest of this gap analysisretrieval or a new connector. |
| TAM | 3 | Claims 9-12. Structure is sound and deliberately written as retrievable components rather than guesses, but no figure of any kind is asserted: volume (claim 9verified) names a CMS public use file without a count, price (claim 10unconfirmed) names the fee…the rest of this reasoningschedule without an allowable, addressable share (claim 11unconfirmed) is an explicit unsourced Generator estimate, and the 16M COPD population base (claim 12unconfirmed) is recollection. Scored 3 on what the claims would imply if all four held; a Medicare-only volume would in any case be a floor, not a market. | The CMS DME by Referring Provider public use file for E0471 allowed services and the DMEPOS fee schedule allowable — both free downloads. Hypercapnic-CRF prevalence has no connector reach here and likely needs a human-attached primary…the rest of this gap analysistable. |
| FTO / IP | 2 | Claims 25 and 26. No patent search was run and no patent number is asserted anywhere in this file — the field is unexamined, not clear. Scored low rather than blank because claim 25unconfirmed names two specific live risk surfaces…the rest of this reasoning(volume-assured/target-ventilation control algorithms of the iVAPS/AVAPS kind, and tcCO2 sensor head construction, heating and calibration) that sit on two well-capitalised incumbents' active products. claim 26unconfirmed's expectation that the ~1990 bi-level-with-back-up-rate foundational patents have expired is inference from the modality's age only. This carries the rubric's only automatic kill. | A patent search on the two named surfaces against ResMed, Philips Respironics and SenTec assignee portfolios, via PatentsView or EPO OPS; low cost once an API key is confirmed available in this environment, which it has not been. |
| Moat / defensibility | 2 | Claims 19, 24, 25. The proposed moat is the combined-function integration, but there is no IP behind it (claim 25unconfirmed) and the two incumbents named in claim 24unconfirmed already ship both halves separately — adding a tcCO2 channel to an existing…the rest of this reasoningbi-level platform is a plausible fast-follow. claim 19unconfirmed's failure to find a combined device is an absence-of-evidence result and is not credited here as open space. | Same patent search as FTO, plus a product-code-keyed clearance scan to establish whether any incumbent has already cleared a combined channel; cheap, and it is the same work item as the two rows above. |
| Mechanism & clinical risk | 4 | Claims 15 and 16. HOT-HMV (Murphy et al., JAMA 2017) is a randomised trial reporting extended admission-free survival, and claim 16unconfirmed states the titration target was overnight transcutaneous CO2 — which is the single load-bearing mechanism…the rest of this reasoningclaim, since the whole device is that loop moved into the home. Scored 4 rather than 5 because claim 16unconfirmed was read from a search summary and not the paper's methods, the PMID is itself unverified, the trial was UK-conducted, and claim 29unconfirmed (adherence to nightly IPAP > 20 cm H2O and correct at-home sensor placement outside trial supervision) is an unsourced assumption. | Open the JAMA paper's methods section via the PubMed record for PMID 28528348 to confirm tcCO2-guided titration; free and fast. |
| Capital intensity & time-to-revenue | 3 | Claims 7, 8, 17, 18, 28. If the 510(k) route and existing E0471 capped-rental payment hold, first revenue does not wait on a new code or a new coverage decision, which is unusually fast. Offset by a two-subsystem hardware build (ventilator…the rest of this reasoningplus reusable sensor with a consumable), and by claim 28unconfirmed's unresearched assumption that DME suppliers will stock and service the extra sensor inside the existing rental allowable. | Supplier-economics primary work (DME channel margin against the E0471 allowable) — a licensed report or human-run supplier interviews; the most expensive item in this column. |
| Competitive intensity | 2 | Claims 19, 22, 23, 24. Scored as a crowded space: claim 24unconfirmed names two entrenched incumbents with marketed volume-assured bi-level devices plus SenTec in tcCO2. claim 19unconfirmed is explicitly an absence-of-evidence keyword-search result and is…the rest of this reasoningnot scored as white space. The only opening is claim 23unconfirmed's 2024 Philips consent decree restricting US sales, which the Generator did not check for whether it has since been lifted — a temporary dislocation is not a structural gap. | Check the current status of the Philips consent decree (court docket or Philips investor filings) and pull openFDA recall records for Philips Respironics; both cheap, and the decree status is the one that changes the read. |
| Strategic fit & portfolio balance | — | Left unscored deliberately. No claim in this file speaks to the portfolio this candidate would sit in. The nearest claims, 20 and 21, establish the opposite of a fit argument: the Generator's own design position is that RAPID gives…the rest of this reasoningthis file no coverage-timing tailwind and none should be credited. Guessing a number here would not trace to a claim. | A portfolio-level statement from the requesting department, or a Stage 5/round-table position — not a primary-source lookup, and not Verifier work. |
To finish this packet:
domains: — run /research**Cheapest next test:** is still the unfilled _template.md placeholder — name the test, or the criterion checks nothingSource file: knowledge-base/candidates/idea-017.md
Nothing on this page is a recommendation. Shortlist review, legal and clinical sign-off and capital allocation are human-only decisions, and no agent in this repo may make or simulate one. This is a compilation of what the candidate file says.
_shared/plain-language.md