Evidence · idea-010

claim 1verifiedUS

21 CFR 862.1600 exists, is titled "Potassium test system", describes a device intended to measure potassium in serum, plasma and urine, and classifies it Class II

jurisdiction: US — confidence: verified — source: found via web search (eCFR result surfaced; section text not read in full by the Generator). Checkable in the exact form written: connectors regulation 862.1600 --expect "potassium". Note for the Verifier: the identification text as summarised says *serum, plasma and…

the full check

urine* — this candidate measures capillary whole blood, which may or may not sit inside that generic type and is precisely the kind of gap that decides whether a predicate argument holds. [Verifier 2026-08-25] connectors regulation 862.1600 --title 21 --expect "potassium" --jurisdiction US → verified (source_ref: eCFR versioner title-21, issue 2026-08-19, section=862.1600). Identification text read verbatim: “§ 862.1600 Potassium test system. (a) Identification. A potassium test system is a device intended to measure potassium in serum, plasma, and urine. Measurements obtained by this device are used to monitor electrolyte balance in the diagnosis and treatment of diseases conditions characterized by low or high blood potassium levels. (b) Classification. Class II.” Every element of the claim as written is confirmed, including the sample matrix. Sample-matrix finding, and it is a real one: the identification names serum, plasma and urine and does not name whole blood, capillary or fingerstick. This candidate measures capillary whole blood, which appears nowhere in the generic type's identification. That is a genuine gap in the predicate argument and not a nitpick — but it is not dispositive, and the Verifier found primary-source evidence cutting the other way: see the potassium-predicate claim below for K200865, a cleared potassium test system whose IFU covers heparinized whole blood. Whether capillary whole blood sits inside this generic type is an FDA substantial-equivalence judgment, not a database fact. [Audit 2026-08-25] Re-run cold: regulation 862.1600 --title 21 --expect "potassium" --jurisdiction US → verified, same source_ref (eCFR versioner title-21, issue 2026-08-19, section=862.1600), and the identification text returned today is character-for-character the text quoted above, including “serum, plasma, and urine”. The citation supports the claim. The sample-matrix finding is confirmed on primary text: 862.1600's identification names serum, plasma and urine and does not name whole blood, capillary or fingerstick.

claim 2verifiedUS

21 CFR 862.1225 exists and is the "Creatinine test system" regulation, Class II

jurisdiction: US — confidence: verified — source: the K070068 record summary via web search, which lists 862.1225 as the regulation number. Checkable: connectors regulation 862.1225 --expect "creatinine". [Verifier 2026-08-25] connectors regulation 862.1225 --title 21 --expect "creatinine" --jurisdiction US →…

the full check

verified (source_ref: eCFR versioner title-21, issue 2026-08-19, section=862.1225). Verbatim: “§ 862.1225 Creatinine test system. (a) Identification. A creatinine test system is a device intended to measure creatinine levels in plasma and urine. Creatinine measurements are used in the diagnosis and treatment of renal diseases, in monitoring renal dialysis, and as a calculation basis for measuring other urine analytes. (b) Classification. Class II.” The claim as written (section exists, is the Creatinine test system regulation, Class II) is fully confirmed. Sample-matrix finding — narrower than the Generator assumed and narrower than 862.1600: this identification names only plasma and urine. It does not name whole blood, and it does not even name serum. Yet K070068, cleared under this very regulation, has an IFU covering “capillary, venous, and arterial whole blood” and states the meter is “calibrated to provide plasma equivalent results to laboratory methods” — which is the documented mechanism by which a whole-blood device sits inside a plasma-worded generic type. This is directly load-bearing for the potassium channel's identical matrix gap. [Audit 2026-08-25] Re-run cold: verified, same source_ref, identification text identical to the quotation above including “plasma and urine”. Citation supports claim.

claim 3verifiedUS

FDA product code CEM is "Electrode, Ion-Specific, Potassium" and sits under 21 CFR 862.1600

jurisdiction: US — confidence: verified — source: a web-search summary only; the openFDA classification record was not read, and the Generator's first guess at this code (CGZ) was wrong and turned out to be chloride. Treat CEM as a lead, not a fact. Checkable: connectors classification --product-code CEM, with…

the full check

connectors classification --device-name "potassium" as the recovery query if it misses. [Verifier 2026-08-25] connectors classification --product-code CEM --jurisdiction US → verified (source_ref: openFDA device/classification product_code=CEM). Detail: “Device class 2: Electrode, Ion Specific, Potassium (regulation 862.1600, review panel CH).” Both halves of the claim confirmed: the code's name matches (modulo hyphenation — FDA's field reads “Ion Specific”, the claim wrote “Ion-Specific”) and it does sit under 21 CFR 862.1600 at Class II. The Generator's earlier wrong guess (CGZ/chloride) is superseded; CEM is now a fact, not a lead. No recovery query was needed. [Audit 2026-08-25] Re-run cold and reproduces byte-for-byte. One addition the first pass missed and that matters for any predicate sweep: CEM is not the only product code under 21 CFR 862.1600. connectors classification --product-code MZV --jurisdiction US → verified (source_ref: openFDA device/classification product_code=MZV): “Device class 2: Test, System, Potassium, Enzymatic Method (regulation 862.1600, review panel CH).” So the potassium generic type carries at least two codes — CEM (ion-selective electrode) and MZV (enzymatic) — and a predicate sweep run on CEM alone is incomplete. This candidate's ionophore-membrane ISE channel points at CEM; the cleared Piccolo potassium device (K200865) sits under MZV.

claim 4verifiedUS

FDA product code CGL is "Electrode, Ion Based, Enzymatic, Creatinine" and is the classification product code on K070068

jurisdiction: US — confidence: verified — source: web-search summary of the K070068 record. Checkable: connectors classification --product-code CGL. [Verifier 2026-08-25] connectors classification --product-code CGL --jurisdiction US → verified (source_ref: `openFDA device/classification…

the full check

product_code=CGL

). Detail: “Device class 2: **Electrode, Ion Based, Enzymatic, Creatinine** (regulation **862.1225**, review panel CH).” Code name and regulation both match the claim exactly. **[Audit 2026-08-25]** Re-run cold: **verified**, source_ref and detail reproduce exactly. The second half — that CGL is the classification product code **on K070068** — was checked separately against the openFDA 510(k) record's own fields:

product_code = 'CGL'`. Both halves verified.

