What "Pre-Reads" Actually Means in Radiology (And Why So Many Searches Get It Wrong)

Pre-reads are searched 1,600 times a month and rarely defined precisely. Here's what a pre-read is, how it differs from a final read, and why the layer exists.

⚠️ Important scope note — please read before continuing

RAD365 provides offshore preliminary reads for human patients in the USA only. RAD365 does not provide final reads — the facility's own licensed radiologist completes every final interpretation. For veterinary and pet imaging there are no preliminary reads; veterinary studies are read as complete final reports by specialist veterinary radiology groups.

By Trisha Seal — August 28, 2026. Trisha writes on RAD365's 24/7 preliminary-read workflow and PACS integration operations, where preliminary interpretations are delivered inside a facility's existing worklist for human patients in the USA.

Roughly 1,600 people a month type pre-reads into a search engine. Almost none of them find a precise definition. They find provider pages that assume the reader already knows, job postings that use the phrase in passing, and forum threads where three clinicians use it to mean three subtly different things. For a term that sits at the center of how emergency and overnight imaging actually gets handled in US hospitals, that is a strange gap — and it has consequences well beyond vocabulary.

The confusion is not academic. Departments evaluating outside coverage routinely compare quotes for deliverables that are not the same deliverable. Administrators assume a pre-read is billable when it is not. Clinicians occasionally treat a pre-read as though it were the diagnosis of record, which it never is. This piece clears up what the term covers, where the boundary sits between a pre-read and a final read, and why an entire service category grew up around a step that produces nothing billable.

The short answer, before the long one

A pre-read is a rapid, documented, non-final interpretation of an imaging study, issued to support a clinical decision that cannot wait for the signed report. It is produced inside the facility's existing worklist, it is used immediately, and it is superseded by a final interpretation completed and signed by the facility's own licensed radiologist. That final report — not the pre-read — is what enters the patient record and what gets billed. Everything else about the category follows from that one distinction. RAD365's own preliminary reads service line is built entirely on this boundary for human patients in the USA.

Pre-read vs prelim vs wet read vs final read

Three of these four terms mean the same thing. The fourth is categorically different, and conflating it with the others is the single most common mistake in the category.

Term	What it is	In the medical record?	Billable?
Pre-read	Rapid non-final interpretation supporting an immediate decision	No — workflow documentation	No
Prelim	Identical to a pre-read; the formal shorthand	No — workflow documentation	No
Wet read	Identical again; a film-era holdover term still in daily use	No — workflow documentation	No
Final read	Complete interpretation signed by the facility's own licensed radiologist	Yes — diagnosis of record	Yes

The "wet read" label survives because it describes something physical that no longer happens. Film emerged from the processor still wet, and a radiologist would look at it on the way past and give the ordering physician an immediate impression before it dried and was filed properly. The film is long gone; the workflow step it named is not. Departments that switched to digital human radiology workflows two decades ago kept the word, which is part of why the terminology drifted in the first place.

Why the pre-read category exists at all

A step that generates no billable revenue does not become a national service category by accident. It exists because reading capacity has not kept pace with demand, and because the shortfall is concentrated in exactly the hours when studies cannot wait.

37,482 → 47,119–52,591

US radiologist workforce, 2023 to projected 2025 (JACR)

3.0% vs 1.9%

Post-COVID annual attrition versus pre-pandemic (JACR)

~3,116

Radiologists' worth of growth projected to be erased by 2055 by elevated exits alone (JACR)

Those figures come from a Journal of the American College of Radiology (JACR) workforce study, and the shape of them is more interesting than the headline. The workforce is growing. It is also leaking, at an attrition rate that rose from roughly 1.9% a year before the pandemic to roughly 3.0% after it — and the JACR researchers project that the elevated exit rate on its own could erase the equivalent of about 3,116 radiologists' worth of growth by 2055. Recruitment is winning the count and losing the compounding.

The second force is structural. According to a 2026 ACR Bulletin workforce update citing Harvey L. Neiman Health Policy Institute research, the average US radiology practice size nearly doubled between 2014 and 2023 — from 45 to 89 physicians — as radiologists per practice grew 85%, from 9.7 to 17.9. That is consolidation, driven partly by staffing pressure, and for a single community hospital it changes the hiring maths entirely: the candidate you are trying to recruit is being recruited by a group four times your size.

The same ACR/Neiman HPI analysis adds a detail that matters for anyone planning subspecialty coverage: subspecialty radiologists are 37% more likely to leave the workforce than generalists, and radiologists overall are now exiting practice at more than twice the rate they were less than a decade ago. Depth of coverage, in other words, is the part of the rota most exposed to attrition.

