Why Preliminary Reads Exist: The Radiologist Shortage in Data
Radiologist supply, imaging demand and attrition data explain why outsourced preliminary reads exist as a category. Real 2026 figures, cited inline.
Preliminary reads did not become a purchasable service category because hospitals wanted another contract to manage. They became one because the arithmetic of who is available to read a study at 3 AM stopped working. The data below — radiologist supply projections, exam-volume growth, attrition rates and the cost of a single additional salary line — is the whole explanation. Everything else about how outsourced preliminary reads operate follows from it.
⚠️ Important scope note — please read before continuing
RAD365 provides offshore preliminary reads for human patients in the USA only — we do not provide final reads; your 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 — September 1, 2026. RAD365 has run offshore preliminary read operations for 12+ years across 20+ clients, delivering non-final interpretations inside each facility's existing PACS worklist. Final interpretations are always completed and signed by the facility's own licensed radiologist.
Projected US radiologist supply growth 2023–2055 under static residency levels (Neiman HPI)
Projected growth in imaging demand by 2055 (Neiman HPI)
Exam volume growth vs. radiologist workforce growth, 2018 to early 2024 (Radiology Business)
Rate at which radiologists now leave practice vs. under 10 years ago (ACR Bulletin / Neiman HPI 2026)
Why Outsourced Preliminary Reads Exist as a Category
Start with supply. The Harvey L. Neiman Health Policy Institute projects that the US radiologist supply will grow roughly 25.7% between 2023 and 2055 if residency positions stay static, while imaging demand over the same horizon is projected to rise 16.9% to 26.9%. Read carelessly, that looks like a wash. Read properly, it is the problem: supply and demand growth are roughly comparable, which means the gap that exists today does not close on its own. Absent workforce expansion or a reduction in per-person imaging utilisation, a department that is short-staffed for nights in 2026 is structurally short-staffed for nights in 2040.
The nearer-term picture is tighter than the long-range model. Workforce and volume analysis reported by Radiology Business found radiology exam volume rose about 31% — roughly 5% a year — between 2018 and early 2024, while the radiologist workforce grew only about 24% over the same period. Seven percentage points of divergence across six years does not sound dramatic in the abstract. In a department it looks like a worklist that is never empty at shift change.
Then attrition. The 2026 ACR Bulletin workforce update, citing Neiman HPI research, found radiologists are now leaving practice entirely at more than twice the rate they were less than ten years ago, and that subspecialty-trained radiologists are about 37% more likely to exit than general radiologists. The subspecialty figure matters most for coverage planning, because subspecialty depth is exactly what rotas assume will be reachable when a complex study lands unexpectedly. These are attrition and composition figures, not a causal claim about any single hospital — but they describe a bench that is thinner and less predictable than the one most staffing models were built around.
Finally, cost. Salary.com puts the average US radiologist salary at $531,890 per year, with a typical range of roughly $472,000 to $637,000 — before recruitment, credentialing, benefits and the on-call premium. That is the number a CFO is actually being asked to approve when the proposed fix for overnight coverage is "hire another radiologist." For a facility whose off-hours volume does not justify a full clinical FTE, a per-study preliminary layer is the alternative that makes the coverage arithmetic solvable rather than merely expensive. It sits alongside, not instead of, the facility's existing human radiology operation.
What Counts as a Preliminary Radiology Interpretation
Preliminary radiology interpretations are rapid, non-final reads issued so a clinical decision can be made immediately — admit, transfer, operate, discharge. The deliverable is a documented, time-stamped impression of what appears to be there right now. It is not the diagnosis of record, it is not signed as the final, and it is not billed as the professional component.
The terminology drifts by department. "Prelim," "pre-read" and "wet read" all name the same step; today's post uses preliminary reads because that is the operational term that appears in coverage contracts. What matters is the boundary rather than the vocabulary: the preliminary interpretation supports the decision that cannot wait, and the facility's own licensed radiologist afterwards completes and signs the final interpretation with priors and the full clinical picture in hand. Programs that run a preliminary layer well pair it with structured peer review and QA, sampling preliminary-versus-final pairs so the discrepancy rate is a measured number rather than an assumption.
In-house off-hours staffing vs. a per-study preliminary model
| Attribute | Additional in-house radiologist for off-hours | Per-study preliminary read layer |
|---|---|---|
| Cost structure | Fixed salary line (~$531,890 average, per Salary.com) plus benefits and on-call premium | Variable, scaled to studies actually read |
| Coverage on absence | Single point of failure; locum cover at premium rates | Team-based; coverage does not depend on one individual |
| Recruitment exposure | Full exposure to a market with rising attrition | None — capacity is contracted, not recruited |
| Ramp time | Months of search, credentialing and onboarding | Phased by shift, typically weeks |
| Who signs the final | Facility's own licensed radiologist | Facility's own licensed radiologist |
| Best fit | Sustained, predictable off-hours volume justifying a clinical FTE | Nights, weekends, holidays, overflow and backlog |
What the process actually looks like
- Studies land where they always did. Acquisition and routing are unchanged — the study appears on the facility's existing PACS worklist through the existing DICOM pathway. No new viewer, no parallel archive, no change to how technologists submit.