claim 5verifiedUS

K070068 is a real FDA 510(k): "Nova StatSensor Creatinine Hospital Meter, Control Solutions, Linearity Solutions, and Test Strips" (Nova Biomedical), received 2007-01-08, decision 2007-05-07, decision "Substantially Equivalent", regulation 862.1225, product code CGL, clinical chemistry panel

jurisdiction: US — confidence: verified — source: an accessdata.fda.gov 510(k) database page surfaced by web search; the Generator did not open the record or the summary. Checkable twice over: connectors 510k K070068 and connectors clearance-summary K070068 --expect-indication "creatinine". **This is the one hard…

the full check

predicate anchor in the file and it covers only the creatinine channel.** [Verifier 2026-08-25] Checked twice over, as the claim invited. (1) connectors 510k "StatSensor Creatinine" --applicant "Nova Biomedical" --jurisdiction US → verified (source_ref: openFDA device/510k k_number=K070068). Raw record fields confirm every element of the claim: k_number=K070068, applicant='Nova Biomedical Corp.', date_received=2007-01-08, decision_date=2007-05-07, decision_code=SESE (Substantially Equivalent), product_code=CGL, advisory_committee_description='Clinical Chemistry'. (2) connectors clearance-summary K070068 --expect-indication "creatinine" --jurisdiction US → verified (source_ref: FDA 510(k) summary K070068, https://www.accessdata.fda.gov/cdrh_docs/pdf7/K070068.pdf). Indications for Use, verbatim and load-bearing: “The Nova StatSensor Creatinine Hospital Meter is intended for in vitro diagnostic use by health care professionals and for Point-Of-Care usage for the quantitative measurement of creatinine in capillary, venous, and arterial whole blood. Creatinine measurements are used in the diagnosis and treatment of renal diseases and in monitoring renal dialysis. Not for use in neonates. ... The Creatinine Meter is calibrated to provide plasma equivalent results to laboratory methods.” Three things this settles that a bare 510k check would have missed: (a) the predicate is cleared for capillary whole blood — the exact matrix this candidate needs — and for point-of-care use by health care professionals, matching this candidate's intended_use: setting and operator; (b) the plasma-equivalent calibration language is how a whole-blood device lives under a plasma-worded regulation; (c) the predicate's IFU is broader than this candidate's intended use, explicitly covering renal-dialysis monitoring, which this candidate excludes — a narrowing, which is the favourable direction for substantial equivalence. Scope limit, unchanged: this anchors the creatinine channel only. It says nothing about potassium. [Audit 2026-08-25] Both calls re-run cold and both reproduce exactly — 510k → verified, openFDA device/510k k_number=K070068, “NOVA STATSENSOR CREATININE HOSPITAL METER ... / Nova Biomedical Corp. decided 2007-05-07”; clearance-summary K070068 → verified, FDA 510(k) summary K070068, and the IFU returned today matches the text quoted above word for word, including “capillary, venous, and arterial whole blood” and “calibrated to provide plasma equivalent results to laboratory methods”. The citation supports the claim in full. One addition from the product-code enumeration run during this audit: clearances --product-code CGL --since 30y returns 7 of 7 clearances, among them K171059 “Nova StatSensor Creatinine Hospital Meter System” (Nova Biomedical, 2017-05-26) — a later clearance of the same platform than the K070068 the file anchors on, and probably the more current predicate to cite. Left as a pointer for the Generator; a Stage 0 scan carries no confidence tag.

claim 6verifiedUS

K091547 appears in the FDA 510(k) database alongside the StatSensor creatinine family

jurisdiction: US — confidence: verified — source: the K-number was returned by web search next to K070068; the Generator did not read this record and does not know what device it is. Recorded as a lead for connectors 510k K091547, explicitly not claimed as a predicate. [Verifier 2026-08-25] `connectors…

the full check

510k K091547 --jurisdiction US

 → **unconfirmed** (source_ref: none). Detail: “No 510(k) record found for device_name='K091547'”. That is a null query, not a finding:

connectors/fda.py::check_predicate_510k builds its query as device_name:"<arg>"

, so passing “K091547” searched openFDA for a device literally *named* “K091547”, which no device is. **[Audit 2026-08-25 — the first pass's conclusion was wrong and the claim is now RESOLVED.]** The first pass concluded that resolving this “needs either a connector change to support

k_number: lookups or a human opening accessdata directly.” **That is incorrect.** connectors clearance-summary takes a **K-number** as its positional argument and is exactly the right tool here. connectors clearance-summary K091547 --jurisdiction US → **verified** (source_ref: FDA 510(k) summary K091547

, https://www.accessdata.fda.gov/cdrh_docs/pdf9/K091547.pdf). K091547 is therefore a **real FDA 510(k) with an FDA-posted summary**, which is the whole of what this claim asserts. **What the device actually is** (Indications for Use, verbatim, OCR artifacts and all): “The **Nova Max Plus Blood Glucose and β-Ketone Monitoring System** Monitor is intended to be used for the quantitative measurement of **glucose** or β-hydroxybutyrate (**β-ketone**) in **fresh capillary whole blood**. It is intended for use by people with diabetes mellitus **in the home** and by health care professionals in clinical settings ... specifically indicated for the quantitative measurement of glucose in fresh capillary whole blood samples obtained from the **fingertip**, forearm and palm ... The Glucose Monitor is **calibrated to provide plasma equivalent glucose results to laboratory methods**.” **Three things this settles.** (a) The claim as written is verified — K091547 is in the FDA 510(k) database, and it is a **Nova Biomedical** strip-and-meter product, so “alongside the StatSensor family” is fair as a *family and format* statement. (b) It is **not** a potassium predicate and **not** a creatinine predicate — it is a glucose/β-ketone meter, so it changes nothing about the potassium channel. (c) It is nonetheless a **useful format analog and a second instance of the plasma-equivalent calibration device**: a single-use strip taking a fingertip capillary drop, multi-analyte on one platform, cleared for **home lay use** — which is the OTC route carried in

device_class:`. No web search was substituted at any point.