Put together, these numbers explain the category without any appeal to marketing. Hospitals need reading capacity in defined windows, cannot reliably hire it, and can buy a preliminary layer that is contractible this quarter. The preliminary reads layer is a response to a supply problem, not a preference.

How the pre-read workflow actually works

The study is acquired and filed. The technologist completes the exam and it lands in the facility's PACS exactly as it always has. Nothing about acquisition changes.
Routing rules select it. Predefined rules — by modality, priority, time of day, or ordering location — determine whether the study goes to the preliminary rota. The rules live in the facility's own environment; there is no manual push step for technologists.
A radiologist reads it in the facility's PACS. The preliminary reader works inside the existing system rather than a separate portal, using the same priors and the same worklist context the in-house team sees.
The preliminary impression returns to the worklist. It appears where the ordering clinician already looks. No second system to check, no email attachment to chase.
Critical findings escalate immediately. Anything meeting the facility's critical-finding criteria triggers the written escalation path — named contact, defined channel, defined time window — with the communication documented in the audit trail.
The facility's radiologist completes the final. The signed final interpretation supersedes the preliminary impression and becomes the diagnosis of record and the billable event.
Discrepancies are tracked. Pre-read and final are compared over time. That comparison is the quality instrument for the whole layer, and it feeds naturally into a facility's broader peer review and QA program.

Step seven is where most of the value in a mature arrangement sits, and it is the step most often skipped. A pre-read layer without discrepancy tracking is a service you are buying on faith. With it, you have a monthly number that tells you whether the layer is safe, and a shared basis for conversation when it is not.

What the term should mean when you're the one buying

If you are evaluating providers, treat the word "pre-read" as a placeholder until the provider fills it in. Ask what precisely is delivered: a triage flag, a structured impression, or a full narrative preliminary. Ask whether the reader works in your PACS or theirs. Ask for turnaround committed by modality and priority, reported for your site rather than blended across a client base. And ask how discrepancy against your finals is measured and shared.

Four questions, all of them answerable in a first call. They also happen to separate the providers with an operational model from the providers with a marketing page — which, given how loosely the term is used across the category, is the harder distinction to make from a website alone.

Pre-reads in radiology: frequently asked questions

Understanding the Term

What does "pre-reads" mean in radiology?

A pre-read (short for preliminary read) is a rapid, non-final interpretation of a study issued to support an immediate clinical decision — most often overnight, during overflow, or while clearing a backlog. It is a workflow step, not the diagnosis of record. The pre-read tells the treating clinician what appears to be there right now, so a disposition decision can be made without waiting; the facility's own licensed radiologist still completes and signs the final interpretation afterwards.

Is "pre-reads" the same thing as a "prelim" or a "wet read"?

Yes. Pre-read, prelim, and wet read all describe the same non-final interpretation step. The terms come from different eras — "wet read" is a holdover from when film came wet out of the processor and a radiologist glanced at it before it had dried — but they function identically inside a modern PACS-based workflow. Whichever word a department uses, the deliverable is a documented preliminary impression that precedes, and does not replace, the signed final report.

Why do so many people search "pre-reads" but seem confused about what it actually covers?

Because the term is used loosely across marketing material, hospital job postings, and vendor documentation without a consistent definition, and it is easy to conflate a pre-read with the signed final report that actually goes into the patient's chart. One provider's page describes a pre-read as a triage flag; another describes something close to a full report. Buyers researching the category end up comparing deliverables that are not the same deliverable, which is why the definitional question keeps getting asked.

Does a pre-read ever become part of the official medical record?

No. The pre-read supports the clinical decision in the moment. The final, signed interpretation from the facility's own licensed radiologist is what becomes part of the official patient record and what gets billed. The pre-read is retained as workflow documentation and is used for quality comparison against the final, but it is not the diagnosis of record and should never be presented to a patient or a payer as one.

Who typically performs pre-reads?

Radiologists working in a dedicated preliminary-interpretation capacity — either an on-shift overnight physician internally, or, increasingly, an outside preliminary-read team working inside the facility's existing PACS. The distinction that matters operationally is not who employs the reader but whether the reading capacity is continuously staffed. A rota with unassigned hours produces delays regardless of whether the reader sits down the corridor or in another time zone.

Why the Pre-Read Category Exists

Why did outsourced pre-read services become common in US radiology?

