Prior authorization APIs required January 1, 2027

Your medical policy, running as a DTR questionnaire.

Payers already publish the rules — in bulletins, medical policies and clinical editing updates written for humans. The prior authorization rule requires those same rules to arrive at the point of care as a FHIR Questionnaire that fills itself in. PolicyCodify does that conversion, and keeps it current when the policy changes.

What you publish today

Medical Policy MP-114 — Lumbar MRI

Advanced imaging of the lumbar spine is considered medically necessary when conservative therapy of at least six weeks has been completed and documented, or when the member presents with progressive neurologic deficit, suspected malignancy, or suspected infection.

Documentation must include the duration and type of conservative treatment and the findings on physical examination.

What DTR needs

// Questionnaire + CQL, prepopulated
{
  "resourceType": "Questionnaire",
  "item": [{
    "linkId": "conservative-tx",
    "text": "Weeks of conservative therapy",
    "type": "integer",
    "extension": [{
      "url": "...cqf-expression",
      "expression": "ConservativeWeeks"
    }]
  },{
    "linkId": "neuro-deficit",
    "text": "Progressive deficit?",
    "type": "boolean"
  }],
  "enableBehavior": "any"
}

Policy prose in. Questionnaire and CQL out.

The gap is not that payers lack rules. It is that the rules exist as sentences, and the regulation needs them as artifacts a provider's EHR can execute without a phone call.

01
Read the policy as published

Bulletins, medical policies, clinical editing updates, coverage criteria — in the format you already maintain, including the PDF nobody wants to touch.

02
Extract the decision logic

Every condition, threshold, duration, exception and documentation requirement, separated from the prose around it and mapped to the data an EHR can actually supply.

03
Emit the artifacts

FHIR Questionnaire with CQL expressions for prepopulation, structured so the clinician answers only what the record cannot already answer.

04
Keep it current

Policies change quarterly. The conversion is re-run against the new version and the diff is reviewable — so the artifact and the published policy never drift apart.

Why this is harder than it sounds

Every payer that has tried this in-house has hit the same four walls.

01

Prose hides structure

"Or", "unless", "at least", "documented" — one sentence can carry three conditions and an exception. Getting that wrong makes a questionnaire that asks for the wrong evidence.

02

Prepopulation is the whole point

A questionnaire that asks a clinician to retype what is already in the chart will be ignored. The value is in the CQL that answers before anyone is asked.

03

The policy moves

A one-time conversion is a project. What a payer actually needs is a pipeline that survives the next quarterly update without a rebuild.

04

Nothing tells you it is wrong

An artifact that asks the wrong question still validates, still returns 200, and still looks correct — right up until the documentation does not support the authorization.

Who this is for

Anyone who owns published clinical criteria and now owns a deadline.

Health plansCommercial, Medicare Advantage, Managed Medicaid and Marketplace lines with published medical policy.
Third-party administratorsRunning utilization management on behalf of plans and self-funded groups.
UM & PA vendorsPlatforms that need to ship DTR support without staffing a policy-analysis team.
Delegated groupsRisk-bearing provider organizations administering their own criteria.

Start with one policy.

Send a single published policy and we will return the Questionnaire and CQL it produces, so the question stops being theoretical. No cost for the first one, and no obligation after it.

We reply from a person, not a sequence. Your policy is not shared, and not used to train anything.

Thank you — we will be in touch shortly.

Who is behind it

PolicyCodify is built by people who have spent more than twenty years on the payer side of American healthcare — claims operations, core administrative platform implementations, utilization management and payment integrity, across Medicaid managed care, Medicare Advantage, Marketplace and commercial lines.

That work included interoperability delivery against the federal prior authorization and patient access requirements, and production systems that read published rules and turn them into logic that runs. This is the same problem, aimed at the documentation side.

We are a small team by choice. You will talk to the person doing the conversion, not to an account manager.