STAT radiology reads: 30-min median, 99.4% SLA (2026 data)

AstraRad's measured STAT median is 30 minutes, from the arrival of the last image of the study to the radiologist signing the final report, against a contractual STAT tier of under 1 hour. A STAT radiology read is a study flagged for immediate interpretation because the finding may change what the treating team does within the hour, so the tier is a promise about a clock: the report arrives in the middle of a clinical decision. Urgent studies are contracted under 4 hours and routine under 24, and 99.4% of reports across all three tiers were signed inside their window over the trailing 12 months.

Whether any of those figures mean anything depends on which event starts the clock and which event stops it. Most published turnaround numbers define neither.

STAT read turnaround benchmarks for 2026

Typical contractual tiers across the US teleradiology market run 30 to 60 minutes for STAT, 2 to 4 hours for urgent, and 24 hours for routine, and those tiers are the right yardstick for any vendor's claims:

Read that first column as what vendors agree to, not as what departments experience. The two are not the same number and should never be quoted as if they were. Observed turnaround runs materially longer: AAG Health's 2025 benchmark review puts emergency CT and MRI at 60 to 120 minutes and outpatient CT and MRI at 24 to 72 hours. A contract tier tighter than the observed benchmark is the entire reason to buy one, and it is also why the contract tier alone tells you nothing until a median is published beside it.

The contractual column is the one that misleads buyers. A contractual ceiling is the worst case a vendor has agreed to accept a penalty for, and a median is what the middle study experienced. Those are different questions. A panel contracted at 60 minutes may sign the middle STAT study in 25 minutes, or it may brush 58 minutes every night and remain in compliance the whole time. Both pass the same contract review. Only one of them returns your trauma CT while the surgeon is still deciding.

That gap appears nowhere in an RFP that asks only for the guarantee, which is why the guarantee is the easiest figure for a vendor to hand over and the least useful one to receive. Ask for the median alongside it, and ask for the median by tier. A blended figure across routine outpatient volume says very little about the hour your emergency department is asking about. A vendor that measures its own performance answers both questions in a sentence, and a long pause at the second one is itself a finding.

Marketing averages and contractual guarantees also disagree with some regularity, occasionally on the same vendor's own website: a homepage average sitting two clicks from a services page that quotes a slower contractual range. When two pages under one brand carry two different numbers, the number nobody reconciled is the number nobody audited.

The clinical case for the fast end of those tiers has been measured. A propensity-matched cohort of emergency patients with acute abdominal pain, published in the American Journal of Emergency Medicine, found that CT report availability within 30 minutes of physician assessment was associated with shorter emergency department and inpatient length of stay among patients triaged at higher severity. That interval covers ordering, transport and scanning as well as reporting, so it is wider than a report turnaround clock and is not the same 30 minutes this page is about; what carries over is the direction the tiers are built around.

Stroke is the exception to all of this. Facilities running an acute stroke program typically negotiate a dedicated protocol with direct radiologist notification, and a general STAT tier, AstraRad's included, is the wrong instrument for that job. Benchmark stroke coverage separately from ordinary STAT volume.

What an AstraRad STAT read includes

Every tier ends in a signed final report, never a preliminary, including overnight. The ACR Practice Parameter for Radiologist Coverage of Imaging Performed in Hospital Emergency Departments recognizes two models for off-site coverage, a preliminary interpretation later finalized by the local radiologist or a final interpretation as hospital bylaws allow, and only the second one is on offer here. Your day team inherits an empty finalization queue in the morning, because the physician who read the study at 3 a.m. already signed it. The mechanics of that distinction, and the discrepancy evidence behind it, are laid out in STAT vs preliminary vs final reads.

The tiers are identical for every client and printed on the rate card. Pricing is per report, with no minimums, no subscriptions, and no platform fees, and STAT priority carries a multiplier printed on the same card, so the cost of escalating a study is knowable before you send it. The structure is on our pricing page.

The SLA page is the canonical dated source for every figure on this page. If a number here ever disagrees with that page, the SLA page wins. The same figures, each carrying its as-of date and its own citable anchor, are collected on the facts page for anyone quoting them in a shortlist or an RFP response.

Every turnaround number hides two definitions

The clock starts at last-image arrival. Turnaround is measured from the moment the last image of the study lands on our platform, whether it arrives as DICOM from your PACS or through portal upload. Transmission time from your facility to ours belongs to the network between us, and no reading provider can honestly claim to govern it. Stating the boundary out loud is what makes it auditable: every study carries receipt, assignment, and signature timestamps, so you can lay our start event against your own send logs.

The clock stops at the signature on the final report, not at a preliminary impression. One physician, one document, one timestamp.

We publish the median, not the mean. A mean bends under a large volume of trivial studies and bends further once a vendor starts excluding outliers, and the exclusions never appear in the marketing copy. The median describes the middle STAT study: half sign faster, half take longer, and the 1-hour contractual ceiling bounds the tail. Compliance is then reported as its own figure, the percentage of all reports that met their tier, which is where the trailing 12-month figure comes from.