Largely because supply growth has been undercut by attrition. A Journal of the American College of Radiology (JACR) workforce study found the US radiologist workforce grew from 37,482 in 2023 to a projected 47,119–52,591 by 2025 depending on the residency-growth scenario — but with post-COVID attrition running at roughly 3.0% a year versus 1.9% before the pandemic, researchers project that elevated exit rate alone could erase the equivalent of about 3,116 radiologists' worth of growth by 2055. Headcount rising while net capacity leaks is exactly the condition under which a purchasable preliminary-interpretation layer becomes a standing part of the market rather than an emergency measure.

What's driving hospitals to add a pre-read layer instead of just hiring more radiologists?

Consolidation has already absorbed much of the available hiring headroom. According to a 2026 ACR Bulletin workforce update citing Harvey L. Neiman Health Policy Institute research, the average US radiology practice size nearly doubled between 2014 and 2023 — from 45 to 89 physicians — as radiologists per practice grew 85% (9.7 to 17.9), a sign of consolidation driven partly by staffing pressure. For an individual hospital, that means competing for a recruit against far larger groups. Buying a defined coverage layer is a decision a department can execute this quarter; a successful recruitment round often is not.

Does adding a pre-read layer actually reduce the workload on in-house radiologists?

Yes — it absorbs routine volume, backlog, and off-hours studies so the facility's own radiologists can focus on signing finals and handling complex cases instead of triaging every incoming study personally. The relief is not only in study count. It removes the interruption load of being paged about studies that turn out to be unremarkable, which is a meaningful share of the fatigue that drives the attrition the workforce data describes.

How the Workflow Actually Works

Does adding pre-reads change how a study moves through PACS?

No — a pre-read layer is inserted into the existing PACS worklist. The study still lands, gets read, and returns to the same system technologists already use. There is no separate portal for staff to learn and no second place to check for results. If a prospective arrangement requires technologists to push studies somewhere else or clinicians to log into an outside viewer, that is an integration shortcut, not a pre-read workflow, and it will erode the time saving it was bought to create.

How fast is a typical pre-read turned around?

Turnaround targets and STAT thresholds are set per facility based on case mix and protocol, but the point of a pre-read layer is that it is materially faster than waiting for the next on-shift radiologist to become available. What to hold a provider to is not an advertised average but a committed target by modality and priority, reported per facility rather than blended across a client base, so an outlier site cannot hide inside a healthy overall number.

What happens if a pre-read flags a critical finding?

It is relayed immediately to the on-shift provider per the facility's written escalation protocol, with full audit documentation, ahead of the final signed report. The escalation path — who is called, on which number, within how many minutes, and what happens if that person does not answer — should be agreed in writing before the first study is read, not improvised the first time a critical finding appears at 3am.

Can a pre-read layer cover routine daytime volume, not just nights?

Yes — pre-read layers are increasingly used for daytime overflow and backlog clearance, not only the traditional overnight window. A department carrying a persistent unread backlog, or absorbing a seasonal volume spike, faces the same structural problem as an uncovered night: studies waiting on capacity that does not exist. The staffing model is identical; only the hours differ.

Cost, Compliance & Choosing a Partner

Is a pre-read billable to Medicare or insurance?

No. Only the final, signed interpretation is billable. The pre-read is a workflow and support layer that precedes it. That is why a pre-read layer should be evaluated as an operating cost against turnaround, throughput and clinician time recovered — not modelled as a new revenue line, which it is not.

What should a facility check before adding a pre-read service?

Turnaround commitments by modality, how critical findings are escalated, whether the provider works inside the facility's existing PACS without new software, and how discrepancies between pre-read and final are tracked over time. Ask for the discrepancy reporting format specifically. A provider that already produces that report routinely is measuring its own quality; a provider that offers to build one on request is not yet doing so.

Are pre-reads used the same way in veterinary imaging?

No. Veterinary imaging has no preliminary stage at all. Veterinary studies are read as complete final reports by specialist veterinary radiology groups, so there is no pre-read step to buy, schedule or escalate. Anyone mapping a human-side preliminary workflow onto a veterinary practice is describing a category that does not exist there; the veterinary equivalent is simply a final report, delivered once.

Does adding a pre-read layer require replacing the facility's PACS system?

No — a properly built pre-read layer reads inside the facility's existing PACS. No new software or hardware purchase is required. Connectivity is established through the existing DICOM and HL7 interfaces, with routing rules that determine which studies reach the preliminary rota, and the finished preliminary impression returns to the same worklist the department already works from.

How does a facility measure whether its pre-read layer is actually working?

Track the discrepancy rate between pre-read and final interpretations, turnaround time by modality and priority, and whether STAT findings are consistently relayed within the facility's written protocol window. Reviewed monthly, those three measures show whether the layer is genuinely reducing time-to-decision or simply adding a step. A discrepancy rate that is never reported is not the same as a discrepancy rate that is low.