78-day tracking window — what 20,775 job postings reveal about America's radiologist shortage.
About 47% of "open jobs" on aggregators are re-posts of the same role. The real shortage is concentrated and stubborn — 1,470 currently-active openings have been sitting unfilled for over two months. Most aren't underpaid — they're geographically inconvenient.
Of 12,187 closed postings with complete lifespans, the distribution is sharply bimodal: a quarter vanish in under a week, but a third grind for 30–60 days. There is no "average" radiologist opening.
| Lifespan bucket | Share of closed jobs |
|---|---|
| < 7 days | 24.9% |
| 7–14 days | 11.6% |
| 15–30 days | 33.4% |
| 31–60 days | 27.8% |
| 61–90 days | 2.3% |
1 in 4 radiology jobs vanishes in under a week — filled fast, pulled, or revealed as a duplicate. But 30%+ stay open longer than a month. The market doesn't have a single "fill rate" — it has two markets layered on top of each other.
Half of currently-active jobs have been open for over 30 days; roughly 1 in 5 (1,470) have been sitting unfilled past 60 days.
| Age threshold | Stuck jobs | % of all active |
|---|---|---|
| > 30 days | 3,713 | 49.7% |
| > 60 days | 1,470 | 19.7% |
| > 90 days | 0 | 0% (window limit) |
| State | Stuck >60d | Total active | % stuck |
|---|---|---|---|
| NE — Nebraska | 36 | 53 | 67.9% |
| MN — Minnesota | 29 | 70 | 41.4% |
| WA — Washington | 38 | 109 | 34.9% |
| GA — Georgia | 38 | 132 | 28.8% |
| WI — Wisconsin | 31 | 108 | 28.7% |
| MT — Montana | 14 | 50 | 28.0% |
| OR — Oregon | 26 | 95 | 27.4% |
| MA — Massachusetts | 54 | 209 | 25.8% |
| CA — California | 100 | 390 | 25.6% |
| PA — Pennsylvania | 82 | 326 | 25.2% |
| FL — Florida | 116 | 661 | 17.5% |
| TX — Texas | 65 | 443 | 14.7% |
| NY — New York | 48 | 342 | 14.0% |
| Population tier | Stuck >60d | Total active | % stuck |
|---|---|---|---|
| Low-population (<2M) | 105 | 410 | 25.6% |
| Mid-population (2–6M) | 220 | 972 | 22.6% |
| High-population (>6M) | 849 | 4,629 | 18.3% |
Nebraska has 53 radiology openings — 36 have been sitting unfilled for over two months. The lowest-population states show stuck-rates 40% higher than the highest-population states. The shortage isn't national — it's concentrated in places radiologists won't move to.
| Group | Median max salary | Sample size |
|---|---|---|
| Stuck states (NE, MN, WA, GA, WI, MT, OR, MA) | $550,000 | 1,124 |
| Fast-cycling states (FL, TX, NY, NC, OH, IN) | $725,000 | 3,454 |
| State | Stuck >60d | Fast-filled <14d | Gap |
|---|---|---|---|
| OH — Ohio | $500K | $750K | −$250K |
| FL — Florida | $699K | $1,000K | −$301K |
| CA — California | $530K | $600K | −$70K |
| NY — New York | $525K | $600K | −$75K |
| GA — Georgia | $625K | $690K | −$65K |
| MA — Massachusetts | $501K | $500K | +$1K |
| TX — Texas | $799K | $750K | +$49K |
| MN — Minnesota | $850K | $622K | +$228K |
In most states, money is the lever — stuck jobs pay less; raising offers would likely move the needle. But in Texas and Minnesota the inverse holds: stuck positions pay MORE than fast-filling ones. The constraint there isn't compensation — it's lifestyle, on-call burden, location, or workflow. Exactly what AI automation and remote-friendly setups can change.
Across 18 sources, the same radiologist position is listed on average 1.89 times.
