idea-010 · 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: the user. Here that is three people at once: the heart-failure or nephrology prescriber who has to make a RAASi or MRA dose decision, the medical assistant or nurse who has to get a usable fingerstick drop, and the patient who gives it. The device's intended use is professional point-of-care use by trained clinic staff, in outpatient clinics only, not at home, and not in emergency, critical-care or dialysis settings.
How to read this: every slide is a condition, not a conclusion. A bracketed claim reference points into knowledge-base/candidates/idea-010.md. The number is the claim's position in its ## Claims list, and the tag is copied from that file as it stood on 2026-09-29 and never adjusted here. [no claim] means nothing in the file speaks to the condition. Claims 60–62 were appended by this deck's build. See docs/deck-spec.md.
What changed since the last build: on 2026-09-29 the Verifier moved claims 8, 14 and 16 to verified. Those are the Dual 510(k) (premarket notification)/CLIA-waiver pathway, CLFS (clinical laboratory fee schedule) payment and Medicare volume. None of the three says whether anyone wants the device. Every stakeholder claim this lens rests on is still unconfirmed or unverified, because the adoption connector returns unverifiable by design. The rubric scores no factor for user demand, so this is the expected shape for the lens. It is not a finding against the idea.
Would have to be true: At a RAASi or MRA titration visit in a US outpatient heart-failure or nephrology clinic, the prescriber would have to lack a current potassium result while the patient is still in the room, so the dose decision is deferred to a phone call or a later visit.
Where it stands: Nobody has checked the "deferred decision" moment. This build put it on record claim 60unverified. The file's assumed segment is clinics that send potassium to an outside lab and wait a day, and that is the Generator's assumption, not market data claim 18unconfirmed. The reason the moment matters is the clinical premise that fear of hyperkalemia leads clinicians to underdose, stop and fail to restart RAASi and MRAs. The cessation figures behind it were taken from a search summary, and no primary paper was read claim 20unconfirmed. The monitoring cadence that makes an in-visit result recur (baseline, about 2 weeks, about 1 month, then periodically) comes from a guideline summary, and the guideline itself was not read claim 21unconfirmed. As context only: Medicare fee-for-service billed 319,876 standalone potassium services in the office/non-facility setting claim 16verified. That count says nothing about when the result reaches the prescriber.
What would settle it: A human would need to attach and read two things: the primary papers behind the cessation figures (claim 20unconfirmed), and the HF/CKD guideline that sets the cadence (claim 21unconfirmed). Then someone needs to ask a sample of outpatient HF and nephrology prescribers how the last several titration decisions were made: in the visit, by phone the next day, or at the next visit. No connector reaches either; adoption is unverifiable.
If it's false: If prescribers already have a same-visit potassium result, whether from an on-site analyzer or a pre-visit draw, there is no moment for the device to own. Slides 2–6 become moot, the addressable segment in the viability deck (Slide 3) shrinks toward zero, and Factor 3 loses its share leg.
Would have to be true: An outpatient HF or nephrology prescriber would have to change a RAASi or MRA dose within the visit on the fingerstick potassium result. The alternative, which kills the value, is ordering a confirmatory venous draw and deferring the decision anyway.
Where it stands: This is the file's own named highest-value primary-research question. It is recorded as an assumption, and the Verifier formed no view on it claim 28unconfirmed. A second condition sits inside it: the in-visit test has to replace the titration blood draw rather than add to it. If the rest of a titration panel still needs a venous specimen, the patient gets both a draw and a fingerstick claim 33unconfirmed.
What would settle it: Structured interviews or a vignette study with US outpatient HF and nephrology prescribers. Each would be shown a POC potassium value in the titration band and asked what they would order next, with the confirmatory-draw rate as the primary measure. For the replacement half, a human would read which analytes the HF/CKD guidelines ask for at a titration visit. No connector reads clinical guidelines.
If it's false: The device would work and would not be used, as the file says. Titration would still wait for the lab, the clinical premise on Slide 1 goes unaddressed, and every revenue line in the viability deck loses its volume. This is the load-bearing condition for this lens (Slide 8).
Would have to be true: Four more parties would have to say yes besides the prescriber. (a) The clinic would have to hold or obtain a CLIA (Clinical Laboratory Improvement Amendments) Certificate of Waiver and absorb its QC (quality control), documentation and competency burden. (b) Where the clinic is hospital-owned, the health system's laboratory director or POC testing committee would have to approve a fingerstick potassium method. (c) The medical assistant or nurse would have to get an adequate drop. (d) Whoever buys the meter and strips would have to be the clinic, or someone the clinic can persuade.
Where it stands:
What would settle it: (a) and (d): a practice survey of outpatient HF and nephrology practice administrators. (b): interviews with POC coordinators or laboratory directors at a handful of health systems that own HF and nephrology clinics. (c): a sample-adequacy usability study in real clinics with real patients (see the feasibility deck, Slide 6). The split of site types in (b) is the business case's A2 falsifier: CMS (Centers for Medicare & Medicaid Services) Provider of Services and CLIA laboratory files cross-referenced to practice locations.
If it's false: Any single "no" in this chain blocks adoption on its own terms. A laboratory director who won't approve a capillary potassium method removes every hospital-owned site from reach, whatever the prescriber wants. The site-mix question is then also a reimbursement question (viability deck, Slide 1).
