STAT vs preliminary vs final radiology reads, explained

"Is a STAT read just a fast preliminary that somebody else checks and signs in the morning?" Buyers ask us that in almost those words, and the question collapses two separate things. STAT sets a clock. Preliminary and final reads tell you what the document is. A preliminary read is a rapid initial interpretation issued to guide immediate care before the report of record exists, and a later signature always supersedes it. A final signed report is the complete interpretation from a board-certified radiologist, and it is the only version that is billable and legally definitive. A STAT study can come back as either one, so the tier by itself tells you nothing about what lands in the chart.

Peer-reviewed comparisons put major discrepancies between preliminary and final interpretations at 2 to 3 percent, with 2.6 percent the figure quoted most often.

STAT vs preliminary vs final: separate the clock from the document

The urgency tier and the report status answer different questions: STAT says how fast, preliminary and final say whether what arrives is the record.

A STAT study can get a preliminary, a final, or one after the other. A routine outpatient MRI usually goes straight to final. Contracts that blur the two axes produce one specific surprise: a vendor promising STAT reads in 30 minutes may be promising an impression, with the report of record arriving hours later over a different physician's name.

So put two questions to every tier in a contract. How fast does the report arrive, and is it signed when it arrives? Then push on the second one: which physician signs the report of record, and at what hour. If the answer involves a different radiologist the next morning, you're buying preliminary coverage whatever the tier is called on the price sheet.

You'll also still hear wet read. It dates to the film era, when a radiologist gave an impression while the X-ray film was literally still wet from the developer. In current usage it means a rapid preliminary, most often in the ED. The chart still needs the signed report afterward.

How often do final reads change preliminary findings?

Major discrepancies between preliminary and final interpretations run in the range of 2 to 3 percent across the peer-reviewed literature, with 2.6 percent the figure quoted most often. Roughly 1 in 40 preliminary reports gets revised in a way that can change management.

The percentage is the part everyone quotes, and it is the less interesting half of the finding. What costs you is the interval. In a two step model the clock starts when the preliminary posts and stops when the second signature lands, and every clinical decision taken in between rests on a document that may not survive the morning. An ED reading a hundred overnight studies will see two or three of them change in a way that matters. By the time the amended report exists, one of those patients has been discharged and another is already on a treatment path chosen from the earlier impression. Somebody now has to find them. A nurse works the phone list, the ED attending reopens a chart that closed six hours ago, and the day radiologist writes an addendum for a change nobody was in the room for. None of that surfaces as a discrepancy rate. It surfaces as your morning, and it scales with the width of the interval, which is why the only lever that reliably shrinks it is making the first interpretation the signed one.

Read the literature with two caveats in hand. Much of it studies resident preliminaries finalized by attendings, and experienced teleradiologists issuing preliminaries post lower rates than residents do. The structural fact survives either way: any two step interpretation opens a window during which care proceeds on a report that can change.

The second caveat is age. Most of the widely circulated public content on this question dates to the early 2010s, when overnight preliminaries finalized by the day group were how coverage worked. That has shifted. Signed final reports overnight are now the standard among teleradiology providers that compete on quality, because that is the one change that closes the window instead of managing it.

For comparison, AstraRad signs the first interpretation and independently double reads 1 in 20 signed reports, blind to the original. Major discrepancies run under 0.3% of signed reports over the trailing twelve months, and each one is reviewed and closed with the radiologist who signed it. The counting method behind that figure is published on our SLA and quality page.

Payment and liability both attach to the signature

Both the professional fee and the legal exposure for an imaging study follow the physician who signs the final report, not the author of the first impression.

Billing. Medicare rules and most commercial payer policies pay the professional component to the physician who renders the complete, signed interpretation. A preliminary is generally not separately billable on its own. So in a preliminary plus overread model you are buying two radiologist touches on every study and generating one billable event, and the overnight fee carries no revenue against it. CMS states the interpretation rules in the Medicare Claims Processing Manual, Chapter 13, which pays for only one interpretation of an emergency department study, the one that directly contributed to the diagnosis and treatment of the patient, and treats a review of findings without a complete written report as not separately payable.

Liability. The final signed report is the legal document of record. When a preliminary and a final disagree, the disagreement itself becomes discoverable, and the interval between them is the period in which the patient was managed on information the record later contradicted. Closing that interval to zero costs nothing at the point of care.

Accountability. One study, one accountable physician. Split the interpretation across two readers and you split peer review, discrepancy tracking, and the credentialing file that has to account for both of them. Ask which physician's name a plaintiff's attorney finds on the overnight study, and how fast your vendor can produce that physician's fellowship training and state licensure.

When do preliminary reads still make sense?

Preliminary reads still make sense in residency training, in groups that deliberately finalize in house, and as a verbal impression given minutes ahead of a fast signature.

Academic training programs. Resident preliminaries with attending finalization are how radiologists learn to read. The discrepancy cycle is the point, backed by structured overread and feedback that closes the loop with the trainee.

Groups that finalize in house on purpose. A group that wants its own radiologists on every signature, for continuity or credentialing reasons, may buy preliminary overnight coverage deliberately and finalize each morning as policy. In these first two settings a signed final overnight service, AstraRad included, is the wrong fit.

A verbal impression ahead of a fast signature. In hyperacute settings a phone call to the ED within minutes, followed by the signed report shortly after, gives the treating team speed while the record closes behind it. This is a sequencing choice inside a final read model.

Outside those three, a preliminary plus overread pipeline doubles the work and pays once for it: a finalization queue that eats your day team's first hours and a standing 2 to 3 percent revision rate somebody has to administer. Imaging centers, hospitals without a residency program, and groups buying nighthawk coverage mostly land in the same place, where the first interpretation is the signed one and the morning queue is empty.

Every AstraRad tier ends in a signed final report

AstraRad signs the first interpretation at every priority tier, at every hour: STAT in under 1 hour, urgent in under 4 hours, routine in under 24 hours, each delivered as a signed final report. The clock behind each tier runs from last-image arrival to radiologist signature. There is no preliminary layer to reconcile and no morning overread queue.

The panel is board-certified and fellowship-trained by subspecialty, every radiologist is physically located in the United States, and each one is licensed in the state where your patients are located. A 2 a.m. stroke protocol CT goes to a neuroradiologist who is on shift and awake, and it comes back as a STAT CT final read against the same clock as a 2 p.m. study. What a sub hour STAT tier demands of a roster is set out in how a STAT read gets signed that fast. Coverage spans CR X-ray, ultrasound, mammography, CT including coronary CTA, MRI including cardiac, nuclear medicine, and PET-CT, each routed to the matching section in neuro, MSK, body, cardiac, breast, pediatric, chest, ER, oncologic, or nuclear imaging.

Quality is enforced identically at every tier: 1 in 20 signed reports goes to a second subspecialist who cannot see the first interpretation. Studies arrive by DICOM from your PACS or through portal upload, with no integration project to schedule, and results come back over HL7 or FHIR. Your PACS stays your PACS. The platform is DICOM conformant and HIPAA aligned, under a BAA signed before the first study moves.

Across STAT, preliminary, and final reads, one question decides the rest: which document lands in the chart, and when. A report that can still change is a report somebody still has to chase; a signed final read closes the chart the first time.

Reports are billed under per-report pricing: no minimums, no subscriptions, no platform fees, with priority as a printed multiplier, so the cost of a STAT final report against a routine one is visible before your first study goes out. The contact page is the fastest way to get those multipliers in writing. If you are still mapping the coverage models against each other, the plain language explainer on what teleradiology is and how it works covers the mechanics end to end, and the rest of the definitions live in the teleradiology resource library.