![\"CT](\"https:\/\/rtmedical.com.br\/wp-content\/uploads\/2026\/08\/tempo-laudo-radiologia-crescimento.jpg\")
How the study was built<\/h2>\n
Researchers analyzed 2.9 million Medicare fee-for-service claims for outpatient exams performed in offices and hospitals between 2014 and 2024. The data source matters: Medicare covers the over-65 population that drives much of the cross-sectional imaging demand, and its administrative records carry both service and interpretation timestamps — which lets you reconstruct turnaround time (TAT) without relying on voluntary reporting from each practice.<\/p>\n
TAT is one of the most closely watched metrics in radiology, and for good reason. It works as a productivity barometer, it shows up in service-level agreements and, above all, it defines how long a patient and a referring physician sit waiting on a clinical decision. When TAT climbs consistently for several years, what you are measuring is not seasonal noise — it is a structural mismatch between supply and demand.<\/p>\n
The time series splits into two clearly different regimes. Between 2014 and 2021 — seven years — TAT rose only 14%, from 0.091 to 0.104 days. Converting to the unit a reading room actually thinks in, with $TAT_{h} = TAT_{days} \\times 24$, that means going from roughly 2h10 to 2h30. After 2021 the curve changes character: up 60% in 2023 and 27% in 2024, closing the period at 0.251 days — about six hours.<\/p>\n
CT leads the decline at 381% over ten years<\/h2>\n
The modality breakdown is the most useful part of the paper for anyone running a department. Over the decade, TAT grew 381% in computed tomography, 278% in MRI, 267% in ultrasound and 126% in conventional radiography. The degradation is broad, but it bites hardest in the cross-sectional modalities.<\/p>\n
That makes physical and operational sense. A chest CT that ten years ago was reconstructed at 5 mm now ships with hundreds or thousands of thin slices, multiplanar reformats and often pre- and post-contrast series. Each of those layers adds eye-on-screen time and raises the odds of an incidental finding that must be described and given a recommendation. Radiography, by contrast, has barely changed in information density — which is why it deteriorated least. Worth noting that dose per exam has been falling across the same window, as the US CT radiation dose data<\/a> show: what grew was not protocol aggressiveness, it was the amount of data to read.<\/p>\n Here the authors are candid: they do not know. Their hypothesis is that practice leaders adopted containment measures to absorb demand — opening extra shifts, hiring moonlighting coverage, reactivating retired or near-retirement radiologists. None of those options is sustainable in the medium term, and the Neiman team itself lands on the uncomfortable conclusion: the radiology workforce may be operating “at or near maximum capacity.”<\/p>\n An earlier study from the same group had already described a “hockey stick” effect, with the TAT curve bending sharply upward from 2022 — but it only ran through 2023. What is new here is the addition of 2024, and the demonstration that slower growth does not mean a return to the previous baseline. The line is still climbing, just less steeply.<\/p>\n For anyone managing workflow, the practical message is that TAT gains now come less from “reading faster” and more from removing friction. That includes revisiting how the PACS worklist prioritizes cases, eliminating manual data re-entry between RIS and reporting system, standardizing structured templates and cutting low-value studies — and there is evidence that clinical knowledge on the ordering side reduces unnecessary requests<\/a>.<\/p>\n There is also a warning here about workforce policy. Radiology was the first specialty to physically decouple from the hospital building, and decisions that ignore that come with a bill: exactly what happened when radiologists left the Veterans Affairs network after the remote work ban<\/a>, with a direct hit to the reporting backlog. In Brazil, where teleradiology underpins much of the coverage outside major cities, the same logic applies: the bottleneck is rarely the scanner, and almost always the person reading it. Screening decisions feed the same queue — see the debate on whether screening ultrasound is still needed in the DBT era<\/a>.<\/p>\n The work has boundaries that invite caution. These are outpatient Medicare fee-for-service data, which excludes the pediatric population, commercial plans, Medicare Advantage and most emergency and inpatient volume. The reported TAT is also a central measure, and central measures hide the tail: a six-hour median is perfectly compatible with a small fraction of exams waiting days. For patient experience, the tail is what hurts.<\/p>\n Two forces will fight over this curve in the coming years. On the pressure side, population aging and screening expansion. On the relief side, automated triage and prioritization tools, structured reporting and language-model support in drafting. Whether those tools genuinely shorten TAT or simply shift time toward validation tasks is the question the next batch of data will have to answer.<\/p>\nWhy the curve decelerated in 2024<\/h2>\n
What changes in day-to-day practice<\/h2>\n
Outlook and limitations<\/h2>\n