When James Duncan, M.D., a radiologist at Washington University in St. Louis, experienced intense pain in his abdomen in 2010, he rushed to a local emergency room. His doctors suspected kidney stones, but they wanted to be sure, so they ordered a CT scan. Duncan remained motionless as the machine captured a detailed, 3D image of his abdomen. He knew that the test was done when the machine stopped whirring. So he was surprised when the scanner kicked back on after a few seconds.

“I later learned that the technician running the CT mistakenly believed that the first scan didn’t include the top of my kidneys, and decided to acquire more images ‘just to be sure,’ ” Duncan says. “The irony: I was getting ready to give a lecture on reducing radiation exposure from medical imaging. And there I was, reluctantly agreeing to a CT scan and then getting overexposed.”

Duncan will never know whether that specific scan caused any long-term harm, because it’s almost impossible to link radiation exposure from any one medical test to a future illness. But like other researchers, he knows that doctors today order millions of radiation-based imaging tests each year, that many of them are unnecessary, and that the more radiation people are exposed to, the greater their lifetime risk of cancer.

X-rays have been used for almost 120 years, but the introduction of computed tomography, or CT scans, in the 1970s, was revolutionary. The new tests, which use multiple X-ray images, allowed doctors to see with unprecedented precision the inner workings of the human body, and earned the inventors of the device the 1979 Nobel Prize in medicine. Use of the tests grew quickly, rising from fewer than 3 million per year in 1980 to more than 80 million now.

But recent research shows that about one-third of those scans serve little if any medical purpose. And even when CT scans or other radiology tests are necessary, doctors and technicians don’t always take steps to limit radiation exposure.

All of that exposure poses serious health threats. Researchers estimate that at least 2 percent of all future cancers in the U.S.—approximately 29,000 cases and 15,000 deaths per year—will stem from CT scans alone. Even some standard X-rays, which expose you to much smaller amounts of radiation, can pose risks if you undergo multiple ones.

“No one says that you should avoid a CT scan or other imaging test if you really need it, and the risk posed by any single scan is very small,” says Marvin M. Lipman, M.D., Consumer Reports’ chief medical adviser. “But the effect of radiation is cumulative, and the more you’re exposed, the greater your cancer risk. So it’s essential that you always ask your doctors why they are ordering an imaging test and whether your problem could be managed without it.”

Given those risks, why are we—and our doctors—so scan-happy?

For one thing, patients aren’t necessarily aware of the danger. A new Consumer Reports survey of 1,019 U.S. adults found that people are seldom told by their doctors about the risks of CT scans and other radiology tests. It’s no surprise, then, that only 7 percent of those who had a nondental X-ray and 2 percent of those who had a CT scan thought they might have received the tests unnecessarily. And only 4 percent ever told their doctor they did not want a CT scan. “That’s worrisome,” says Lipman’s colleague at Consumer Reports, Orly Avitzur, M.D. “Patients need to take the lead on this because their doctor may not.”

Other studies show that doctors themselves often underestimate the dangers CT scans pose. Moreover, some doctors may actually have a financial incentive to order the tests.

“Health care professionals shouldn’t have the right to image children or adults unless they first show that they can do it safely and appropriately, and most of the time in this country, that’s not happening,” says Stephen J. Swensen, M.D., medical director at the Mayo Clinic in Rochester, Minn. “If the scan isn’t necessary or emits the wrong dose of radiation, the risks far outweigh the benefits.”