U.S. adult CT radiation doses have decreased substantially since 2014, with chest scans showing up to 31% reductions, driven by advances in scanner technology, reconstruction methods, and improved clinical protocols according to a new study analyzing 5.2 million CT acquisitions.
- 31% reduction in chest CT scans with contrast material compared to 2014 benchmarks
- Study analyzed 5.2 million CT acquisitions from 592 U.S. facilities across 10 examination categories
- Average 22% decrease in diagnostic reference levels and 9% decrease in achievable doses overall
- Key drivers include iterative reconstruction, machine learning techniques, improved detector efficiency, and protocol optimization
- Lower doses reduce already small cancer risks from CT while maintaining diagnostic benefits
U.S. adult CT radiation doses have decreased substantially since 2014, with up to a 31.2% drop seen in chest scans with contrast material, according to a study published July 28 in Radiology.
The finding is from a retrospective analysis that included approximately 5.2 million adult CT acquisitions from 592 U.S. facilities across 10 examination categories, nearly fourfold more than prior U.S. benchmarks, noted lead author Kalpana Kanal, PhD, of the University of Washington in Seattle, and colleagues.
“The observed reductions are consistent with broad changes in CT technology and clinical practice over the past decade,” the group wrote.
Approximately 10 years have elapsed since the data underlying the most recent U.S. national benchmarks were collected, underscoring the need for updated reference values, as current national benchmarks may not reflect recent advances in scanner technology, reconstruction methods, and protocol changes, the authors explained. To address the gap, the researchers aimed to establish updated U.S. national adult CT diagnostic reference levels (DRLs) and achievable doses (ADs) derived from 2025 data.
The researchers analyzed approximately 5.2 million adult CT acquisitions performed at 592 U.S. facilities between January 1 and December 31, 2025, using the Imalogix (Imalogix Inc., Bryn Mawr, Pa.) radiology optimization platform. The 10 most common adult CT examination categories were mapped using standardized RadLex Playbook identifiers, with dose metrics normalized at the acquisition level.
According to the results, relative to the last national benchmarks in 2014, volume CT dose index (CTDIvol) achievable doses (ADs) decreased by an average of 9% and diagnostic reference levels (DRLs) decreased by an average of 22%. The largest reductions appeared in chest examinations with contrast material, where DRLs fell 31%, and chest examinations without contrast, where DRLs fell 27%.
Bar graphs show comparison of volume CT dose index (CTDIvol) (A) achievable doses (50th percentile) and (B) diagnostic reference levels (75th percentile) between 2014 (see reference 4 by Kanal et al) and 2025 (present study) for 10 common CT examination types performed without contrast (w/o) and with contrast (w/). Across all 10 adult CT examination categories, CTDIvol achievable doses and diagnostic reference levels generally decreased from 2014 to 2025, with the largest reductions for torso examinations (chest; abdomen-pelvis; and combined chest, abdomen, and pelvis), intermediate reductions for neck and cervical spine, and minimal change for head examinations.RSNA
“Although this study was not designed to attribute causality, the observed reductions are consistent with broad changes in CT technology and clinical practice over the past decade, including adoption of iterative and machine learning–based reconstruction techniques, refinements in tube-current and tube-potential modulation, improved detector efficiency, and increasing protocol optimization,” the group wrote.
In an accompanying editorial, Cynthia H. McCollough, PhD, of the Mayo Clinic in Rochester, MN, wrote that the attention paid to dose reduction technologies and best practices over the past two decades is bearing fruit.
“The good news that CT doses continue to decrease comes at a time when large radiation epidemiologic studies have reported observed associations between the use of CT and cancer incidence,” she wrote.
However, such studies used the older, higher doses, and the reported associations may no longer be significant at the lower doses of today, McCollough added.
“Future studies at today’s lower doses will be needed to estimate absolute increases in cancer risk caused by CT. But whatever those numbers may be, based on the assumed linearity of the association between dose and cancer incidence, lowering CT doses will lower the already small risks from CT, ensuring that patients can safely receive the benefits of this powerful diagnostic tool,” she concluded.
Read the full study here.
