Some U.S. states are just “stuck” when it comes to filling vacant radiology jobs, according to a new analysis from RadBoard.io. The report casts doubt on the severity of the overall radiologist shortage.
Filling jobs is considered stymied when more than 50 positions stay open for longer than two months. That trend appears worst in the Nebraska and Minnesota markets, RadBoard’s analysis found. In contrast, Florida and Texas markets rank at the top of fast-cycling states where jobs fill most quickly.
“The [radiology] shortage isn’t national — it’s concentrated in places radiologists won’t move to,” wrote researcher Kirill Lopatin, who studied the full lifespan of 20,775 radiology job postings between March 9, 2026 and May 26, 2026. Notably, about 47% of open jobs turned out to be re-posts of the same role, according to the report.
“National ‘shortage’ numbers based on aggregator scrapes need [roughly a] 47% deflation before being trusted,” Lopatin wrote. For the period, the analysis found that the same radiologist position was listed an average of 1.89 times across 18 sources.
Of 10,979 unique postings, about one in five, or a total of 1,470 active openings, remained unfilled for over two months. Lopatin reasoned that these jobs were “geographically inconvenient.” RadBoard dubbed them “zombie jobs.”
“In most states, money is the lever — stuck jobs pay less,” the report explained. “Raising offers would likely move the needle. But in Texas and Minnesota, the inverse holds: stuck positions pay more than fast-filling ones. There, the constraint isn’t compensation. It’s lifestyle, on-call burden, location, or workflow — exactly the things AI automation and remote-friendly setups can change.”
Churn is most evident in the rate in which jobs appear and disappear from job post sources. About one-quarter of jobs are gone within seven days, either filled, pulled, or revealed as a duplicate, according to Lopatin. About one-third are cleared between 15 and 30 days. About 28% are closed between 31 and 60 days. Roughly 2.3% languish from 61 to 90 days.
“The market doesn’t have a single ‘fill rate’ — it has two markets layered on top of each other,” noted Lopatin, who called out the subspecialty boards [American Society of Emergency Radiology] ASER and [Society for Pediatric Radiology] SPR for having the highest volume of job reposts compared to other sources.
Lopatin also looked into job postings with the RSNA, the American College of Radiology (ACR), and NEJM, pointing out that they’re contracted for 30-day periods. LinkedIn job postings refreshed about every nine days, the reported noted.
“Looking only at the reliable extreme — postings that closed in under 14 days — reveals a clear subspecialty hierarchy of demand,” Lopatin found. Pediatric radiology and emergency radiology posted fast-fill rates of 79.9% and 78%, respectively, while musculoskeletal and breast imaging hit the low end at 27.2% and 29.3%, respectively.
Body and general radiology fill slowly because the labor pool is larger, but the work is increasingly commoditized — these are also the segments most exposed to AI automation, according to Lopatin.
RadBoard’s analysis further differentiated radiology practice operating models for their effect on attracting applicants. PE-backed platforms filled 99% of 988 openings within 60 days, whereas academic systems filled 55% of 165. The difference was in the speed of decision, remote-friendly options, partnership tracks, and no-call structures, according to the report.
In this report, RadBoard also predicted that approximately 30% of U.S. radiology job postings will reference AI tools by year-end, suggesting the AI adoption gap is narrowing.
Find the full report here.