claim 7unconfirmedUS

No cleared strip-based, fingerstick-volume, single-analyte potassium meter was identified by the Generator, and none is being named. "No specific predicate identified for the potassium channel" is the honest state of this field

jurisdiction: US — confidence: unconfirmed — source: Generator's failed search. Checkable in the negative-lead sense: connectors 510k "potassium meter" and connectors classification --device-name "potassium"; an openFDA null is a null and would not establish that no such device exists. [Verifier 2026-08-25]…

the full check

unconfirmed — and this is the most consequential line in the file, so the sweep is reported in full. A bounded device-name sweep was run, deliberately without --applicant: connectors/fda.py returns "refuted" if applicant else "unconfirmed" on a name miss, so passing an applicant would manufacture a false refuted from a mere name-field miss. That is exactly the false refutation a prior pass in this repo recorded, and it was avoided here by construction. Name-matching misses (all unconfirmed, source_ref none — each means only that no openFDA record carries that string in its device_name field): 510k "potassium meter", "potassium strip", "potassium test strip", "handheld potassium", "fingerstick", "capillary potassium", "Piccolo Xpress". None of those nulls is evidence that no such device is cleared. connectors classification --device-name "potassium" returned verified but is a substring artifact that must not be recorded as evidence: it matched “Potassium Periodate” (product code KKR, Class I, regulation 864.1850, hematology panel) — a staining reagent, an entirely different device. It says nothing whatever about a potassium meter. What the sweep did find, and it partially cuts against the claim: cleared potassium-measuring devices unquestionably exist — most importantly K200865, “Piccolo Potassium Test System” (Abaxis, Inc., decided 2021-10-20), whose IFU covers heparinized whole blood at a point-of-care location (full IFU quoted below in this same line). Also surfaced: K791175 (Nova sodium/potassium analyzer, 1979), K833159, K872190, K853405, K113726 (epoc). A Stage 0 pointer carrying no confidence tag: clearances --product-code CEM --since 10y returned 5 clearances in ten years — K253491 (Roche cobas pro ISE), K241037 (Radiometer ABL90 FLEX PLUS), K220396 (Medica EasyStat 300), K230890 (Randox ISE Electrodes), K200544 (Diamond Diagnostics SmartLyte Plus). Net finding as of the first pass, stated precisely: the claim as narrowly written — no cleared strip-based, fingerstick-volume potassium meter was identified — was not contradicted by that sweep, though at that point it rested on device-name nulls (see the audit below, which replaced them with a real enumeration); every cleared potassium device located is a benchtop analyzer, rotor analyzer, cartridge or ISE module, none strip-and-fingerstick. But the broader framing “no predicate for the potassium channel” is now materially weaker than the file states: K200865 is a cleared device under the potassium test system generic type covering whole blood at point of care. Whether it is an adequate predicate is an FDA substantial-equivalence judgment, not a database fact, and is left for a human. connectors denovo "potassium test system" --jurisdiction US → unconfirmed: “No local cache at data/fda_denovo.csv” — nothing was searched, so the 510(k)-vs-De Novo fork is untouched and no De Novo absence may be inferred.

claim 8unconfirmedUS

FDA operates a Dual 510(k) and CLIA Waiver by Application pathway, with a guidance document of that name

jurisdiction: US — confidence: unconfirmed — source: an fda.gov guidance download URL surfaced by web search; the guidance was not read. Structurally unconfirmable here — no connector reads FDA guidance documents — but it is load-bearing for the schedule claim, so it is written rather than omitted. **[Verifier…

the full check

2026-08-25]** unconfirmed — structurally, exactly as the claim predicted. No connector in this repo reads FDA guidance documents; connectors regulation serves eCFR regulation text only, and a claim about a pathway FDA operates is a question about FDA policy and interpretation, not about regulation text. For context only, and explicitly not cited as settling this claim: connectors regulation 493.15 --title 42 --jurisdiction US → verified (source_ref: eCFR versioner title-42, issue 2026-08-13, section=493.15), which sets out CLIA's statutory waiver criteria — test systems that “(1) Are cleared by FDA for home use; (2) Employ methodologies that are so simple and accurate as to render the likelihood of erroneous results negligible; or (3) Pose no reasonable risk of harm to the patient if the test is performed incorrectly.” That is the CLIA-side criterion a waiver application must meet; it does not confirm that FDA operates a combined “Dual 510(k) and CLIA Waiver by Application” submission pathway or that a guidance of that name exists. Settling this requires a human to attach the named FDA guidance PDF as a manual citation. Load-bearing for the schedule claim under Factors 1 and 7 and still unverified. [Audit 2026-08-25] The 42 CFR 493.15 lookup this line cites as context did not reproduce on re-run today: regulation 493.15 --title 42 --jurisdiction US returned unconfirmed three times — twice “eCFR request failed: ... Read timed out (read timeout=30)”, once “eCFR returned HTTP 503”. That is an upstream availability failure, not a refutation; title-21 sections resolved normally in the same session, so this is a transient eCFR problem and 493.15 should be re-run before anyone relies on it. No tag changes either way, because 493.15 was correctly recorded here as context and was never cited as settling this claim.

claim 9unconfirmedUS

The Abbott i-STAT CHEM8+ cartridge measures potassium among other analytes, is 510(k)-cleared, and is CLIA moderate complexity, not waived

jurisdiction: US — confidence: unconfirmed — source: a distributor product page via web search. The moderate-complexity designation is a CLIA/CMS fact with no connector; only the clearance half is checkable, and the Generator did not identify its K-number. [Verifier 2026-08-25] **unconfirmed — and the check…

the full check

produced a genuine surprise that argues against the claim as written.** The Generator did not have a K-number; the Verifier located one. connectors 510k "CHEM8" --jurisdiction US → verified (source_ref: openFDA device/510k k_number=K191360): “i-STAT CHEM8+ cartridge with the i-STAT 1 System / Abbott Point of Care, Inc. decided 2020-02-14” — so a cleared CHEM8+ record exists. But connectors clearance-summary K191360 --expect-indication "potassium" --jurisdiction US → unconfirmed (phrase miss) (source_ref, omitted by the first pass and supplied here: FDA 510(k) summary K191360, https://www.accessdata.fda.gov/cdrh_docs/pdf19/K191360.pdf), and the actual IFU is about a different analyte entirely. Verbatim: “The i-STAT CHEM8+ cartridge with the i-STAT 1 System is intended for use in the in vitro quantification of ionized calcium in arterial or venous whole blood in point of care or clinical laboratory settings. Ionized calcium measurements are used in the diagnosis and treatment of parathyroid disease, a variety of bone diseases, chronic renal disease and tetany.” The word potassium does not appear. Correctly recorded as unconfirmed, not refuted: K191360 is evidently an analyte-specific clearance within the CHEM8+ family, and the panel's potassium channel is very likely cleared under a different K-number that this sweep did not identify. So the claim's potassium half is not established at Stage 3. Its second half — CLIA moderate complexity, not waived — has no connector at all (CLIA complexity categorisation is a CMS/CDC database, not openFDA, and is not cached in data/), so it remains entirely unchecked. Also verified in passing as a related record: 510k "i-STAT" → K103195, “I-STAT 1 WIRELESS ANALYZER” (Abbott Point of Care, 2011-02-08).

