Slowing the increase in the population dose resulting from CT scans

Radiat Res. 2010 Dec;174(6):809-15. doi: 10.1667/RR1859.1. Epub 2010 Aug 23.


The annual number of CT scans in the U.S. is now over 70 million. The concern is that organ doses from CT are typically far larger than those from conventional X-ray examinations, and there is epidemiological evidence of a small but significant increased cancer risk at typical CT doses. Because CT is a superb diagnostic tool and because individual CT risks are small, when a CT scan is clinically indicated, the CT benefit/risk balance is by far in the patient's favor. Nevertheless, CT should operate under the ALARA (As Low As Reasonably Achievable) principle, and opportunities exist to reduce the significant population dose associated with CT without compromising patient care. The first opportunity is to reduce the dose per scan, and improved technology has much potential here. The second opportunity is selective replacement of CT with other modalities, such as for many head and spinal examinations (with MRI), and for diagnosing appendicitis (selective use of ultrasound + CT). Finally, a fraction of CT scans could be avoided entirely, as indicated by CT decision rules: Clinical decision rules for CT use represent a powerful approach for slowing down the increase in CT use, because they have the potential to overcome some of the major factors that result in some CT scans being undertaken when they are potentially not clinically helpful. In the U.S. and potentially elsewhere, legislative approaches are a possible option, to improve quality control and reduce clinically unneeded CT use, and it is also possible that upcoming changes in heath care economics will tend to slow the increase in such CT use.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Humans
  • Neoplasms, Radiation-Induced / epidemiology
  • Neoplasms, Radiation-Induced / etiology
  • Radiation Dosage
  • Risk
  • Tomography, X-Ray Computed / adverse effects*
  • Tomography, X-Ray Computed / statistics & numerical data