Hospice eligibility agent, at a glance

AttributeDetail
What it isAn AI agent for hospice eligibility and admission prioritization
Built forIntake coordinators, clinical leads and operations directors at hospice and palliative providers
Core use casesAutomated hospice eligibility check • Referral-to-admission conversion probability
Time per referralUnder 5 minutes
How rules are setPlain-language instructions edited by clinical staff. Live the moment you save.
Data layerSnowflake read and write. Azure Document Intelligence for scanned and unstructured documents.
TriggersIntake portal, EHR, CRM, API call, webhook or scheduled task
Decision modelThe agent screens and explains. Your clinical team reviews. A physician certifies terminal illness.

The problem

A hospice referral means a patient with weeks left, not years. Medicare covers hospice care when a person is entitled to Part A and a physician certifies a life expectancy of six months or less if the illness runs its normal course. The referral packet has to show clinical evidence that supports that prognosis, such as diagnosis, functional decline and decline indicators, for a family already in crisis.

When referral volume is high, intake teams cannot always review every packet quickly. A rushed review can admit a patient whose record does not support the prognosis, which risks payment denials and recoupment when the claim is reviewed. A slow review delays care for the patient and family.

The wider problem is the process. Reviewers can weigh the same evidence differently, and eligible referrals are not ranked against each other. Every guidance change can mean an engineering cycle, and the load lands on clinicians who should be at the bedside. That is the gap a hospice eligibility agent has to close.

What is the hospice eligibility agent?

The hospice eligibility agent screens each referral in four steps:

  • Reads the referral’s clinical documents.
  • Extracts key values, such as the prognosis statement and Palliative Performance Scale (PPS) score.
  • Applies your organization’s scoring rules.
  • Returns Eligible, Pending Review or Ineligible.

Caregence™ completes this in under five minutes per referral.

Use Case 1: How does hospice eligibility check work?

A referral submitted through your intake portal fires the workflow automatically within Caregence platform. Five steps run start to finish.

#1
Get Data

Pulls every patient document from the hospice system: physician certification, face-to-face encounter notes, decline indicators, admission form, clinical notes.
Output
Complete referral packet in one place

#2
Extract Data (OCR)

Azure Document Intelligence parses unstructured and scanned PDFs.
Output
Structured clinical values: PPS, prognosis, hospitalizations, decline indicators

#3
Intelligent Agent

Applies your custom scoring logic to the extracted values and reasons over the gaps.
Output
Total score, determination and the factors behind it

#4
Update Referral Status

Writes the determination back to the hospice system and the portal record.
Output
Eligible, Pending Review or Ineligible, visible instantly

#5
Send Email

Dispatches the outcome to whichever recipients you configured.
Output
Clinician notified with score breakdown and notes

Eligibility criteria and thresholds

Caregence™ looks for evidence in the referral documents:

  • A prognosis statement
  • Functional decline indicators
  • PPS score
  • Recent hospitalizations
  • Documentation completeness

It totals the points and returns a status. Your hospice management sets the weights and thresholds, because they differ by diagnosis and by Medicare contractor guidance. Caregence. by Inferenz, does not set the clinical standard.

Use Case 2: Which eligible referrals should you pursue?

Eligibility tells you who qualifies. It does not tell you who converts. When hundreds of eligible referrals compete for limited slots, the second workflow scores each one on its likelihood of becoming an admission, using your own historical data.That score becomes your admission priority order; the referrals that are most likely to convert rise to the top of the queue, without anyone having to manually reorder them.

#1A
Get historical data

Retrieves the organization's full admission history: past referral profiles, outcomes and discharge records

#1B
Get patient data

Fetches the current patient's clinical and referral profile

#2
Intelligent Agent

Pattern-matches the current patient against historical conversion data and applies weighted scoring

#3
Manage Data

Writes the probability score and its reasoning and the portal

#4
Send Email

Delivers the conversion report to intake coordinators and clinical leads

What drives the conversion probability score

The score looks at these factors, weighed against your own admission record:

  • Caregiver availability
  • Geography
  • Capacity
  • Documentation completeness

It also flags risks to admission: unclear family decisions, capacity limits and missing paperwork.

Caregence™ shows the score and the reasoning behind it. The score helps coordinators plan which referrals to work first. It does not decide whether a patient is admitted, and the coordinator makes the decision. It runs only on referrals already marked eligible.

The scoring logic is plain-language text inside the Intelligent Agent. Any authorized clinical lead can read and edit it.

Everything your team owns, without a developer

Logic & rulesProcess & flowIntegration & output
Eligibility criteria and point weightsMandatory versus optional documentsNotification recipients and conditions
PPS, prognosis and decline thresholdsWorkflow branching conditionsDatabase fields to read and write
Decision cut-off scoresTrigger source and escalation pathsCRM, EHR and scheduling outputs
Probability weighing factorsReview routing by patient typeFollow-up workflow triggers

Works with the intake portal you already have

Caregence is portal-agnostic. It sits above your systems as an orchestration layer rather than replacing them, so there is no migration.

Portal actions.

A button click or form submission in your existing intake portal

API and webhooks.

Any EHR, CRM or scheduling tool can POST to a Caregence endpoint

Scheduled runs.

Nightly batch eligibility reviews or weekly probability recalculations

Conditional chains.

An ‘eligible’ determination can automatically trigger the probability workflow

A timestamped record of every step

Every workflow step is timestamped against the patient record: the inputs pulled, the logic applied, the score produced, the result written and the email sent. Medicare medical review looks at the medical record, including the physician’s signed certification and narrative. This log supports those documents and does not replace them.

Manual intake versus AI intake agent

DimensionManual intakeCaregence
Time per determinationManual Varies with packet size and staffingCaregence Under 5 minutes per referral
Scoring consistencyVaries by reviewer and by dayCaregence The same configured rules applied to every referral
Changing the criteriaIT ticket and a development cycleText edit, live on save
Referral prioritizationIntuition and whoever called lastConversion probability score with reasoning
Audit evidenceManual Notes in the chartCaregence Timestamped log of every step
Capacity ceilingFixed by headcountScales with referral volume

Configure rules, logic and actions once. Admit right. Care better.

Bring one week of real referrals. We will run them through the workflow and show you the scores against your own past decisions.

Book a 20-minute walkthrough