idea-019 · 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. An adult with Parkinson disease, multiple system atrophy or pure autonomic failure who stops standing because standing makes them grey out; the spouse or carer who would put the garment on them each morning; the neurologist who would have to write it instead of a pill.
How to read this: every slide is a condition, not a conclusion. A bracketed claim reference points into knowledge-base/candidates/idea-019.md — the number is the claim's position in its ## Claims list, and the tag is copied from that claim as it stands right now, never adjusted here. A [no claim] marker means nothing in the file speaks to the condition at all. See docs/deck-spec.md.
What this candidate's evidence looks like from this chair. The file is unusually strong on the literature and unusually empty on the user: claims 5, 6 and 7 verify a published randomized comparison and two registered trials of this exact mechanism, while every statement about what a patient, carer, prescriber or supplier would actually do is unconfirmed by the Generator's own labelling (claims 28, 29, 30) or newly on record here as unverified. The rubric scores no factor for whether a user wants this, so that gap is not visible in any score in the file.
Would have to be true: There would have to be a substantial US population of adults with primary autonomic failure who are still symptomatic on standing after non-pharmacologic measures — people whose day is shaped by how long they can stay upright before presyncope.
Where it stands: The clinical fraction is the best-evidenced number in the file: a systematic review of 55 studies and 10,463 patients with Parkinson disease gives pooled orthostatic hypotension prevalence of 33% (95% CI 28-37%), the neurogenic subtype at 26%, and symptomatic OH at approximately 20%, with North America at 28% and a trim-and-fill adjustment to 24.9% [claim 22: verified]. The denominator it multiplies is not: the approximately one million US Parkinson patients and approximately 90,000 new diagnoses a year are permanently unconfirmed absent a human-attached source [claim 21: unconfirmed], and no claim in the file carries any prevalence figure for multiple system atrophy or pure autonomic failure at all [no claim] — even though both are named in the intended use. That "symptomatic" is a real, enrollable state is corroborated by a registry record: NCT03482297 enrolled MSA and PAF patients with an OHSA composite of at least 4 [claim 6: verified], and 19 autonomic-failure patients with measurable orthostatic symptom burden were studied in the published crossover [claim 5: verified].
What would settle it: The Scores table names the source for the denominator — CMS Chronic Conditions Data Warehouse Parkinson-disease prevalence among Medicare FFS beneficiaries, a human fetch, since the market connector returns unverifiable by design and the CCW figure is in any case a Medicare-FFS floor. For MSA and PAF, a named epidemiological source has to be attached as a new claim before anything can be multiplied.
If it's false: Factor 3's population base is the only partially-evidenced half of the TAM row, and it is carried by claims 21 and 22 together. If 21 does not hold, the rate in 22 has nothing to multiply, and every downstream slide on this deck is about a population nobody has sized.
Would have to be true: A person with Parkinson disease, MSA or PAF — a population in which tremor, rigidity, bradykinesia and impaired dexterity are common — would have to don a powered abdominal garment every morning, keep it charged, and still be doing so months later, rather than abandoning it after the first few weeks.
Where it stands: This is the file's own flagged assumption, and it is flagged as permanently unconfirmed absent primary research [claim 28: unconfirmed]. Nothing has been observed, because no device of this design exists as a prototype and no bench, animal or human data exist for it [claim 8: unconfirmed]. The two registered binder studies cannot stand in: NCT03482297 is a study-day protocol [claim 6: verified] and NCT01971008 used an elastic binder in a short crossover [claim 7: verified], so neither measures sustained home use, which is exactly what this condition asserts. The tolerability half — whether a wearer keeps using it through discomfort, breathlessness or skin irritation — is now on record as an assumption [claim 39: unverified].
What would settle it: Two steps, cheapest first. NCT01971008 has posted results (hasResults=true, recorded at claim 7) which nobody in this repo has read — pull them for discontinuation, adverse events and adherence. Then a home-use diary or device-abandonment study in an autonomic clinic, which is human-run: the adoption connector refuses this class of question by design and a model's estimate of it would be fabricated, not weak.