| Company / Source | Total posts | Unique jobs | Re-post ratio |
|---|---|---|---|
| ASER (subspecialty board) | 56 | 5 | 11.2× |
| SPR (subspecialty board) | 64 | 6 | 10.7× |
| Loma Linda Univ. Faculty | 58 | 10 | 5.8× |
| SIR (subspecialty board) | 71 | 13 | 5.5× |
| Indiana University Health | 74 | 14 | 5.3× |
| Tufts Medicine | 72 | 15 | 4.8× |
| Trinity Health | 117 | 25 | 4.7× |
| SAR | 103 | 23 | 4.5× |
| Oregon Health & Science Univ. | 93 | 22 | 4.2× |
| UVA Health | 89 | 24 | 3.7× |
| Position | Times re-posted |
|---|---|
| Indiana University Health — Radiologist (IN) | 28× |
| Bayhealth — Radiologist (DE) | 26× |
| Vanderbilt University Medical Center — Radiologist (TN) | 20× |
| Ochsner Health — Chest Radiologist (LA) | 19× |
| Dartmouth-Hitchcock — Radiologist (NH) | 19× |
| Source | Median lifespan | Closed postings |
|---|---|---|
| 9.0 days | 2,952 | |
| SPR | 12.0 days | 235 |
| AuntMinnie | 14.6 days | 2,596 |
| Indeed | 15.1 days | 177 |
| Hospital sites (direct) | 15.1 days | 484 |
| SIR | 24.0 days | 507 |
| RSNA | 28.7 days | 1,628 |
| ACR | 30.0 days | 1,582 |
| NEJM | 30.0 days | 767 |
| PracticeLink | 31.4 days | 401 |
LinkedIn postings live 9 days because recruiters re-list constantly to game freshness. ACR, NEJM, and RSNA postings live ~30 days because they run for a contracted period. Same opening, different signal. National "shortage" numbers from aggregator scrapes need a ~47% deflation before being trusted.
Some employers close 99% of openings within 60 days. Others can't get past 55%.
| Company | Active jobs | Stuck >60d | % stuck |
|---|---|---|---|
| Radiology Partners (PE-backed) | 988 | 12 | 1.2% |
| SAR | 103 | 10 | 9.7% |
| Jefferson Health | 42 | 12 | 29% |
| Diversified Radiology | 40 | 12 | 30% |
| MUSC Health | 40 | 14 | 35% |
| Cleveland Clinic | 165 | 75 | 45% |
| Oregon Health & Science University | 23 | 11 | 48% |
| AdventHealth | 21 | 11 | 52% |
| University Health Associates | 38 | 21 | 55% |
| Radiology Associates of North Texas | 35 | 20 | 57% |
| Hospital-based private practice (composite) | 73 | 61 | 84% |
| Sanford Health – Marshfield Clinic | 16 | 15 | 94% |
| Medicus Healthcare Solutions | 27 | 27 | 100% |
PE-backed groups average $702K salary max — Cleveland Clinic-tier academic offers $515K. Yet the PE-backed platform fills 99% of 988 openings within 60 days; the academic system fills only 55% of 165. The difference: speed of decision, remote-friendly options, partnership tracks, no-call structures. PE-backed: 32% AI mentions, 44.7% remote, 37.6% no-call. Academic: 13% / 8.7% / 11.2%.
Looking only at the reliable extreme — postings closed in under 14 days — reveals a clear subspecialty hierarchy of demand intensity.
| Subspecialty | Fast-fill % | Closed sample |
|---|---|---|
| Pediatric radiology | 79.9% | 154 |
| Emergency radiology | 78.0% | 82 |
| Interventional radiology | 43.8% | 210 |
| Chest / cardiothoracic | 38.3% | 235 |
| Mammography | 34.8% | 394 |
| Cardiac | 34.2% | 117 |
| Body / abdominal | 33.4% | 725 |
| General radiology | 32.3% | 384 |
| IR | 32.2% | 676 |
| Breast imaging | 29.3% | 150 |
| MSK | 27.2% | 394 |
Pediatric and emergency radiology fill at ~2× the speed of body/MSK/general. Hospitals can't operate without them — and the candidate pool is structurally smaller. Body and general fill slowly because the labor pool is larger but the work is increasingly commoditized — the segments most exposed to AI automation.
Body and chest radiology — most directly addressable by CT-report automation — represent 15.5% of active US radiology openings.
| State | Active body openings |
|---|---|
| FL — Florida | 121 |
| TX — Texas | 97 |
| NY — New York | 58 |
| PA — Pennsylvania | 51 |
| CA — California | 44 |
| OH — Ohio | 44 |
| CO — Colorado | 38 |
| NC — North Carolina | 37 |
| IL — Illinois | 37 |
| MA — Massachusetts | 30 |
The share of new postings mentioning AI tools climbed from 14.7% to 21.2% across consistent 14-day windows over the tracking period. ~80% of postings still don't reference AI at all.
| Period | Rate |
|---|---|
| Mar 9–23 | 14.7% |
| Apr 6–20 | 17.2% |
| Apr 20–May 4 | 21.2% |
| May 4–18 | 19.4% |
| Ownership type | % mentioning AI |
|---|---|
| PE-Backed | 32% |
| Independent | 21% |
| Academic | 13% |
| Hospital System | 10% |
| Staffing | 6% |
| Government | 3% |
6.5 percentage points of growth in AI mention rate over 6 weeks. If the trend holds, ~30% of US radiology job postings will reference AI tools by year-end. PE-backed groups are running 10× ahead of government-owned hospitals on this dimension.
Download the print-ready report with all 11 pages of charts, layouts, and full source notes.
Download PDF →