The decline, shown — serial weights, functional scores, wounds, hospitalisations.
Hospice eligibility documentation review reads a long-stay chart of two hundred to fifteen hundred pages into the longitudinal decline narrative the coverage determination asks for: serial weights, functional scores, wound progression, and hospitalisations, each cited to its source page, before the probe letter arrives rather than after it lands.
The manual is published. So is the reviewer.
We do not build for a case type until we can name the standard that defines a correct output and the person who grades the file against it. In hospice eligibility documentation, both are written down, and so are the denial rates each round of review produces.
Long-stay charts run 200 to 1,500 pages of the same few numbers.
A chart under review is rarely one production. It arrives as the admission packet, the interim IDG notes, a recertification packet for each benefit period, and whatever the referring hospital sent across. Two hundred pages is a short one; fifteen hundred is a long-stay patient across several benefit periods, with roughly 850 pages sitting near the middle of the range.
Length is not what makes it hard. The evidence of decline is scattered: serial weights buried in nursing flowsheets, PPS and FAST scores recorded unevenly from visit to visit, wound measurements in handwritten aide notes, hospitalisations visible only in a discharge summary someone scanned crooked. Recertification packets repeat the prior period wholesale, so page count grows while evidence does not.
The volume behind it is not small. In 2024, 1.91 million Medicare beneficiaries were enrolled in hospice, up 4.4% year over year and 53.1% of Medicare decedents. That population produces roughly 6.7 million certifications of terminal illness a year, of which the audited subset runs on the order of 100,000 to 250,000 charts.
Four steps, starting with one real chart.
We build case types with the people who work them, not ahead of them. The read comes back before anything is scoped, so you can judge the output against a chart you already know.
Send one real file
One chart, as you already hold it: EMR export, scanned packet, referring hospital production, or all three in the same upload. Everything is handled under a signed BAA from the first byte.
We return the read
The decline record as we would build it: serial weights, functional and PPS scores, wound measurements, hospitalisations and interval changes, each dated and cited to the page it came from. In the demo case, 342 pages across 2 packets, with a wrong-patient page quarantined before the read starts.
We scope the build
Your output format, your benefit-period structure, the measures your medical director wants surfaced first, and the export your compliance team already files with. If the case type is not buildable to that standard, we say so.
You run it
Your clinical and compliance staff run charts through it on your own schedule. The certification, the narrative, and the response to any reviewer stay with the people licensed to make them.
The rules the platform never breaks.
HIPAA, under a signed BAA
Every file is handled under our Business Associate Agreement, from the first byte.
Never trains a model
Your records are never used to train any AI model — ours or anyone else's.
Every line cited
If we can't cite it, we don't say it. Every sentence links to its source page.
Deleted after delivery
Files are deleted 30 days after delivery, with a full audit log of every access.
Four capabilities behind every decline record.
The hospice decline record is one deliverable of the same platform that reads, sorts, and cites the whole chart.
Condition Progression Tracking
Interval change across benefit periods, laid out in the direction it actually moved rather than restated period by period.
Medical Table Extraction
Serial weights, vitals, albumin, and functional scores lifted out of nursing flowsheets as dated values, not as a scanned grid.
Medical Chronology
The admission-to-current timeline builds itself from visit notes, IDG notes, and hospital discharge summaries, synced to every source page.
Missing Records Identification
Gaps named before a reviewer names them: a benefit period with no recorded weight, a face-to-face encounter with no document behind it.
Hospice eligibility documentation, answered.
Send one chart. We'll tell you what we can read.
No obligation. If the case type is buildable we'll scope it; if it isn't, we'll say so.