Why do imaging claims get denied even when prior authorization was obtained?

Because authorization is tied to a specific CPT code, modality, contrast protocol, facility, and time window — and any mismatch between what was authorized and what was actually performed or billed triggers a denial, regardless of whether the study was clinically appropriate. Premier's national provider survey found a meaningful share of denied private-payer claims were already prior-authorized before they were denied.

What's the difference between a clinical denial and an operational denial in imaging?

A clinical denial disputes medical necessity. An operational (front-end) denial disputes a process detail — the wrong CPT code, an expired authorization, a site-of-service mismatch — even when the study itself was medically appropriate. The five failures in this article are all operational, not clinical.

Can an expired prior authorization be fixed after the study is performed?

Rarely. Most payers will not retroactively authorize a study that's already been completed. That's why authorization expiration needs to be re-verified at every reschedule, not just at the original scheduling date.

What is a modality mismatch in imaging billing?

It's a gap between what a payer authorized (a specific modality, contrast protocol, or body region) and what was actually performed — for example, an authorization for MRI without contrast when contrast was clinically added, or an authorization for CT abdomen when the study performed was CT abdomen and pelvis.

Does Corena diagnose or interpret imaging studies?

No. Corena is an operations agent that checks studies for read-ready completeness — including authorization, coding, and modality gaps — before the radiologist reads them. It recommends what to fix; a named human on your team reviews and approves. It never interprets, reads, or diagnoses.

Prior authorization·July 16, 2026·7 min read

5 Front-End Failures Behind Imaging Denials

Most imaging denials aren't clinical calls — they're operational breakdowns that happen before the study is ever read. Here are the five most common ones, and how to catch each before it turns into a rejected claim.

Imaging claims are most often denied for five front-end, operational reasons: a modality mismatch between what was authorized and what was performed, an authorization that expired or lapsed before the study date, a CPT or contrast code that doesn't match the study actually done, a site-of-service error, and missing or incorrect prior authorization altogether. Payers initially denied 11.8% of claims in 2024, up from 11.5% the year before (Kodiak Solutions), and none of that requires a radiologist to have made a clinical error — it requires a gap between what the front desk scheduled, what the tech performed, and what the coder billed.

Denials are a workflow problem, not a clinical one

It's tempting to treat imaging denials as a billing-department fire to fight after the fact. But most of the failure points that generate them happen upstream — at scheduling, at check-in, at the modality, at coding — long before a claim ever reaches a payer's desk. Kodiak Solutions' 2024 revenue cycle benchmarking, drawn from over 2,100 hospitals, put the initial claim denial rate at 11.8%, up from 11.5% in 2023 and well above pre-pandemic levels. Separately, a national Premier survey found that private payers deny close to 15% of all claims initially, and that a meaningful share of those denials hit claims that were already prior-authorized — the authorization existed, and the claim still bounced.

That second number matters most for imaging operations, because it means "we got the auth" is not the finish line. The five failures below are the specific, recurring ways an authorized, medically appropriate study still ends up denied — and each one has a catch point that exists before the study is billed, in most cases before it's even read.

1. Modality mismatch — what was authorized isn't what was performed

The payer authorized an MRI without contrast. The radiologist's protocol, or the patient's clinical presentation at the table, called for contrast. The tech does the right clinical thing and adds it — and the claim goes out for a study the authorization never covered. This also shows up as a swapped modality entirely (CT authorized, ultrasound performed after a scheduling error) or a body-region mismatch, where a CT abdomen authorization doesn't cover a CT abdomen and pelvis — they're different CPT codes with different authorization rules, and payers routinely deny the mismatch.

Why it happens operationally: authorization is locked in at scheduling, days or weeks before the study, but the final protocol is often a clinical decision made at or near the point of care. Nobody reconciles the two.

How to catch it early: compare the authorized CPT/modality against the actual performed protocol before the claim drops — not after. If a tech or radiologist changes the protocol at the table, that change needs to trigger a same-day check against the auth on file, not a discovery three weeks later in the coding queue.

2. Expired or lapsed authorization

Authorizations carry an expiration window, typically 30 to 90 days depending on payer. Studies get rescheduled — patient illness, equipment downtime, insurance changes — and the new date slides past the authorization's validity window without anyone re-checking it. The study happens, it's clinically appropriate, and it's still denied because the paperwork expired.

Why it happens operationally: authorization status is checked once, at the time it's obtained, and treated as a permanent fact rather than a fact with a shelf life. Rescheduling workflows in most scheduling systems don't automatically re-flag authorization validity.

How to catch it early: treat every reschedule as a trigger to re-verify authorization status and expiration date, not just appointment availability. The check needs to happen at confirmation, not at check-in the morning of.