If it's false: The candidate has no user. Claim 24's addressable share explicitly filters on "sufficient manual dexterity or caregiver support to don and operate a powered abdominal garment daily", so Factor 3's reachable share goes to whatever fraction can, and the mechanism evidence in claims 5-7 becomes irrelevant to a product nobody wears twice.
Would have to be true: Three other parties would each have to say yes on the same prescription: a neurologist or autonomic specialist would have to write a DME prescription instead of, or alongside, a drug; a DMEPOS supplier would have to stock, fit, deliver and bill a device that has no code of its own; and a carer would have to be available on the mornings the patient cannot manage the garment alone.
Where it stands: Prescriber willingness, including acceptance of the DME documentation and prior-authorization burden, is an explicit assumption and permanently unconfirmed absent interviews or a preference study [claim 29: unconfirmed]. The shape of that burden is a verified fact rather than an opinion, and it is adverse: NCD 280.6 conditions coverage of pneumatic compression on a documented four-week conservative-therapy trial for lymphedema and a six-month one for CVI ulcers, with physician oversight, a written treatment plan defining pressure and frequency, and ongoing monitoring [claim 9: verified]. The supplier link in the chain had nothing behind it and is now on record [claim 37: unverified], against a file that identified no HCPCS code specific to this device [claim 10: unconfirmed]. Whether Medicare pays at all is the viability deck's Slide 2, referenced rather than restated here [claim 41: unverified]. Carer availability is folded into claim 28's "with or without caregiver help" [claim 28: unconfirmed].
What would settle it: Structured interviews with autonomic-specialist prescribers at movement-disorder centres, using NCD 280.6's own conditions of coverage as the concrete burden description; and interviews with national DMEPOS suppliers about willingness to carry a non-coded, powered, prescription garment. Neither is reachable by any connector in this repo — both are human acts.
If it's false: A prescription device with no prescriber and no distribution channel never reaches a patient, whatever the mechanism does. Factor 2 already scores the missing code as an obstacle; a missing channel is the same obstacle one step further downstream, and it is not scored anywhere.
Would have to be true: The user would have to accept the whole cost, not the brochure half: minutes of donning help every morning, an inflating bladder worn under clothing in public, sustained pressure on the abdominal wall through every upright period, the risk of the device inflating at the wrong moment — and whatever money is left after Medicare pays.
Where it stands: There is no price anywhere in this file. The Generator asserts no selling price and no allowable, and the checkable half — the published DMEPOS allowable for E0650/E0651/E0652 — could not be looked up because data/hcpcs_payment_rates.csv is absent from this environment [claim 23: unconfirmed]. So the out-of-pocket residual cannot be stated, only assumed, and it is now on record as an assumption [claim 36: unverified]. The comparator cost is a pharmacy-benefit copay for midodrine or droxidopa, which is itself unconfirmed for a tooling reason — this repo has no drug connector at all [claim 11: unconfirmed]. The bodily costs are uncharacterized in both directions: skin and abdominal-wall injury under sustained pressure, interference with respiration or with abdominal stomas and feeding tubes, and reduced effectiveness at high BMI are named failure modes of an unbuilt device [claim 31: unconfirmed], and the wrong-moment risk is the feasibility deck's Slide 4 [claim 48: unverified]. Whether the wearer tolerates the pressure well enough to keep going is claim 39 [claim 39: unverified].
What would settle it: The DMEPOS fee-schedule allowable for E0650-E0652 once the cache lands — the file names the exact missing file — which gives the Part B coinsurance base a patient would face; then willingness-to-pay interviews against a real number rather than a hypothetical one. Tolerability data can be pulled cheaply from NCT01971008's posted results before anything is built.
If it's false: The comparator is a tablet on the pharmacy benefit. A device that costs more out of pocket and more effort per day than a pill has to win on something else, and the something else is claim 30 — the supine-pressure argument on Slide 5, which is itself unevidenced as a preference.