claim 10unconfirmedUS

Premise-correcting claim, deliberately written against this candidate's interest: a CLIA-waived potassium test already exists — the Abaxis/Abbott Piccolo Xpress waived-test list appears to include 84132QW (potassium) and 82565QW (creatinine). The gap this candidate claims is therefore form factor and per-test cost, not the absence of a waived potassium test

jurisdiction: US — confidence: unconfirmed — source: a CMS waived-test transmittal PDF and a coding page, both surfaced by web search and neither read in full. If this claim is verified it does not kill the candidate, but it should cap Factor 8 and it refutes the more attractive story. [Verifier 2026-08-25]…

the full check

unconfirmed overall — but the clearance half is now VERIFIED, and it is the half that hurts this candidate. Split the claim in two. (a) A cleared potassium test system on the Piccolo platform exists — verified. connectors 510k "potassium test system" --jurisdiction US → verified (source_ref: openFDA device/510k k_number=K200865): “Piccolo Potassium Test System / Abaxis, Inc. decided 2021-10-20”. connectors clearance-summary K200865 --expect-indication "potassium" --jurisdiction US → verified (source_ref: FDA 510(k) summary K200865, https://www.accessdata.fda.gov/cdrh_docs/pdf20/K200865.pdf). IFU verbatim: “The Piccolo® Potassium Test System, used with the Piccolo® blood chemistry analyzer or the Piccolo Xpress® chemistry analyzer, is intended to be used for the in vitro quantitative determination of potassium, in heparinized whole blood, heparinized plasma, or serum in a clinical laboratory setting or point-of-care location.” (Note in passing that the IFU's own list of conditions potassium measurement is used to diagnose includes “in vitro hemolysis”.) (b) That it is CLIA-WAIVED — unconfirmed, and nothing here checked it. A 510(k) IFU saying “point-of-care location” is not a CLIA waiver; waived status lives on the CMS CLIA waived-test list, which has no connector and is not in data/. The specific assertion that 84132QW and 82565QW appear on that list is doubly unchecked: it rests on CPT codes (connectors cpt → unverifiable, AMA-licensed) and on a CMS transmittal nobody read. Consequence, and it is adverse to the candidate: the Generator wrote this claim against its own interest and the checkable half of it stands up — a cleared potassium test system covering whole blood at point-of-care demonstrably exists. The narrowed gap (single-use strip + one fingerstick drop vs. a rotor analyzer) is intact, since Piccolo is a benchtop rotor analyzer requiring heparinized sample, but the more attractive story is refuted on the clearance half. Per the Generator's own instruction this should cap Factor 8; the Verifier computes no scores.

claim 11unconfirmedUS

CPT 84132 exists and is descriptor-defined as "Potassium; serum, plasma or whole blood"

jurisdiction: US — confidence: unconfirmed — source: coding aggregators via web search. Structurally unconfirmable in this pipeline: CPT descriptors are AMA-licensed and connectors cpt returns unverifiable by design (see idea-008's Stage 3 result). Settling it requires a human to attach AMA CPT Professional or…

the full check

a licensed vendor feed. Written anyway because "whole blood" appearing in the descriptor is what makes an existing code fit this device without a new code. [Verifier 2026-08-25] connectors cpt 84132 → unverifiable (source_ref: none). Detail: “CPT codes are AMA-licensed and cannot be scraped or reproduced verbatim (phase2-spec.md Section 3). Rescope this claim to HCPCS, or attach a licensed AMA/vendor source manually.” (Note: cpt takes no --jurisdiction flag; the claim is US and no cross-jurisdiction risk arises.) connectors hcpcs 84132 --jurisdiction US → unconfirmed: “No local cache at data/hcpcs_level_ii.csv” — and 84132 is CPT/HCPCS Level I in any case, not Level II, so that route cannot resolve it either. Recorded as permanently unconfirmed absent a human-attached licensed source (AMA CPT Professional 2026, or a licensed vendor feed such as the AMA CPT API, Optum EncoderPro or Find-A-Code), entered as a manual citation. No web search was run, no aggregator is cited, and the Verifier did not reason its way to the descriptor's wording. Load-bearing consequence, stated plainly: the candidate's entire favourable reimbursement leg — that “whole blood” already appears in the 84132 descriptor, so an existing code fits this device and no new code is needed — runs exclusively through this licensed text and is therefore unverified at Stage 3 and unverifiable in this pipeline by design.

claim 12unconfirmedUS

CPT 82565 exists and is descriptor-defined as "Creatinine; blood"

jurisdiction: US — confidence: unconfirmed — source: coding aggregators via web search. Same AMA-licensing wall; expect unverifiable. [Verifier 2026-08-25] connectors cpt 82565 → unverifiable (source_ref: none), identical AMA-licence detail as for 84132. connectors hcpcs 82565 --jurisdiction US →…

the full check

unconfirmed (“No local cache at data/hcpcs_level_ii.csv”). Existence and descriptor wording of 82565 both require a human-attached licensed AMA source. Permanently unconfirmed; no aggregator consulted.

claim 13unconfirmedUS

A CLIA-waived test billed by a waived-certificate site must carry the QW modifier

jurisdiction: US — confidence: unconfirmed — source: CMS waived-test transmittal material via web search, not read in full. No connector serves CLIA certificate or modifier rules. [Verifier 2026-08-25] unconfirmed. No connector serves CLIA certificate rules or claim-modifier policy. connectors cpt is…

the full check

unverifiable (AMA-licensed) and the CMS waived-test transmittal that carries the QW modifier list is not present in data/ and has no connector reading it. For context only and not cited as settling this claim, regulation 493.15 --title 42 → verified and gives the statutory criteria for a certificate of waiver, but says nothing about billing modifiers. Needs the CMS CLIA waived-test transmittal plus a licensed CPT source, both human-attached. [Audit 2026-08-25] The 42 CFR 493.15 lookup this line cites as context did not reproduce on re-run today: regulation 493.15 --title 42 --jurisdiction US returned unconfirmed three times — twice “eCFR request failed: ... Read timed out (read timeout=30)”, once “eCFR returned HTTP 503”. That is an upstream availability failure, not a refutation; title-21 sections resolved normally in the same session, so this is a transient eCFR problem and 493.15 should be re-run before anyone relies on it. No tag changes either way, because 493.15 was correctly recorded here as context and was never cited as settling this claim.