Put three things to any provider you evaluate, in writing: the start event, the stop event, and whether the number is a median or a mean. The answers are more diagnostic than the number itself. Our full evaluation checklist is in how to choose a teleradiology company.

Who gets called when a scan shows a critical finding?

When a reading radiologist identifies a critical or unexpected urgent finding, it is flagged at the point of interpretation and the referring clinician or your designated contact is notified directly, before sign-off, along the escalation path defined during onboarding. A fast signature closes the record. It does not close the loop on a finding that needs somebody paged. The communication is documented inside the report: who was notified, when, and by what channel. The signed final report then follows within the study's SLA tier.

This matters for accreditation and for malpractice exposure alike. The ACR Practice Parameter for Communication of Diagnostic Imaging Findings places findings that suggest a need for immediate or urgent intervention, naming pneumothorax, pneumoperitoneum and a significantly misplaced line or tube among the examples, in the tier that calls for nonroutine communication instead of an ordinary report delivery. A reviewer works from the report; if the call is not in the report, the reviewer has no evidence that the call happened, and a year later neither does the radiologist who made it. Because receipt, assignment, and signature timestamps are already on every study, critical results documentation falls out of the normal workflow instead of requiring a separate compliance exercise.

Quality control runs on the same volume as speed. A standing quality assurance program with formal peer review is what the ACR White Paper on Teleradiology Practice asks of a teleradiology provider, covering physician education, error reduction, longitudinal follow-up, a second opinion when local caregivers raise a concern, and remediation. One in every 20 signed reports goes to a second subspecialist who cannot see the first interpretation, and major discrepancies run under 0.3% of signed reports; every one is reviewed at the monthly discrepancy meeting and closed with the reader who signed it. Both figures are published together, because a fast median only means something next to the rate at which signed reports are later found to be wrong.

Five ways a turnaround claim fails inspection

"Industry-leading turnaround times" is the most common phrase on teleradiology websites and the least informative. Some established providers publish a Turnaround Times section with no numbers in it at all. Read that as a disclosure: a company that measures its turnaround publishes the result, because the figure sells.

The claims worth disqualifying:

• No number at all. "Fast," "rapid," and "industry-leading" commit a vendor to nothing.
• A number with no definition. 20 minutes from what to what? Median or mean? STAT only, or blended across all studies?
• Numbers that conflict across the vendor's own pages. Two unreconciled figures under one brand mean no audit trail underneath either.
• A marketing average with no contractual tier behind it. The SLA in the contract is the number you can enforce.
• No trailing compliance figure. A median with no compliance percentage beside it hides the tail, and the tail is where your 2 a.m. trauma case lives.

Every item on that list is answered on the SLA page, which carries the counting method, the measurement window, and the date each figure was refreshed.

Sub-hour STAT capacity shows up in three places

Sub-hour STAT radiology reads earn their keep in emergency coverage, overnight coverage, and overflow:

• ED and trauma coverage, where STAT CT reads carry most of the volume: head CT, CT angiography, abdomen and pelvis.
• Overnight coverage, replacing or supplementing an on-call rotation with subspecialist final reads, the arrangement covered in nighthawk radiology.
• Overflow and surge, where your own radiologists keep the routine list and STAT-eligible studies route to us only when your queue backs up.

The first two are hospital problems before they are radiology problems, and they are usually bought that way. What an ED throughput gap or an uncovered night costs a hospital, and how a hospital medical staff office credentials a remote panel to close it, is set out on teleradiology for hospitals.

A sub-30-minute median is a rostering fact before it is a technology fact. Someone has to be awake at the hour your scanner finishes, licensed in the state where your patient is located, and trained in the anatomy on the screen. Meeting all three conditions at once, every hour of the year, is what 240 board-certified fellowship-trained subspecialists buy you. All of them are physically located in the United States, they work scheduled day, evening and overnight shifts, and every study routes to the matching subspecialty section, so a pediatric abdominal CT at 2 a.m. reaches a pediatric radiologist who is on shift and awake. Volume across the panel runs roughly 600,000 reads per year, and current headroom of 25,000 additional studies per month is why a STAT study sent on a holiday night meets the same tier as one sent on a Tuesday morning.

Starting does not require an integration project. Send DICOM from your PACS or upload through the portal, encrypted end to end, HIPAA aligned under a signed BAA, with HL7 or FHIR report delivery when you want it.

One last thing about the arithmetic. Our clock starts at last-image arrival, and your patient's clock started earlier, on the table. The minutes between those two events belong to your network and your scanner protocol, and any provider who folds them into a marketing average is quoting a number they did not measure. Ask for the multipliers and the tier definitions in writing before your first STAT study goes out. The fastest route to both is to book a coverage consultation. STAT is the sharpest of the teleradiology use cases we publish numbers for; the rest of the coverage picture starts there.