3. CPT and contrast code misalignment

This is a coding-and-documentation failure, not a scheduling one. The three contrast variants — without, with, and with-and-without — are separate CPT codes, and the code billed has to match what the radiologist's report documents was actually administered. A common version: the request is authorized and coded as CPT 70551 (brain MRI without contrast), but the study performed and reported is with-and-without contrast, which requires CPT 70553 — a code the original authorization never covered. Multi-phase studies have a parallel trap: billing a single-phase code when a multi-phase protocol was actually run.

Why it happens operationally: the order, the authorization, and the final report are handled by three different people (scheduler, tech, coder) at three different points in time, and none of them is looking at all three documents side by side.

How to catch it early: before the claim is coded, reconcile the authorized CPT code against the radiologist's documented protocol — contrast administered, phases run, region imaged. If they don't match, that's a hold-and-fix, not a bill-and-appeal.

4. Site-of-service errors

Authorizations are frequently facility-specific — a hospital outpatient department authorization doesn't transfer to a freestanding imaging center, and vice versa, even for the identical CPT code and identical clinical indication. Patients get rerouted for scheduling convenience, equipment availability, or network changes, and the authorized facility doesn't travel with them.

Why it happens operationally: site assignment is often a capacity or logistics decision made independently of the authorization record, especially in multi-site imaging networks and RBM-managed environments.

How to catch it early: whenever a study is rerouted to a different location than originally scheduled, treat the authorized site as a hard constraint to re-verify — not a detail that carries over automatically.

5. Missing or incorrect prior authorization

The most direct failure: the study is performed without an authorization on file at all, or with an authorization number that's incorrect, mistyped, or belongs to a different patient or study. Once an advanced-imaging study — MRI, CT, PET, nuclear medicine — is performed without confirmed authorization, the denial is close to non-recoverable: payers generally won't retroactively authorize a study that's already been done.

Why it happens operationally: urgent add-ons, same-day orders, and manual transcription of authorization numbers between systems all create openings for this to slip through, particularly under time pressure.

How to catch it early: authorization status has to be a hard gate before the study is performed, not a box checked after — with the number itself validated against the specific patient, CPT code, and facility, not just confirmed as "present."

11.8%

of claims were initially denied by payers in 2024, up from 11.5% in 2023 — Kodiak Solutions, 2024 revenue cycle benchmarking

~15%

of private-payer claims are initially denied, and a meaningful share of those were already prior-authorized before the denial — Premier Inc. national provider survey

None of these five failures requires a diagnostic error. They're gaps between what was scheduled, what was authorized, what was performed, and what was billed — and every one of them is catchable before the claim goes out, if something is checking for the mismatch at the right moment.

Where this fits before the claim, not after

Corena is a read-ready QC agent — it checks a study for operational completeness before a radiologist reads it, not after a claim is denied. As part of that pre-read pass, Corena flags exactly these kinds of gaps: an authorization that's expired, a CPT or contrast code that doesn't match the protocol actually run, a modality or site mismatch against what's on file. It surfaces the flag and routes it to a named human on your team to review and approve or resolve. Corena does not interpret images, diagnose, or make the final call — it recommends, and a person decides. The goal is to catch the operational gap while the study is still in the queue, before it becomes a denial you're appealing weeks later.

Request a demo to see how Corena's pre-read QC flags auth, coding, and modality gaps before they reach billing.

Frequently asked questions

Why do imaging claims get denied even when prior authorization was obtained?: Because authorization is tied to a specific CPT code, modality, contrast protocol, facility, and time window — and any mismatch between what was authorized and what was actually performed or billed triggers a denial, regardless of whether the study was clinically appropriate. Premier's national provider survey found a meaningful share of denied private-payer claims were already prior-authorized before they were denied.
What's the difference between a clinical denial and an operational denial in imaging?: A clinical denial disputes medical necessity. An operational (front-end) denial disputes a process detail — the wrong CPT code, an expired authorization, a site-of-service mismatch — even when the study itself was medically appropriate. The five failures in this article are all operational, not clinical.
Can an expired prior authorization be fixed after the study is performed?: Rarely. Most payers will not retroactively authorize a study that's already been completed. That's why authorization expiration needs to be re-verified at every reschedule, not just at the original scheduling date.
What is a modality mismatch in imaging billing?: It's a gap between what a payer authorized (a specific modality, contrast protocol, or body region) and what was actually performed — for example, an authorization for MRI without contrast when contrast was clinically added, or an authorization for CT abdomen when the study performed was CT abdomen and pelvis.
Does Corena diagnose or interpret imaging studies?: No. Corena is an operations agent that checks studies for read-ready completeness — including authorization, coding, and modality gaps — before the radiologist reads them. It recommends what to fix; a named human on your team reviews and approves. It never interprets, reads, or diagnoses.

Sources

See Corena check a study for read-readiness — before it reaches the radiologist.

Corena flags what is missing and routes exceptions to a named human. Clean studies keep moving; they do not wait on a click.