Would have to be true: Two different beliefs would have to land. A named consensus guideline for managing neurogenic orthostatic hypotension would have to recommend abdominal compression, so a prescriber can point at something; and the patient would have to feel a difference in the first weeks — more standing time, fewer grey-outs — or they will stop.
Where it stands: The guideline half cannot be checked as written: claim 4 names no PMID, guideline title or document identifier, so the literature-detail connector has nothing to run against, and the Verifier correctly declined to go hunting for a document the Generator never identified [claim 4: unconfirmed]. The published evidence that does exist is real and it is about a different device: PMID 27271310 reports symptom burden falling 21.9±3.6 → 16.3±3.1 (P=0.032) with the servo-controlled binder at 40 mmHg versus no change on placebo, in 19 autonomic-failure patients [claim 5: verified] — but that is the Vanderbilt investigational binder, and no data of any kind exist for this candidate's device [claim 8: unconfirmed]. The sham-controlled confirmation that a sceptical clinician would ask for is registered, completed, and unpublished: hasResults=false [claim 6: verified]. The week-one experience a patient would need is now on record as an assumption [claim 38: unverified], as is the belief that absence of supine pressure exposure reads as a meaningful advantage rather than a theoretical one [claim 30: unconfirmed].
What would settle it: Append a specific nOH consensus-guidance identifier as a new claim — the file says in terms that claim 4 needs one before it can move — and run literature-detail against it. Separately, re-check NCT03482297 for posted results, and ask the Vanderbilt investigators whether an outcome manuscript exists; that is a human email, not a connector call.
If it's false: With no guideline to cite and no published outcome for this device, the prescriber in Slide 3 has nothing to act on and claim 29 fails for a reason that has nothing to do with willingness. Mechanism & clinical risk is the strongest row in the file's Stage 2 draft table, and it rests on claim 5 — which is about somebody else's device.
Would have to be true: In the first handful of real users, you would have to see the garment still being put on daily after the novelty period, prescribers writing a second prescription for a second patient, and patients describing a change they noticed themselves. Any user stopping because of donning difficulty, abdominal discomfort, or the device firing while they were seated would falsify the conditions above.
Where it stands: None of this can be observed yet, because there is no prototype [claim 8: unconfirmed]. The cheapest real act available today needs no hardware at all: NCT01971008 has posted results that have never been read here [claim 7: verified], and NCT03482297 measured upright time as a secondary outcome [claim 6: verified] — but claim 28's own note is the right caution, that both are short study-day protocols and neither measures sustained home use [claim 28: unconfirmed].
What would settle it: Pull the posted NCT01971008 results via trial-detail and read the adverse-event and discontinuation tables — free, minutes, and the identifier is already in the file. Then a structured interview round with autonomic-clinic patients, carers and prescribers, which no connector reaches and which a human has to commission.
If it's false: If even the posted trial results show poor tolerability of binder wear in this population, claims 28 and 39 are in trouble before a single engineering dollar is spent — and that is the cheapest possible order in which to find out.
Every condition above with nothing verified behind it. Read this slide first. Five of the six conditions on this deck are in this position, and after write-back most of them now carry a claim number — a short [no claim] list is not good news here, it means the assumptions are recorded, not answered.
data/hcpcs_payment_rates.csv lands, then
willingness-to-pay interviews.
If only one thing from this chair could be checked: whether adults with Parkinson disease, MSA or pure autonomic failure will actually don, charge and keep using a powered abdominal garment every day over months — [claim 28: unconfirmed].
Why this one rather than the others. Every other condition on this deck is downstream of it: a prescriber's willingness (claim 29) is worth little if patients abandon the device, the out-of-pocket question (claim 36) only bites on a device still in use, and the mechanism evidence in claims 5-7 describes study-day performance in a supervised setting. It is also the one condition where the disease itself works against the product — the intended population is defined by a movement disorder, and the device asks for daily fine-motor work. The file has already labelled it an assumption; this deck's contribution is to say it is the first thing to check and that the posted results of NCT01971008 [claim 7: verified] are the cheapest place to start.
Naming it is not a recommendation, a gate, or a kill. It is the answer to "check what first."