Understanding prior authorization: a practice guide

What prior authorization is

A payer's requirement that it approve a service before it is delivered, as a condition of coverage. It applies to some procedures, imaging, medications, durable equipment and courses of treatment, and it varies by payer, by plan and — importantly — by product line within the same payer.

An authorization is not a guarantee of payment. It is a determination that, based on the information submitted, the service meets the plan's coverage criteria. The claim still has to be correct, the patient still has to be eligible on the date of service, and the service delivered still has to match what was approved.

The steps, and where each one fails

Identifying the requirement

Before anything else, someone has to know whether authorization is needed. Getting this wrong in one direction wastes effort on unnecessary requests; getting it wrong in the other produces a denial for a service already delivered, when retroactive approval is often unavailable.

The reliable place to catch this is during eligibility verification, at the point the service is scheduled.

Assembling the documentation

Payers publish coverage criteria, and reviewers apply them. Broadly, they are looking for:

• The diagnosis, and how long the condition has been present
• Severity, and its functional impact on the patient
• Conservative treatments already tried, for how long, and with what result
• Objective findings — examination, imaging, standardised assessment scores where relevant
• Response to any previous related intervention
• The clinical rationale for the specific service requested now

Requests that fail usually fail on one of these, most often conservative treatment history or objective findings. A note that says the patient needs the service is an assertion; the criteria ask for evidence.

Submitting

Through whichever channel the payer requires — portal, electronic transaction, phone or fax. Electronic prior authorization standards are improving and regulatory pressure is pushing payers toward faster, more automated processing, but the landscape remains inconsistent and manual channels are still common.

Following up

This is where most authorization processes break. A submitted request that nobody chases can sit until the appointment date arrives. Every open request needs an owner, an expected turnaround, and a next-follow-up date. Reference numbers and representative names should be logged so nothing depends on somebody's recollection of a phone call.

Recording the determination

The authorization number, approved codes, approved units or visits, and the valid-from and valid-to dates need to be on the account where scheduling and billing can both see them. An approval that exists only in an email thread will not be reconciled against the claim.

Monitoring units and expiry

Approvals are finite. A certain number of visits, sessions or units, valid for a defined window. Somebody has to watch consumption against the approval and trigger re-authorization before the limit is reached — not after. This is a recurring obligation, not a one-time task, and it is where courses of treatment quietly become unpaid.

When a request is denied

Get the specific reason, not the summary. Then determine which of three situations you're in:

• Documentation gap — the criteria weren't addressed in the submission. Resubmit with what's missing.
• Administrative error — wrong code, wrong provider, wrong plan, missing information. Correct and resubmit.
• Genuine criteria mismatch — the payer's position is that the service isn't covered for this presentation. This is where appeal or a peer-to-peer review between your provider and the payer's reviewer becomes the route.

Peer-to-peer reviews are worth scheduling promptly when the clinical case is strong. They put a clinician in front of a clinician, and they are often more productive than another round of paperwork.

Making it survivable

Practices that handle authorization well generally do a few unremarkable things consistently:

• Check the requirement when the service is scheduled, not when it's performed
• Maintain a payer-by-payer reference of what needs authorization and through which channel
• Give every open request a named owner and a follow-up date
• Track approved units and expiry dates in one place everyone can see
• Give clinicians a short template of what the payer's criteria actually require
• Review denied requests for patterns — the same payer denying the same service repeatedly is information

None of that is clever. All of it is the difference between authorization being a background process and authorization being the thing that ruins everyone's Thursday.

---

General information about US healthcare revenue cycle management. Not legal, compliance or clinical advice. Payer authorization requirements vary and change frequently; verify current requirements with the applicable payer.