There’s a news frenzy about whether doctors are ordering too many CT scans. Right now this article is the most popular article on US News and World Report. A search of Google News shows 368 news articles on the topic from all over the world.
According to an article in the New England Journal of Medicine, doctors are ordering way too many CT scans. Those “unnecessary” CT scans are putting the lives of patients at risk. The article cites a “straw poll” of pediatric radiologists (which appears to be a panel discussion in which the audience was apparently able to vote on questions posed by the panel discussants) who opined that 30% of the CT scans ordered were unnecessary. Let me translate that into non-medical lingo:
“We are a group of sub-subspecialists who perform very little direct patient care and, based on our anonymous and unsubstantiated opinions using retrospective analysis of negative CT scan results, those doctors who do perform patient care should not have performed 30% of the CT scans they ordered.”
So an off the cuff vote at a roundtable discussion is the best evidence that the authors could use to substantiate their assertion that clinicians order 30% too many CT scans. In fact, their article cites this “straw poll” twice:
Tellingly, a straw poll of pediatric radiologists suggested that perhaps one third of CT studies could be replaced by alternative approaches or not performed at all.
However, if it is true that about one third of all CT scans are not justified by medical need, and it appears to be likely, perhaps 20 million adults and, crucially, more than 1 million children per year in the United States are being irradiated unnecessarily.
So what’s the news bite that all the reporters dish out? Doctors order 30% too many CT scans and they are needlessly risking your lives! Way to go, guys.
Before anyone puts any further validity into the opinions of the pediatric radiologists whose straw poll has taken this country by storm, the pediatric radiologists should be required to publish another opinion statement. At a minimum, this statement should contain the following two paragraphs:
- Because CT scans are potentially dangerous, we as a society of radiologists agree never to dictate the following sentence into our reports for the rest of our careers: “[insert name of any diagnostic radiology test here] shows no sign of acute pathology. If clinical symptoms warrant, CT scan is advised.” Any time that a clinician sees the phrase “CT scan is advised” they know they have to order one. God forbid that something turns up later on a patient who didn’t get a CT. The radiologist would be the first one wagging his finger on the stand stating “My report clearly states that a CT scan was advised. I have no idea why the clinician didn’t order one.”
- As a specialty society whose members have relatively little patient contact and whose opinions can therefore be based solely on RETROSPECTIVE analysis, we hereby PROSPECTIVELY declare that CT scans are unnecessary for each of the following symptoms: [provide the long list of symptoms for which we need no longer do unnecessary tests so that we can save millions of people from radiation sickness].
No fair including diseases – only symptoms. People with appendicitis don’t come in saying “I have appendicitis”, they come in complaining of “abdominal pain.” The list has to be prospective so all us dumb clinicians know how to properly work up our patients.
After reviewing the strong support for the notion that physicians order too many CT scans, I then went to find out more information about the authors who published this “study.” Guess what? Neither David J. Brenner nor Eric J. Hall holds a medical degree. As far as I can tell, they have never even touched a patient. They are both D. Sc.s or “Doctors of Science.” Dr. Brenner also has a Ph.D. and Dr. Hall also has a D. Phil. or Doctorate in Philosophy in Radiobiology. Both are exceedingly bright fellows, but they are both giving opinions and drawing conclusions that are way outside of their specialties. Both would absolutely be excluded as witnesses if they tried to render an opinion in court regarding patient care.
OK, guys, see this funny looking thing. It’s called a stethoscope. Keep it away from the Bunsen burner. Now put down Immanuel Kant’s biography, walk out of your lab, and use it on a few hundred patients with acute abdominal pain. Then go examine a few hundred other people with headaches. Make your diagnoses. When you’re done and you’ve saved society from all their radiation burns, tally up all the people you’ve misdiagnosed, take time off from defending your lawsuits, and then come back and write your article. Go head-to-head prospectively with someone who walks the walk before you try to talk the talk. Want to write a paper about the different effects of radiation exposure? Fine. You know more about that topic than almost anyone. Leave the recommendations for clinical management to those who actually provide patient care.
Yes, CT scans have risks. CT scans have also been around since the early 70’s. Want to do a good paper? Write an article comparing the rates of different cancers before and after the advent of CT scans. They had the ability to compare real data, but instead used a prestigious journal to scare the public by stating that the radiation from a few CT scans is like being near ground zero at an atomic bomb blast. Shame on NEJM for publishing this article.
If you want to become famous, go make a funny YouTube clip. In the meantime, don’t degrade our profession.
P.S. Does this groundbreaking study mean that we should just keep doing invasive angiograms on patients rather than evolving and doing CTs to rule out coronary disease? Don’t want to get heart cancer, ya know.
Something else keeps running through my mind about this whole issue.
Let’s suppose the assertion(s) that 2-3 CT scans are equivalent to an atomic bomb blast and/or one CT scan increases your chance of cancer by one tenth of a percent is/are true.
Will society be willing to accept the trade-off of not receiving a CT scan to decrease one’s potential cancer risk? We won’t do the scan, but you assume all responsibility if we are not be able to diagnose your traumatic incracranial bleed, your aortic aneurysm, your pulmonary embolism, your lung cancer, your cervical spine fracture, etc.?
This is another reason I think the NEJM article has no good purpose. If a physician performs an ultrasound to reduce radiation exposure and therefore misses an important finding, the patients suffer and the physician can be sued for not doing the “proper test.” If physicians order an MRI instead of a CT to reduce radiation exposure, then they find that access to MRIs is limited, insurance won’t cover the exam if the MRI is available, and then patients complain that healthcare costs are going through the roof. If physicians order a CT scan and patients later develop cancer, the physicians are to blame for performing the CT scan and causing the cancer.
Fear the bad outcome.
Anyone have a suggestion as to how we should “do right” in this quagmire?
Do we now need to create informed consents citing the speculative assertions of a bunch of unnamed pediatric radiologists in order to prevent a flurry of class action lawsuits 5-10 years in the future?
Lifting a box then getting RLQ abdominal “twinge.”
A day and a half of RLQ pain with normal white count.
Left upper quadrant pain with no fever, no appetite, and WBC count of 23k.
To those who actually believe that 30 percent of CT scans are “unnecessary,” on how many of the above cases would you have prospectively ordered CT scans?
All of them, right?
After all, they were all positive for appendicitis.
Thanks to Ten out of Ten for the interesting cases.