claim 14unconfirmedUS

The load-bearing negative economic claim: these are laboratory codes paid from the Clinical Laboratory Fee Schedule, not the Physician Fee Schedule, and the allowable for a single electrolyte is low in absolute terms — so the strip's cost to the practice must sit far below the allowable, or the clinic loses money on every test it insources. The Generator asserts no dollar figure

jurisdiction: US — confidence: unconfirmed — source: Generator's reasoning; no fee schedule was read. Checkable in form: connectors payment 84132 and payment 82565 (expected unconfirmed in this deployment — no cached CMS rate file; note also that the connector's cache is a HCPCS rate file and may not carry CLFS…

the full check

at all). [Verifier 2026-08-25] connectors payment 84132 --jurisdiction US and payment 82565 --jurisdiction US → both unconfirmed, source_ref none. Detail (identical for each): “No local cache at data/hcpcs_payment_rates.csv — payment rate for <code> cannot be checked until it exists. See data/README.md for the exact CMS source and expected columns. Do not substitute a web search for this file.” data/ currently holds only README.md. This is a missing-file result: nothing was searched. It is emphatically not a finding that payment is low or absent — reading it that way would invert the meaning of the null. The Scorer's own caveat is independently correct and worth preserving: the connector's cache is a HCPCS/PFS rate file and these are Clinical Laboratory Fee Schedule codes, so even with the cache attached this connector may not carry CLFS at all. What would settle it is the CMS Clinical Laboratory Fee Schedule annual file (a public, licence-free download) for the national limitation amount on 84132 and 82565. Both sides of the load-bearing inequality remain unquantified: no allowable was verified, and strip COGS is a Tier 3 human number no connector returns. Note this claim points negative — leaving it unconfirmed does not favour the candidate.

claim 15unconfirmedUS

No Medicare NCD or LCD specific to point-of-care potassium testing was identified by the Generator

jurisdiction: US — confidence: unconfirmed — source: no coverage search performed. Checkable in form: connectors coverage 84132 (expected unconfirmed). Recorded as an unexamined gap, not as an absence of coverage restrictions. [Verifier 2026-08-25] connectors coverage 84132 --jurisdiction US and `coverage…

the full check

82565 --jurisdiction US

 → both **unconfirmed**, source_ref none. Detail: “No local cache at data/medicare_coverage_policies.csv — Medicare coverage policy for <code> cannot be checked until it exists.” **Nothing was searched**, so this remains exactly what the Generator called it: an **unexamined gap, not an absence of coverage restrictions.** No NCD or LCD was located either way, and no inference about coverage stance is available in either direction. Would be settled by attaching the CMS Medicare Coverage Database export to

data/, then reading any returned policy with WebFetch to establish stance — recalling that coverage` verifies the existence of a policy, not positive coverage.

claim 16unconfirmedUS

TAM component 1 — annual Medicare Part B volume of CPT 84132 and 82565 in the outpatient setting

jurisdiction: US — confidence: unconfirmed — source: Generator states no number. This is the checkable-in-principle component: connectors procedures 84132 and procedures 82565, expected unconfirmed here (no cached CMS utilisation file). Two caveats to carry when it resolves: it is a Medicare-FFS floor, not a…

the full check

market size; and it counts all potassium tests, the large majority of which are central-lab draws this device does not address. [Verifier 2026-08-25] connectors procedures 84132 --jurisdiction US and procedures 82565 --jurisdiction US → both unconfirmed, source_ref none. Detail: “No local cache at data/medicare_procedure_volumes.csv — procedure volume for <code> cannot be checked until it exists.” This is the one genuinely checkable-in-principle TAM component and it is blocked only by a missing human-supplied cache file, not by any licensing wall — attaching the CMS Medicare Part B National Summary Data File or the Physician & Other Practitioners PUF would resolve it. Carry both caveats when it does: (1) a verified volume is a Medicare fee-for-service floor on US volume and never a market size, since Medicare FFS excludes Medicare Advantage, commercial, Medicaid, VA and cash-pay; and (2) the Generator's own sharper caveat — 84132 counts all potassium tests, overwhelmingly central-lab draws this device does not address, so the unfiltered number measures the wrong thing and would need filtering to office place-of-service before it means anything here.

claim 17unconfirmedUS

TAM component 2 — realistic net price per test strip to the clinic, bounded above by the CLFS allowable per the economic claim above and below by strip COGS for a three-channel electrochemical sensor

jurisdiction: US — confidence: unconfirmed — source: Generator's assumption; no price is asserted, because the allowable it derives from is unverified. connectors market is unverifiable by design. [Verifier 2026-08-25] connectors market → unverifiable (source_ref: none). Detail: “Market-size claim ...…

the full check

has no free, authoritative, machine-queryable source. Stays unconfirmed permanently unless a human attaches a licensed report and edits the claim by hand.” Permanently unconfirmed. No aggregator, SEO page or vendor site was consulted or accepted. Compounding this, the claim's upper bound derives from the CLFS allowable, which is itself unverified (see the economic claim above), and its lower bound is strip COGS, a Tier 3 human number — so this is an assumption resting on an unverified assumption at both ends. The Generator correctly asserts no price.

claim 18unconfirmedUS

TAM component 3 — addressable share assumption: the obtainable segment is outpatient heart-failure and nephrology clinics that today send potassium to an outside lab and wait a day for the result, assumed by the Generator at 10-20% of such clinics within five years

jurisdiction: US — confidence: unconfirmed — source: Generator's assumption, not market data. Embeds both a switching assumption and an unmeasured denominator (the count of such clinics), neither connector-checkable. [Verifier 2026-08-25] connectors market → unverifiable (same detail) for the share half,…

the full check

and connectors adoption → unverifiable for the switching half embedded inside it. Permanently unconfirmed. The uncounted denominator (how many US outpatient HF/cardiology and nephrology clinics send potassium to an outside lab) has no connector; it would need Medicare provider-enrolment data (PECOS / Provider Enrollment PUF) by specialty, or a licensed POC-diagnostics report, attached by a human. The 10-20%-in-five-years figure is the Generator's assumption and no Verifier judgement was substituted for the missing research.

