idea-011 · 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.

Continuous wearable ECG hyperkalemia-risk screening — desirability

The chair: the user. Two of them, really: the CKD/heart-failure outpatient wearing the patch between clinic visits, and the prescribing clinician who has to act on what it says.

How to read this: every slide is a condition, not a conclusion. A bracketed claim reference points into knowledge-base/candidates/idea-011.md — the number is the claim's position in its ## Claims list, the tag is copied from it and never adjusted here. A [no claim] marker means nothing in the file speaks to the condition at all. This is the first deck built for this candidate — there is no prior build to compare against. See docs/deck-spec.md.


Slide 1 — The user, and the moment

Would have to be true: There would have to be a large population of adult outpatients with CKD or heart failure, undergoing RAASi initiation or up-titration, who are currently sent back for a repeat venous potassium draw between scheduled visits — a burden real enough that a between-visit screening step is worth wearing continuously.

Where it stands: The comorbidity population is plausible but every population figure in this file is unconfirmed: HF/CKD comorbidity reported at 20-67% with no resolvable source [claim 8: unconfirmed], a US GWTG-HF registry finding 64% of HF hospitalizations had discharge eGFR <60 [claim 27: unconfirmed], and a US Medicare FFS cohort of 621,171 HFpEF patients with rising MRA initiation [claim 29: unconfirmed]. None of these establish the actual moment this device targets — the between-visit draw itself. Whether outpatient RAASi titration in fact requires repeat venous draws between scheduled visits, and what that currently costs a patient in time and travel, has no source in the file at all until this deck's own write-back [claim 31: unverified].

What would settle it: Chart review or clinic-workflow interviews establishing the actual between-visit draw frequency and how patients currently obtain it (return to clinic, outside lab, home phlebotomy).

If it's false: If titration monitoring rarely requires an out-of-clinic draw — if it is folded into a visit that would happen anyway — the premise that this device removes a real burden collapses, and Mechanism & clinical risk and TAM both rest on a moment that doesn't occur as described.


Slide 2 — What they'd have to do differently

Would have to be true: A patient would have to wear a continuous adhesive multi-lead ECG patch between visits, and the prescribing clinician would have to receive an ECG-based hyperkalemia flag and act on it before a lab value confirms anything — ordering a confirmatory draw promptly rather than waiting for the next scheduled one.

Where it stands: This is the candidate's own named assumption, and the file is unusually direct about its status: "clinicians managing RAASi titration would be willing to act on an ECG-based hyperkalemia flag without an accompanying lab value, and would order a confirmatory draw promptly when flagged" was run through the adoption connector and returned unverifiable — "needs primary research... stays unconfirmed permanently unless a human attaches real research" [claim 12: unconfirmed]. The nearest real-world proxy is what clinicians do when a lab potassium value comes back high, not an ECG flag: in a US HFrEF registry, RAAS-inhibitor discontinuation ran 12.7%-20.4% by drug class over 18 months, with CKD the only independent predictor of increased discontinuation risk [claim 20: unconfirmed]. That is action on a confirmed lab number, the opposite of the behaviour this device asks for.

What would settle it: Nothing a connector in this repo can reach — the file says so explicitly. A prospective study or structured clinician interviews specifically asking about action on an ECG-only flag, not a lab value.

If it's false: The device generates alerts nobody acts on differently than they already do, and the entire clinical-workflow premise (screen → confirmatory draw → decision) never triggers the confirmatory draw promptly enough to matter.


Slide 3 — Who else has to say yes

Would have to be true: Beyond the prescriber and patient, a payer or the patient's own wallet would have to cover the device (see the viability deck's Slide 1 for who and from what budget), and — for the patient side specifically — a caregiver or the patient would have to manage a continuous adhesive patch and paired reader without clinical staff supervision.

Where it stands: The payer question is CMS's own, still-proposed RAPID pathway [claim 4: verified], [claim 5: verified], [claim 6: verified], [claim 7: verified] — verified as to what the notice says, not verified as to whether this device qualifies or whether the notice is ever finalized (comment period closed 2026-10-13). No HCPCS code, payment rate, or coverage policy specific to this service has been identified [claim 11: unconfirmed]. Whether a patient or caregiver can correctly apply, wear, and maintain the patch unsupervised for multi-day stretches is this deck's own write-back [claim 33: unverified] — nothing in the file addressed it before this pass.