Would have to be true: The device would have to cost each user less than it saves them. The patient would have to accept an extra fingerstick at each titration visit in exchange for skipping a lab trip. When the strip suppresses a hemolysed result, clinic staff and the prescriber would have to repeat the stick or proceed on clinical grounds, not give up on point-of-care testing.
Where it stands: Patient acceptance of the extra fingerstick is an assumption, and the Verifier formed no view on it claim 30unconfirmed. The file points out that the comparator is a lab trip, not "no test", but it records this as framing, not evidence. What staff do after a suppressed result is an assumption with no connector claim 31unconfirmed. How often that happens is the invalid-result rate in ordinary clinic hands, and the file states plainly that a device voiding one draw in four is not usable however accurate it is claim 23unconfirmed. That rate is owned by the feasibility deck. This slide borrows it because the user pays for it in repeat sticks and lost minutes. No claim in the file speaks to what the patient pays out of pocket, if anything, for a POC test compared with a lab draw [no claim].
What would settle it: A published patient-preference study or a short commissioned patient survey among people in RAASi titration. Prescriber and staff interviews about what they would do after a flagged result. The invalid-result rate itself only comes from the multi-operator bench-and-clinic study in the feasibility deck (Slide 5). The out-of-pocket question needs a reimbursement analyst reading cost-sharing rules for CLFS tests in an office setting, and no connector reads them.
If it's false: If a suppressed result sends staff back to the outside lab, a high invalid rate turns into abandonment, not repeat sticks, and Slide 2's behaviour never forms. If patients refuse the extra stick, per-clinic volume in the viability deck falls claim 48unconfirmed.
Would have to be true: Before US outpatient HF and nephrology prescribers change titration practice on a fingerstick potassium, they would have to see some evidence they themselves accept. Two candidates: a published capillary-versus-venous potassium comparison in an outpatient non-dialysis population, and a guideline that accepts a point-of-care potassium result as a basis for a titration decision.
Where it stands: Whether prescribers require a published comparison is an assumption the Verifier could not reach claim 34unconfirmed. Whether that comparison exists and comes out favourably is a separate question, carried in the feasibility deck claim 67unverified. Whether current guidelines accept a POC potassium result for titration is unchecked. This build put it on record claim 62unverified. The only guideline content in the file is the monitoring cadence, taken from a summary article claim 21unconfirmed. As a format precedent only: FDA (US Food and Drug Administration) has cleared a creatinine strip-and-meter for point-of-care use by health-care professionals on capillary whole blood claim 5verified. That shows a regulator accepted the format, which is not the same as a prescriber trusting it for a potassium decision.
What would settle it: A human would read the current US HF and CKD (chronic kidney disease) society guidelines for their position on POC potassium, and a librarian-grade literature search would look for paired capillary-versus-venous potassium studies in outpatient adults. Both are documents a person can read and neither is reachable by a connector. A literature scan would return pointers only, never a tag. Prescriber interviews (Slide 2) should ask directly what evidence would change their practice.
If it's false: If prescribers require evidence that does not exist yet, desirability waits on a clinical study before first sale. That study moves the viability deck's calendar (Slide 4) and is not in any capital claim.
Would have to be true: In the first handful of real clinics, most in-visit fingerstick potassium results would have to end in a same-visit dose decision with no confirmatory venous draw, and the clinics would have to still be running the test months later. A high confirmatory-draw rate, or a lapse in use after the first few months, would falsify Slides 2–4.
Where it stands: The same-visit decision is claim 28unconfirmed's behaviour and has not been observed claim 28unconfirmed. There is one revealed-behaviour proxy that doesn't need this device to exist: whether practices that placed an incumbent POC chemistry analyzer capable of potassium were still running potassium on it twelve months later. The business case put this on record and nobody has checked it claim 59unverified.
What would settle it: The proxy is the cheaper of the two. It needs reagent reorder data for potassium-capable POC cartridges or rotors at cardiology and nephrology accounts, from a distributor or a willing incumbent, at months three and twelve after placement, with the stated reason for any lapse. That is the business case's A4 falsifier, owned by commercial or business development. The direct observation would need a prototype, so it sits behind the feasibility deck.
If it's false: If practices that already own a potassium-capable POC analyzer stopped using it for potassium, the barrier is not the form factor this candidate changes. The narrowed gap in the file, form factor and per-test cost claim 10unconfirmed, would not be the gap users care about.
Every condition above that has nothing verified behind it. That is all six. The verified claims this deck cites (5 and 16) are format and volume context, and neither speaks to a user's behaviour.
After write-back only one of the six has a bare [no claim] in it. That is not good news: every condition here rests on unconfirmed or unverified stakeholder behaviour, and no connector in this repo can move any of it.
If only one thing from this chair could be checked: would an outpatient HF or nephrology prescriber change a RAASi or MRA dose within the visit on a fingerstick potassium result, or order a confirmatory venous draw and defer the decision anyway claim 28unconfirmed?
This one rather than the others because every other condition on the deck is conditional on it. Patient acceptance, staff tolerance of suppressed results, laboratory-director approval and guideline endorsement only matter if the result, once produced, changes the decision it was built for. It is also the one condition where the device could meet every engineering spec and still fail completely. It can be asked before any strip exists.
Naming it is not a recommendation, a gate, or a kill. It is the answer to "check what first."
_shared/plain-language.md