claim 19unconfirmedUS

Denominator context, explicitly aggregator-grade — US adult heart failure prevalence is commonly cited near 6.7 million and CKD near 35 million from national surveillance summaries

jurisdiction: US — confidence: unconfirmed — source: secondary sources via web search; no primary surveillance data read. Not a TAM input; must not be multiplied by anything. Only a fraction of either population is in active RAASi titration. [Verifier 2026-08-25] unconfirmed. No connector serves national…

the full check

disease-prevalence surveillance data; connectors market → unverifiable. The 6.7M heart-failure and 35M CKD figures remain aggregator-grade secondary numbers with no primary surveillance source read. The file's own instruction is endorsed and repeated by the Verifier: this is not a TAM input and must not be multiplied by anything. Only a fraction of either population is in active RAASi titration, and that fraction is unquantified.

claim 20unconfirmedUS

Clinical premise — fear of hyperkalemia causes clinicians to underdose, discontinue and fail to reinstate RAAS inhibitors and MRAs in heart failure and CKD; one secondary source reports RAASi and MRA cessation rates of roughly 34.7% and 52.8% at one year following a hyperkalemia diagnosis

jurisdiction: US — confidence: unconfirmed — source: review articles and a PMC-hosted paper surfaced by web search; no primary paper was read and the specific percentages were taken from a search summary, not from a source document. Not verifiable by any Stage 3 connector: a Stage 0 literature scan can surface…

the full check

the papers but returns no confidence tag and cannot acquire one. [Verifier 2026-08-25] unconfirmed — no Stage 3 connector answers a clinical-epidemiology claim. The specific figures (RAASi cessation ~34.7% and MRA ~52.8% at one year following a hyperkalemia diagnosis) were taken by the Generator from a search summary, not a source document, and the Verifier did not substitute a web search to shore them up. A Stage 0 literature scan can surface candidate papers but returns a ScanResult that carries no confidence tag and cannot acquire one — pointers, never evidence. Settling this needs a human to read the primary papers and attach them as manual citations with the figures checked against the source tables.

claim 21unconfirmedUS

Clinical premise — guidance for patients at high hyperkalemia risk involves potassium measurement at baseline, ~2 weeks, ~1 month and periodically thereafter, i.e. the monitoring cadence that makes an in-visit test valuable

jurisdiction: US — confidence: unconfirmed — source: a guideline-summary article via web search, not the guideline itself. [Verifier 2026-08-25] unconfirmed. Clinical-guideline content has no connector: connectors regulation serves CFR regulation text only, and a monitoring cadence recommended by a…

the full check

professional-society guideline is neither regulation nor an FDA-database fact. The source remains a guideline-summary article, with the guideline itself unread. This claim underpins the “in-visit test is valuable” value proposition and is unverified at Stage 3; settling it requires a human to attach the guideline.

claim 22unconfirmedUS

The mechanism claim that decides this candidate: capillary fingerstick sampling causes red-cell lysis, hemolysis releases intracellular potassium, and the resulting falsely elevated potassium is the reason a strip-based fingerstick potassium meter does not already exist. The proposed mitigation is an on-strip hemoglobin channel that suppresses rather than corrects the result

jurisdiction: US — confidence: unconfirmed — source: Generator's reasoning from general clinical-chemistry knowledge; no source read. No connector answers this; it is settled only by bench data — paired capillary-versus-venous method comparison across operators, with deliberate lysis challenge — which is Stage 7 work.…

the full check

[Verifier 2026-08-25] unconfirmed — and the Verifier states plainly that NO CONNECTOR IN THIS REPO CAN REACH THIS CLAIM. It is a bench-science question about sample physics — whether capillary fingerstick sampling lyses red cells sufficiently to release intracellular potassium and falsely elevate the result, and whether an on-strip hemoglobin channel can reliably detect and suppress those samples. It is settled only by bench data: a paired capillary-versus-venous method comparison across multiple ordinary operators with a deliberate lysis challenge. No FDA, CMS, eCFR or patent lookup speaks to it, and none was pretended. The Verifier formed no opinion of its own on the physics and did not reason from general clinical-chemistry knowledge to a verdict — doing so would manufacture exactly the fluent, plausible, fabricated finding this gate exists to stop. A separate Stage 0 literature scan is being run on this question in parallel; whatever it returns is a ScanResult carrying no confidence tag, and it must not be converted into evidence here or used to flip this tag. One genuinely relevant primary-source observation surfaced incidentally and is offered as context, not as verification: the cleared K200865 Piccolo Potassium IFU lists “in vitro hemolysis” among the conditions potassium measurement is used to diagnose, and requires heparinized whole blood rather than a native fingerstick drop. This remains the factor that decides the candidate and it is unresolved.

claim 23unconfirmedUS

Failure-mode consequence of that mitigation, stated plainly: a strip that suppresses hemolysed samples produces no result rather than a wrong one, so the product's real performance metric is the invalid-result rate in ordinary clinic hands, not just accuracy on clean samples. A device that voids one draw in four is not clinically usable however accurate it is

jurisdiction: US — confidence: unconfirmed — source: Generator's reasoning. [Verifier 2026-08-25] unconfirmed — no connector reaches it. This is a derived consequence of the mechanism claim above and inherits its status entirely. The real performance metric it identifies — the **invalid/suppressed-result rate…

the full check

in ordinary clinic hands, not accuracy on clean samples — is measurable only by the same bench and multi-operator study, and no database anywhere reports it for a device that does not yet exist. The Verifier notes the reasoning is internally sound but records **no view on the likely magnitude of the invalid rate, which would be an invented number.

claim 24unconfirmedUS

Moat lever — the defensible assets are the hemolysis-detection-and-suppression logic, the potassium-selective membrane chemistry at whole-blood ionic strength, and the strip manufacturing yield curve; the meter itself is a commodity

jurisdiction: US — confidence: unconfirmed — source: Generator's judgement. Not connector-checkable as written. Stronger than idea-009's moat, weaker than a patent estate anyone has actually searched. [Verifier 2026-08-25] unconfirmed — not connector-checkable as written, and independently blocked. The three…

the full check

asserted defensible assets (hemolysis detect-and-suppress logic, potassium-selective membrane chemistry at whole-blood ionic strength, strip manufacturing yield curve) would have to be established as filed or filable IP through connectors patent, which is unavailable in this deployment: a control call returned “PATENTSVIEW_API_KEY not set — request one at https://patentsview.org/apis/keyrequest before running Stage 3 IP checks.” The trade-secret leg (yield curve) is undocumentable by any lookup in principle. Non-verifying Stage 0 context carrying no confidence tag and not evidence for this claim: clearances --product-code CEM --since 10y returned 5 clearances, all benchtop/rotor/ISE-module analyzers from established diagnostics firms (Roche, Radiometer, Medica, Randox, Diamond Diagnostics), and clearances --product-code CGL --since 10y returned 2, both from incumbents (Abbott Point of Care i-STAT Crea cartridge K261039; Nova Biomedical StatSensor Creatinine Hospital Meter System K171059). A scan result cannot upgrade this tag.