What would settle it: A usability study with real patients/caregivers on device application and multi-day wear; the reimbursement questions are the viability deck's to settle.

If it's false: Either the money doesn't reach the device (viability's problem) or the device sits unworn in a drawer after week one (a desirability problem no reimbursement fix solves).


Slide 4 — What it costs them

Would have to be true: The burden of continuous wear, plus the rate of unnecessary confirmatory draws triggered by false alarms, would have to be smaller than the burden the device replaces.

Where it stands: The algorithm's own reported performance sets a real cost here. At the 90%-sensitivity operating point the Mayo model's specificity across three validation sites was only 54.7%-63.2% [claim 17: unconfirmed] — meaning a majority of flagged patients without true hyperkalemia would still need, and get, a confirmatory draw under this candidate's own intended-use framing. That is not a hypothetical cost; it is the shape of a false-alarm rate a patient would experience directly, sourced to the same study that is this device's strongest mechanism evidence. Out-of-pocket cost while no reimbursement pathway is finalized is the viability deck's Slide 1-2 question, referenced rather than restated here [claim 11: unconfirmed].

What would settle it: The false-alarm burden, translated from the Mayo cohort's specificity into an expected confirmatory-draw rate for this device's intended population — an extrapolation this file has not attempted and that a full-text read of PMID 30942845 could narrow.

If it's false: The device trades one burden (a scheduled venous draw) for a worse one (frequent unscheduled draws triggered by false alarms), and the desirability case inverts.


Slide 5 — What would make them believe it

Would have to be true: A prescriber would have to be able to point to a guideline mention, a colleague's experience, or a visible reduction in hyperkalemia-related complications before recommending this to a patient.

Where it stands: No claim in the file speaks to guideline recognition — this deck's own write-back records that no clinical practice guideline (KDIGO, ACC/AHA/HFSA) currently mentions ECG-based hyperkalemia screening as an adjunct to RAASi monitoring [claim 32: unverified]. The mechanism evidence a prescriber could cite instead is thin by design: single-institution, retrospective, abstract-only, with the file's own Verifier noting "not refuted... but not a precise match either" and full-text methodology still unread [claim 3: unconfirmed], [claim 17: unconfirmed].

What would settle it: A guideline search (KDIGO, ACC/AHA/HFSA) specifically for mention of non-invasive hyperkalemia screening; a full-text read of PMID 30942845 for methodology and any external validation.

If it's false: Prescribers have nothing to cite when a patient or a formulary committee asks why this device rather than the status quo, and adoption stalls on that gap regardless of what the underlying accuracy turns out to be.


Slide 6 — The cheapest way to find out

Would have to be true: In the first handful of real users, the confirmatory-draw rate triggered by device flags, and the time between a flag and a clinician acting on it, would have to be observably close to what claim 12 and claim 20 leave open.

Where it stands: The file has already run the cheapest available check on the central adoption question and it came back structurally unanswerable by any connector here [claim 12: unconfirmed]. The next-cheapest observation is not a connector call at all: a small number of real users generating real flags, with a clinician's actual time-to-confirmatory-draw recorded against the claim-20 lab-value baseline [claim 20: unconfirmed].

What would settle it: A small pilot cohort, instrumented for flag-to-draw time and false-alarm rate — the first primary research this candidate's desirability case has ever had.

If it's false: The pilot itself is the falsifier; there is no cheaper substitute.


Slide 7 — Where this deck outruns the file

Every condition above with nothing verified behind it. This candidate has four verified claims total, and all four concern the same thing — what CMS's proposed RAPID notice says, not whether this device qualifies for it or whether anyone wants it [claim 4: verified], [claim 5: verified], [claim 6: verified], [claim 7: verified]. Nothing verified in this file speaks to desirability at all.


Slide 8 — The load-bearing condition

If only one thing from this chair could be checked: whether a prescriber managing RAASi titration would act on an ECG-based hyperkalemia flag — ordering a confirmatory draw promptly — without an accompanying lab value [claim 12: unconfirmed].

The candidate's own Generator flagged this as an assumption rather than a fact, ran it through the one connector built to check adoption claims, and got back a structural answer: this stays unconfirmed permanently unless a human attaches real research. Every other desirability condition on this deck — the burden this device targets, the false-alarm cost, the guideline gap — describes context around this question. None of them resolves it, and nothing in this repo's tooling ever will.

Naming it is not a recommendation, a gate, or a kill. It is the answer to "check what first."