CT use more than triples in the ED, but use of the technology may be linked with a significant drop in hospitalizations

Experts urge providers to take the lead on curbing excessive use of CT

All of the discussion in recent years about the risks from exposure to radiation from computed tomography (CT) scans has hardly dampened enthusiasm for the technology in the ED. To the contrary, a new study suggests that CT use in the ED increased by a whopping 330% between 1996 and 2007, according to a retrospective look at data from the National Hospital Ambulatory Medical Care Survey, which is a national survey of services in emergency departments conducted by the Centers for Disease Control in Atlanta, GA. However, the study also suggests that the increase in CT use may be associated with a dramatic reduction in hospitalizations. In 1996, when the study period began, the rate of hospitalization following a CT scan was 26%. This rate dropped to 12.1% by 2007, when the study period concluded.1

The reduction in hospital admissions can certainly be seen as positive with respect to costs. Further, the data suggest that information gleaned from CT scans played a key role in provider determinations that hospitalization was not necessary. However, the steep rise in CT use is a concern, acknowledges Keith Kocher, MD, MPH, the lead author of the study and an emergency medicine physician in the University of Michigan Health System in Ann Arbor, MI.

"The reality is that we are probably doing too many CT scans," says Kocher. "The data from 2008 is already out, and [CT use in the ED] continued to increase, but this is not sustainable."

The Joint Commission clearly agrees with Kocher. The accrediting agency has issued a Sentinel Event Alert, urging hospitals to carefully consider the potential danger posed by repeated exposure to ionizing radiation, and to take steps to ensure that the doses used in CT scans are as low as possible to achieve study standards, and that staff are thoroughly trained about the issue.

Look for opportunities to optimize

Kocher's study doesn't specifically say where CTs are being used unnecessarily, but it does offer some suggestions on where emergency medicine providers might be able to do a better job of optimizing their CT use. One of these areas is in the use of CTs in patients with flank pain.

"If you look from the beginning of the study to the end of the study, there is about a nine times increase in the number of CT scans being used [in patients with flank pain]," says Kocher. "Most of the time, ED physicians who are doing CTs on these patients are looking for kidney stones, yet it is unclear in my mind whether a CT scan is necessary to diagnose a kidney stone."

Kocher points out that ultrasound, or even simply using clinical judgment, are alternative options to consider with respect to diagnosing patients with kidney stones. Further, he questions whether the ability to diagnose a patient definitively with a kidney stone with the use of CT would necessarily translate into a change in how that patient is treated or managed.

Other areas where decision rules have been developed to guide physicians with respect to testing include head injury, neck injury, and pulmonary embolism, says Kocher, but he also states that good evidence showing when patients should receive CT scans and when they should not is scant. "A lot of those studies just haven't been done," he says. "Also, on some level, every patient situation is very individual, so making blanket statements about how things should be done in every case is challenging, and it requires sophistication."

Lead discussions on CT use

One approach to the issue, says Kocher, might be for physician groups to make data available so physicians can see how much they are using CTs for testing relative to their peers. "The challenge is that this suggests that we should all regress to the mean of the group, so you are relying on the mean being, hopefully, the appropriate amount of testing," he says. "But this approach would certainly curb the extremes."

Kocher also encourages ED managers and other emergency providers in leadership positions to lead discussions about the issue and collect ideas on what strategies might be effective and acceptable in their own work settings. "I don't think they could unilaterally mandate how physicians should be practicing … but there is certainly a lot of space to make some change," he says.

Further, to the extent that such measures can include the larger health care community, the easier it will be for ED physicians to make changes. "When you are working in the ED, you feel like you are on an island, you are doing your best, but there are all these pressures that are external to you," says Kocher, noting that patients often request CTs, primary care practitioners often send patients to the ED for CTs, and consulting specialists often want their patients to have CTs before they are admitted to the hospital. "The more that you can remove that sense of being isolated in the decision … I think that might ameliorate a lot of the pressures."

The skyrocketing costs associated with imaging tests will eventually curb the use of CTs one way or another, suggests Kocher. Consequently, he advises providers to propose some of their own approaches now. "The solutions I have seen are few and very ad hoc," he says. "The truth is if we, as ED physician groups, don't come up with our own strategies, somebody else is going to come up with the strategies for us, and they will be imposed on us. And those situations might not be the solutions that we want."


  1. Kocher K, Meurer W, Fazel R, et al. National trends in use of computed tomography in the emergency department. Ann Emerg Med. 2011 Aug 9. [Epub ahead of print]


  • Keith Kocher, MD, MPH, Emergency Medicine Physician, University of Michigan Health System, Ann Arbor, MI. E-mail:kkocher@umich.edu.