claim 25unconfirmedUS

FTO — no patent search was performed and no patent number is named. The likely estates sit with Nova Biomedical, Abbott Point of Care, Roche, Siemens Healthineers and i-SENS, plus the ion-selective-electrode and glucose-strip art generally

jurisdiction: US — confidence: unconfirmed — source: firm names from web search only. Recorded as an unexamined field, not a clear one. This deployment has no PATENTSVIEW_API_KEY, so even a named number returns unconfirmed; that would still mean "no search performed," never "no blocking art." **[Verifier…

the full check

2026-08-25]** unconfirmed — UNEXAMINED FIELD, and this must not be read as a clear one. Two independent blockers, exactly as the claim anticipated: (1) the claim names no patent number, so there is nothing for connectors patent to look up; (2) the connector is unavailable regardless — a control call (connectors patent US7276147) returned unconfirmed, source_ref none: “PATENTSVIEW_API_KEY not set — request one at https://patentsview.org/apis/keyrequest before running Stage 3 IP checks.” Zero patents were searched by the Verifier. No patent search engine, Google Patents query or generic web search was substituted to route around the missing key. The correct reading of this line remains “no search performed”, never “no blocking art found”. The concern is aggravated rather than relieved by what the clearance sweeps did show: the named likely estates (Nova Biomedical, Abbott Point of Care, Roche, Siemens Healthineers, i-SENS) are the same firms appearing on current CEM and CGL clearances, i.e. actively practising in exactly this art. Even a verified result from connectors patent would be a database fact about filing and grant dates, not an FTO conclusion — that is a Tier 3 legal opinion at Stage 7.

claim 26unconfirmedUS

Capital intensity and time-to-revenue — a dual 510(k) plus CLIA waiver with method-comparison studies at multiple untrained-operator sites, on top of electrochemical strip development and manufacturing scale-up, is materially slower and more capital-hungry than anything else currently in this portfolio. The Generator asserts no dollar or month figure, since both would be invented

jurisdiction: US — confidence: unconfirmed — source: Generator's estimate. Recorded as this candidate's honest principal weakness. [Verifier 2026-08-25] unconfirmed. Cost and calendar estimates are not database facts and no connector returns submission cost, review time or capex; any figure an agent supplied…

the full check

here would be invented, and the Generator correctly asserts none. The load-bearing sub-assumption — the 510(k)-versus-De Novo fork — could not be resolved: denovo "potassium test system" → unconfirmed (“No local cache at data/fda_denovo.csv ... openFDA publishes no De Novo endpoint at all”), meaning nothing was searched and no De Novo absence may be inferred. What the Verifier can report is that the fork is less unfavourable than the file assumes: cleared potassium test systems exist under 862.1600/CEM, including K200865 covering whole blood at point-of-care, so a 510(k) route for the potassium channel is not obviously foreclosed — though adequacy of any predicate is an FDA substantial-equivalence judgment, not a lookup. The CLIA-waiver leg's evidence burden remains unscoped: regulation 493.15 --title 42 → verified for the statutory waiver criteria, but the FDA Dual 510(k)/CLIA Waiver guidance that would size the study package is unreadable by any connector and needs human attachment. [Audit 2026-08-25] The 42 CFR 493.15 lookup this line cites as context did not reproduce on re-run today: regulation 493.15 --title 42 --jurisdiction US returned unconfirmed three times — twice “eCFR request failed: ... Read timed out (read timeout=30)”, once “eCFR returned HTTP 503”. That is an upstream availability failure, not a refutation; title-21 sections resolved normally in the same session, so this is a transient eCFR problem and 493.15 should be re-run before anyone relies on it. No tag changes either way, because 493.15 was correctly recorded here as context and was never cited as settling this claim.

claim 27unconfirmedUS

Competitive intensity — the named field is Nova Biomedical, Abbott Point of Care (i-STAT), Abbott/Abaxis (Piccolo), Siemens (epoc); the Generator did not run any clearance scan and does not know the size or trend of the field

jurisdiction: US — confidence: unconfirmed — source: firm names via web search. Checkable as a Stage 0 pointer only: clearances --product-code CEM and --product-code CGL return scan results that carry no confidence tag. [Verifier 2026-08-25] **unconfirmed — the named-firms judgement is not verified, though the…

the full check

field is now partly measured.** Stage 0 scans were run for pointers only; they return no confidence tag and cannot acquire one, so none of the following verifies this claim. clearances --product-code CEM --since 10y → 5 clearances 2016-2026: K253491 Roche cobas pro ISE indirect K/Na/Cl (2026-02-12), K241037 Radiometer ABL90 FLEX PLUS (2025-01-14), K220396 Medica EasyStat 300 (2023-09-26), K230890 Randox ISE Electrodes (2023-09-08), K200544 Diamond Diagnostics SmartLyte Plus (2020). clearances --product-code CGL --since 10y → 2: K261039 Abbott Point of Care i-STAT Crea cartridge, K171059 Nova Biomedical StatSensor Creatinine Hospital Meter System. Two observations the Generator did not have: (a) the CEM field over ten years is entirely benchtop, rotor, cartridge and ISE-module formats — not one strip-and-fingerstick device, which is consistent with (but does not prove) the Factor 6 reading that the absence reflects an unsolved physical problem rather than an unnoticed opportunity; (b) the firms named by the Generator via web search are corroborated as active in the field by primary clearance records, but Siemens/epoc appears in the sweep as Epocal, Inc. (K113726) rather than under the Siemens name. The switching leg — whether a clinic already owning a Piccolo or using an outside lab would buy a strip meter — is not connector-answerable: connectors adoption → unverifiable by design, and a clean clearances result was not allowed to stand in for it.

