The idea that many primary care physicians would not know what diagnostic imaging tests they should order to diagnose a patient's symptoms might seem incongruous, considering their training and the faith we place in them.
Likewise, it seems implausible that some doctors – perhaps because of time constraints or the confidence they have in their referral base – might fall into a habit of routinely ordering costly CTs, with levels of ionizing radiation at potentially cancer-causing levels, even when an ultrasound, MRI, or a simple X-ray would be a safer choice.
But that is indeed what federal and research reports say has been happening far too often. Additionally, when doctors own the machines, they order more of these tests.
Now, medical journals and federal public policy documents are sounding a refrain: too many doctors are ordering too many expensive, unnecessary and possibly cancer-provoking tests. There has to be a way to curtail it.
That's why the Centers for Medicare & Medicaid Services (CMS) has set aside $10 million for demonstration projects to develop computerized "decision support systems" or DSS programs that assist doctors to make the right choices from among 11 advanced imaging procedures. Thousands of physicians will be recruited, and compensated, to test various systems over a two-year period.
The DSS will give algorithms that help doctors select from the following tests: MRI of the brain, lumbar spine, knee and shoulder, CT of the brain, thorax, sinus, abdomen, pelvis and lumbar spine, and Spect MPI. These are tests associated with high expenditures and utilization in the Medicare fee-for-service population, CMS says. The demonstration will rely on specified medical specialty society guidelines for these procedures.
"Independent of this demonstration, CMS recently examined the trends in ionizing radiation exposure due to diagnostic imaging services among fee-for-service Medicare beneficiaries from 1997 through 2007," the agency said when it announced the demonstration project last week.
"Annual ionizing radiation exposure increased dramatically during this time. Thus, the potential for this demonstration to help reduce unnecessary ionizing radiation exposure has the most definite impact on the quality of care provided to Medicare beneficiaries."
It is almost certainly money well spent.
No one knows exactly how many increased cases of cancer are caused by diagnostic imaging. But the Medicare Payment Advisory Commission's June report, "Aligning Incentives in Medicare," provided this citation:
"Although an individual's risk of developing cancer from a single test is small, these risks are applied to a growing number of patients. A recent study (from the December issue of the Archives of Internal Medicine projected that approximately 29,000 future cancers could be related to CT scans performed in the United States in 2007."
In the New England Journal of Medicine last August, Emory University School of Medicine researchers and others noted that in a sample of 655,613 non-elderly adults who underwent at least one imaging procedure associated with radiation exposure, CT and nuclear imaging accounted for 75.4% of the cumulative effective dose, and 81.8% of the total dose administered in outpatient settings.
In the same study, the authors noted that far too many physicians seem unaware that these doses of radiation carry an increased risk of cancer. "In one study of U.S. health care providers using CT in patients with abdominal and flank pain, less than 50% of radiologists and only 9% of emergency department physicians reported even being aware that CT was associated with an increased risk of cancer.
"An improved understanding of the risks of radiation is clearly needed, and raising such awareness among providers has been the focus of recent efforts," the researchers wrote.
The issue of CT radiation became even more of public concern since last August, when Cedars Sinai Medical Center in Los Angeles realized that incorrect settings on its 64-slice CT machines caused 260 patients who underwent brain scans to be exposed to ionizing radiation eight times the normal doses. About 40% developed hair loss and rashes as a result over an 18-month period, and all have been alerted they are at increased risk of developing cataracts, hospital officials said.
A number of these decision support systems already exist and work well, according to an editorial in the March 11 New England Journal of Medicine David J. Brenner of the Center for Radiological Research at Columbia University Medical Center in New York, wrote about success with the American College of Radiology's (ACR) "appropriateness criteria" DSS, a system the ACR offers to physicians at no cost.
In one study of 200 patients administered radiographic imaging tests without the use of decision rules at a level 1 trauma center, "169 underwent CT scanning, resulting in a total of 660 scans. If ACR appropriateness criteria had been applied (which they were not), 44% of these CT scans would not have been performed, but none of the patients with clinically significant injuries would have been excluded from CT imaging," Brenner wrote.
If the ACR's appropriateness criteria DSS system had been applied, Brenner added, "there would have been an estimated 38% decrease in imaging costs in addition to the 44% decrease in CT use and the associated decrease in radiation exposure."
Other larger studies "suggest that 20 to 40% of CT scans could be avoided if decision guidelines are followed, without compromising patient care," he wrote.
Of course there are many defensive legal reasons, and some dealing with patient demand, that influence decisions about what imaging test doctors order.
But there are too many important reasons why doctors should want to participate in this demonstration project. Payers and providers as well as patients need to make sure that these expensive and potentially harmful tests are ordered only when they do more good than harm.