All use cases

Prevent / Care

Prior-auth narrative drafter (clinician-side)

Your clinician needs to write the coverage justification for a GLP-1 prior authorization — and the patient's whole file already holds the facts it turns on. The draft is ready for them to read and sign.

Close the loopSystem triggerBehavioral triggerIn-appEmailGLP-1 & PeptidesHealth & TelehealthInsurance

What it does

A GLP-1 prior authorization lives or dies on the clinical narrative — the justification that ties the patient's BMI, qualifying conditions, and prior therapy to the plan's coverage criteria. As programs shift toward brand-name fills, that paperwork has surged, and writing each narrative from scratch is slow, repetitive clinician time that holds eligible patients at the coverage gate. The agent assembles the prior-auth narrative from the patient's record — pulling the criteria the plan actually requires and mapping the file's facts onto them — and presents it for the clinician to read, edit, and sign. It is the clinician-side counterpart to the patient-facing coverage navigator: the patient gets walked through their plan, the clinician gets the write-up ready for signature.

How it works

  1. 1
    Trigger. a prior authorization is required for a GLP-1 fill, or coverage navigation reaches the point where a clinical narrative is needed; the agent opens the drafter.
  2. 2
    Decision. the agent reads the plan's prior-auth criteria, pulls the patient's BMI, qualifying conditions, and prior-therapy history, and maps which documented facts satisfy each criterion and which are still missing.
  3. 3
    Action. it drafts the narrative into the plan's required structure, citing the record behind each clinical statement, and presents it in the clinician's tool to read, edit, or sign. A criterion with no supporting fact is flagged as a gap, not invented.
  4. 4
    Follow-through. the signed narrative attaches to the authorization and submission status is tracked; a denial routes to the appeal coach; the clinician signs every narrative, and the agent never asserts a coverage outcome the payer or clinician owns.
Example
Agent (narrative for review)Draft ready for the authorization. It maps her BMI and documented sleep apnea to the plan's criteria and cites prior lifestyle therapy from the record. One gap: the plan also asks for a duration on the prior attempt, which isn't captured — I've flagged it rather than filled it.
ClinicianDuration was about six months — add it and I'll sign.
AgentAdded and sourced. The narrative is ready for your signature, then it attaches to the submission and I'll track where it lands.

Configuration

How the agent is wired for this use case.

Triggera prior_auth_required event for a GLP-1 fill, or a coverage-navigation step that hands off for a clinical narrative
Tools & actions
  • Clinical/EHR system · read BMI, qualifying conditions, and prior-therapy history for the patient
  • Knowledge base · pull the plan's prior-auth criteria and the required narrative structure
  • Clinical/EHR system · draft the narrative, citing the record behind each clinical statement, and present it for clinician edit and signature
  • Billing system · attach the signed narrative to the authorization and track submission status
  • Messaging channel · on a denial, route the case to the appeal flow
Autonomydrafting and criterion-mapping run unattended; the clinician reviews and signs every narrative — clinical = assemble the justification, never diagnose or advise a dose; a missing criterion is flagged as a gap, never fabricated, and no coverage outcome is asserted
Channelsin-app · email
Escalationthe clinician owns the signature on every narrative; a criterion that cannot be evidenced from the record is surfaced to the clinician rather than guessed, and a denial is handed to a human-supported appeal path

What you need

The inputs this use case runs on. Your channels stay yours; the agent supplies the judgment.

Signals

prior_auth_required events; coverage-check results that determine when a narrative is needed; denial events that trigger the appeal route

Data

the patient's BMI, qualifying conditions, and prior-therapy history; the plan's prior-auth criteria and required narrative format; consent to use the record for the authorization

Guardrails

clinician signature required on every narrative; source attribution on every clinical statement; no fabricated criteria and no asserted coverage decision; clinical = assemble, never advise a dose; a full audit log of drafts, edits, and signatures

Metrics it moves

  • clinician-hours-savedup, as narrative writing collapses into reading and signing a sourced draft
  • application-completionup, because eligible patients clear the coverage gate instead of stalling on unwritten paperwork
  • time-to-resolutiondown, since the justification is ready at the moment the authorization is needed

Related use cases

Convert

Insurance coverage check and prior-auth navigator

the patient-facing counterpart that walks the plan and gathers facts

Retain / Save

Coverage-denial appeal coach

the next step when a submitted authorization is denied

Prevent / Care

Clinician copilot — approve/decline decision packet

the same record assembled into a start decision rather than a coverage narrative

See it on your own customer journey

Bring one drop-off, one churn cliff, or one silent segment. We will show you what a proactive agent with memory and judgment does with it.

Book a demo