claim 28unconfirmedUS

Assumption, not a fact: clinicians would act on a point-of-care potassium result — actually changing a dose in the visit — rather than ordering a confirmatory central-lab draw and deferring the decision anyway, which would destroy the entire value proposition while leaving the device technically successful. There is no connector for clinician behaviour and no primary research behind this

jurisdiction: US — confidence: unconfirmed — source: Generator's assumption. [Verifier 2026-08-25] connectors adoption → unverifiable (source_ref: none). Detail: “...asserts what clinicians, payers, or patients would do. No free, authoritative, machine-queryable source answers that — it needs primary…

the full check

research (interviews, survey, published preference study)... Do NOT substitute a model's own estimate of stakeholder behaviour: a synthetic stakeholder opinion is a fabricated claim.” The Verifier produced NO opinion about whether clinicians would act on a point-of-care potassium result in-visit or order a confirmatory central-lab draw. Both suggested checkable proxies were attempted and both were blocked by missing cache files: coverage 84132 and procedures 84132 (see the Reimbursement and TAM claims above). Permanently unconfirmed absent commissioned primary research or a published preference study attached by a human. The Generator is right that this assumption failing would destroy the value proposition while leaving the device technically successful — which makes it the highest-value primary research in the file.

claim 29unconfirmedUS

Assumption, not a fact: outpatient heart-failure and nephrology clinics hold or would obtain a CLIA Certificate of Waiver and would absorb the QC, documentation and competency burden that comes with insourcing a test. The certificate rules themselves are CMS facts; the willingness is the assumption

jurisdiction: US — confidence: unconfirmed — source: Generator's assumption. [Verifier 2026-08-25] connectors adoption → unverifiable for the willingness half; no Verifier opinion was formed about what clinics would choose to do. The certificate-rules half is CMS/CLIA and has partial primary support…

the full check

recorded here as context rather than as verification of the willingness claim: regulation 493.15 --title 42 --jurisdiction US → verified (source_ref: eCFR versioner title-42, issue 2026-08-13, section=493.15), which sets out the certificate-of-waiver criteria and the waived-test list mechanism. The QC, documentation and competency burden a waived site actually carries lives in other CLIA subparts not checked here, and whether clinics would absorb it is pure stakeholder behaviour with no connector. Permanently unconfirmed absent a practice survey. [Audit 2026-08-25] The 42 CFR 493.15 lookup this line cites as context did not reproduce on re-run today: regulation 493.15 --title 42 --jurisdiction US returned unconfirmed three times — twice “eCFR request failed: ... Read timed out (read timeout=30)”, once “eCFR returned HTTP 503”. That is an upstream availability failure, not a refutation; title-21 sections resolved normally in the same session, so this is a transient eCFR problem and 493.15 should be re-run before anyone relies on it. No tag changes either way, because 493.15 was correctly recorded here as context and was never cited as settling this claim.

claim 30unconfirmedUS

Assumption, not a fact: patients would accept an additional fingerstick at each titration visit in exchange for avoiding a separate lab trip

jurisdiction: US — confidence: unconfirmed — source: Generator's assumption. [Verifier 2026-08-25] connectors adoption → unverifiable (source_ref: none), same detail. This is a patient-preference claim and needs a published preference study or commissioned patient research; **the Verifier formed and recorded…

the full check

no view on what patients would accept.** Permanently unconfirmed. Worth noting the claim is more benign than it looks — patients in RAASi titration are already having venous potassium drawn on the same cadence, so the comparator is a lab trip rather than no test — but that observation is the Verifier's framing of the question, not evidence, and does not move the tag.

claim 31unverifiedUS

Prescribers who receive a suppressed (hemolysis-flagged) potassium result during an outpatient titration visit will repeat the fingerstick or proceed on clinical grounds, rather than abandon point-of-care testing and revert to the outside laboratory

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (desirability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 32unverifiedUS

Medical assistants and nurses in ordinary outpatient heart-failure and nephrology clinics can obtain an adequate capillary drop without finger-milking from patients with peripheral edema, cool extremities, or frequently-lanced fingers

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (desirability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 33unverifiedUS

In-visit fingerstick potassium and creatinine replaces the outpatient titration blood draw rather than adding to it, i.e. the remaining analytes a RAASi titration visit typically needs do not still require a separate venous specimen

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (desirability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 34unverifiedUS

US outpatient heart-failure and nephrology prescribers require a published capillary-versus-venous potassium comparison in an outpatient non-dialysis population before they will change titration practice on a point-of-care result

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (desirability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 35unverifiedUS

A count of US outpatient cardiology/heart-failure and nephrology clinic sites that currently send potassium testing to an outside laboratory is obtainable from a public or licensed source

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (viability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 36unverifiedUS

The payment a clinic receives per test exceeds the fully-loaded cost of performing it in office (strip, meter amortisation, quality-control material, and operator time) at the test volume a single outpatient heart-failure or nephrology clinic generates

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (viability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 37unverifiedUS

A clinic that insources a potassium and creatinine test it previously sent out captures the payment for that test itself, rather than the payment remaining with the reference laboratory

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (viability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 38unverifiedUS

Potassium and creatinine reported from a single strip and a single fingerstick drop are separately payable as two laboratory tests rather than bundled into one payment

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (viability deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 39unverifiedUS

The maximum allowable potassium bias and imprecision for a RAASi dose-titration decision is defined by an FDA-recognised or published analytical performance criterion that this device could be designed and validated against

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (feasibility deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 40unverifiedUS

An on-strip hemoglobin channel can detect free hemoglobin from a single fingerstick drop at the concentration at which potassium results are materially biased, and suppress the result before it is displayed

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (feasibility deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 41unverifiedUS

A potassium channel calibrated to report plasma-equivalent results from capillary whole blood holds that calibration across the hematocrit range found in heart-failure and CKD patients

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (feasibility deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 42unverifiedUS

The three-channel strip retains its calibration and its hemolysis-detection threshold across a commercially viable shelf life under ordinary outpatient-clinic storage conditions

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (feasibility deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 43unverifiedUS

FDA will require the on-strip hemolysis detection-and-suppression function to be evaluated within the device's own marketing application rather than accepting it as an unreviewed internal safeguard

jurisdiction: US — confidence: unverified — source: deck-surfaced assumption (feasibility deck, 2026-08-25) — never checked; recorded so the Verifier can see it

claim 44unverifiedUS

The outpatient clinic itself, not a distributor, GPO, or health-system central lab, is the direct purchaser of the meter and strips