- Eligible studies route to the preliminary worklist. Typically after-hours, overflow and backlog volume, scoped by the facility.
- The preliminary interpretation is issued. Rapid, documented, time-stamped, and explicitly non-final.
- Critical findings are relayed per written protocol. The facility's own escalation policy governs who is called, how fast, and how the communication is logged for audit.
- The facility's radiologist signs the final. Independently, with priors and full clinical context. That signed report is the diagnosis of record.
None of those steps require the facility to change PACS, buy software, or restructure its radiology group. That is deliberate: the coverage problem the shortage data describes is a staffing problem, and adding an IT project to a staffing problem does not solve either one. The preliminary reads coverage model is built around that constraint.
Scoping overnight or overflow coverage?
RAD365 has run offshore preliminary read operations for 12+ years across 20+ clients — inside the facility's existing PACS, with the facility's own radiologist signing every final interpretation.
Talk to our preliminary reads team →Preliminary Reads and the Radiologist Shortage: Frequently Asked Questions
The Shortage Behind the Demand
Why is there a radiologist shortage in the US?
Practice consolidation, an aging population driving imaging volume, and radiologists leaving clinical practice at more than twice the rate they were less than a decade ago have combined to outpace new radiologists entering the field, per the ACR's Harvey L. Neiman Health Policy Institute.
Is the radiologist shortage expected to improve by the 2030s?
Not on its own — Neiman HPI research projects US radiologist supply will grow about 25.7% between 2023 and 2055 under static residency levels, while imaging demand is projected to climb 16.9% to 26.9% over the same period.
Which radiology subspecialties are hit hardest by the shortage?
Subspecialty-trained radiologists are about 37% more likely to leave the workforce than general radiologists, per ACR Bulletin workforce data, and pediatric radiology's share of all practicing radiologists fell from 6.4% to 4.6% between 2016 and 2023.
Is imaging volume actually growing faster than the radiologist workforce?
Recent workforce analysis found radiology exam volume rose about 31% (roughly 5% a year) between 2018 and early 2024, while the radiologist workforce grew only about 24% over the same stretch.
Why Outsourced Preliminary Reads Exist as a Category
What does it actually cost to add another radiologist just to cover nights and weekends?
Salary.com puts average US radiologist pay at roughly $531,890/year (typical range ~$472,000–$637,000) — before recruiting, credentialing and benefits — the core economics behind an offshore preliminary layer instead of another full-time salary line.
Does an offshore preliminary read layer replace the need for in-house radiologists?
No — it exists because facilities already have licensed radiologists completing final interpretations; the preliminary layer absorbs routine volume, backlog and off-hours reads so those radiologists aren't stretched across every shift.
Could AI eventually close the radiologist shortage instead of outsourcing?
Published research frames AI-assisted triage as a workflow aid that helps radiologists prioritize studies, not a near-term replacement for the interpretation capacity the shortage has removed.
Does outsourcing the preliminary layer help reduce burnout-driven attrition?
Facilities that shift routine and off-hours volume to a preliminary read partner report in-house radiologists spend more time on final interpretations and complex cases instead of triaging every incoming study — directly tied to the burnout pressures workforce researchers have documented.
How the Preliminary Layer Works Day to Day
Does adding preliminary read coverage change how technologists submit studies?
No — studies land on the facility's PACS worklist exactly as they always have; the preliminary layer works inside the existing system.
How are critical findings escalated overnight?
Critical findings are relayed to the on-shift provider per the facility's own written escalation protocol, with the read, relay time and communication method documented for audit purposes.
Can a facility start with just one or two shifts before expanding coverage?
Yes — most facilities phase in coverage for their hardest-to-staff shifts first (nights, weekends, holidays) before expanding to daytime overflow.
What counts as a preliminary radiology interpretation versus a full report?
A preliminary radiology interpretation is a rapid read supporting an immediate clinical decision; the complete, signed report is completed separately by the facility's own licensed radiologist.
Evaluating a Preliminary Read Partner
What should a radiology director check before trusting an outsourced preliminary read vendor?
Its escalation protocol and audit-trail practices, how it integrates with the facility's specific PACS without new software, and how discrepancies between preliminary and final reads are tracked over time.
How can a facility tell if its current overflow coverage is already failing?
Warning signs include STAT studies waiting behind routine ones, referring physicians chasing reports, and repeated reliance on expensive short-term locum coverage for nights and weekends.
What data should a facility track to know a preliminary reads partnership is working?
Turnaround time by shift and modality, discrepancy rates between preliminary and final reads, and whether in-house radiologist overtime and locum spend decline after onboarding.
How many additional radiologists would it take to close the shortage without any outsourcing?
Neiman HPI's modeling shows supply and demand growth are roughly comparable through 2055 even under current trends — the shortage does not close through workforce growth alone.
Is demand for preliminary reads limited to hospitals, or does it affect smaller imaging centers too?
Independent imaging centers and radiology groups face the same shortage economics as hospitals, often with less ability to absorb a single radiologist's absence.