Cool Down on the CT Scanning?

As the L.A. Times reports, two new studies, from researchers at the University of California-San Francisco and the National Cancer Institute, suggest that hospitals may want to cut down on the volume of CT scans. One of the studies estimates that “about 29,000 future cancers could be related to CT scans performed in the United States in 2007 alone.” The researchers’ advice to patients: keep track of how many CT scans you receive, and argue against them with your doctor. (HT: David Friedman) [%comments]

William G

From what I can gather, MRI gives a greater resolution without the dangers associated with CT. Why does anybody get cat scans anymore? Is it simply a matter of convenience (faster to perform)?


@ Wililam - There's a lot of reasons CT is need in addition to MRI... The cost of a CT is usually lower (due to the time needed to perfom as you mentioned). Also, certain tissues cannot be differentiated reliably with an MRI, but can be imaged well with CT. Another big reason is that the "M" in MRI stands for "Magnetic", so metal near an MRI machine is dangerous. It's not an option for a lot of people with stainless steel implanted devices, pacemakers, difibrillators, etc. There's a multitude of other situations where CT is preferred over MRI, but there's tradeoffs with both.


It's all about incentives: do the unnecessary interventions - tests, scans, exploratory surgery - because there's revenue in this activity.

And, more importantly, being defensive and doing extra tests is your only defence against a medical negligence lawsuit that can end your career.

What incentive is there, I wonder, for running a scare story about the cumulative risk to patients and to the wider population of the extra X-rays?

The moment you ask the mildest of questions, the hospital administrators will seize upon your words: "SHE'S REFUSING TREATMENT!!! EVERYBODY LOOK OVER HERE!" and the only way you'll get the shouting to stop will be signing a truly watertight disclaimer that absolved them of all responsibility for any failure of treatment, for whatever cause, on an irrational patient who's refusing treatment.

It's an economics column: would anybody care to ask "Who profits?"


MRI is excellent for imaging soft tissue. Hard tissues require CTs. CTs are more convenient than standard x rays, but MRIs and CTs are called for in different situations.


Just a word of caution, if your doctor recommends something, even if its wrong, you refuse it and he notes it, your health insurance company will use it as a way cut coverage via resision in the future by playing the "refused care" card.


This is a job for the Freakanomic statistics team. My wife is a CT/MRI Technologist. She read several of the news reports (we haven't found the actual data), and she says that this is not necessarily comparing apples to apples. One example is no one mentions how many thousands and thousands of patients by early detection of cancer.

Also, these reports mention "since 1980" as the comparison with today. In 1980, CT was fairly new and was not capable of detecting many things that today's CTs can. Plus modern CTs expose the patient to much less radiation than earlier models.

Another note. One of the reports stated that the amount of radiation exposure is greater in patients seen in the evenings over those seen during the day. Machines don't care what time it is. There has to be something else going on. Maybe later shifts get the patients who need more studies, or who need repeats? (sorry, I can't find the source right now0?

On the other hand, insurance and the drive to reduce costs does cause more CTs to be given. Versus MRI, they are cheaper and in some cases, just as effective. This issue gets complicated when MRI is the better modality, but because of cost, CT is done first, the MRI if the CT is inconclusive.

One example is in suspected stroke patients who come into the emergency room. Some hospitals send these patients to CT first, even though the more expensive MRI can find evidence of stroke much better and earlier in the process than CT.

As with all medical issues, it is complicated and someone with a statistics background should sit down with some imaging specialists and review the data and conclusions


Eric M. Jones

Every hour flying in a jet over 30,000 is the rough equivalent of a chest X-Ray. If the background radiation from Radon in your house was part of the Real Estate pitch, you wouldn't have bought it . I used to stare at the bones in my feet inside one of those X-Ray shoe fitting machines.

Everything has a cost.


So, a few months ago, I went to the emergency room for a kidney stone. They verified that I had the stone with not one, but two, CT scans. The shocking thing is that they didn't bother to check the results of the first scan before doing the second! (I only found out that I got two scans after the fact, when I got the bill -- when you're face down and in pain, you don't really notice that they ran you through the machine twice.)

Joel Upchurch

I think there are two side to the discussion. There has been some evidence that physician owned medical imaging equipment can lead to over use by physicians trying to make sure the equipment earns it's keep.

OTOH, many of these about radiation risks studies are based on what is called the Linear No Threshold hypothesis, which assumes that there is no safe threshold under which radiation will not cause damage and increased risk on cancer. Studies conducted in areas with high levels of natural background radiation, such as Ramsar in Iran, make the LNT hypothesis somewhat questionable.


CT scans are required for imaging bone when the plain films do not provide adequate resolution, or when the anatomy of the bone is complex and can't be easily seen with a plain film.

An MRI is for soft tissue. When looking at a bone with it, all you see are the marrow contents, and not the hard outer cortex. A CT is preferred.

I frankly am getting tired of the assumption that physicians order tests to 1) directly benefit financially, 2) practice defensive medicine or 3) avoid work. The vast majority of physicans are just trying to do right by their patients.


In an emergency setting, the time difference between getting a CT scan and an MRI is crucial. A CT can be run in seconds (it takes longer to transport the patient to the scan than the scan itself). An MRI on the other hand, could take 45 min or more. That's simply unacceptable in an emergency setting, especially in stroke, where there is a 3 hour time limit from symptom onset before giving clot-dissolving drugs such as tPA becomes very dangerous instead of helpful.

KJP - the financial benefit argument really only applies to radiologists and some orthopedics. Most doctors, as you state, don't benefit financially from any tests or scans that they order. But point 2 does apply - there is no downside from the physician point of view of ordering a CT, especially in emergency medicine when you don't know a patient or they are nonverbal. If you don't order a CT, you could face a lawsuit where you would be judged by a group of your supposed peers who have no experience in triage or medical decision making. If you order a CT, your patient gets a higher bill, but you have piece of mind (and possibly useful information). There is no disincentive!


Mark S.

This section of the article is key:

"Researchers' conclusions are based on the assumption that the patients receive a normal dose of radiation, but that is not necessarily a good assumption. Smith-Bindman and her colleagues studied the radiation doses received by 1,119 adult patients at four San Francisco Bay Area hospitals between Jan. 1 and May 30 of 2008.

Estimating the amount of radiation received by the patients, they concluded that dosing was highly variable both between institutions and within them as well. Some patients got below-normal doses."

It would be helpful to
-measure the radiation of individual instruments
-assess the variety of procedures/doses used by radiologists
-determine the highest/lowest combinations
-match up with the disease conditions for which the procedure was ordered in the first place.

Then the risk benefit of of the CTS with a correctly functioning instrument can be assessed quantitatively.

The article does not mention if the differences in radiation dose levels applied was due to clinical findings that some of these higher dose levels achieve better resolution or if it was just a legacy standard procedure that no one had thought to investigate and update.

From my knowledge on how the health care system works and on how difficult it is to communicate best practices and have every provider to adhere to them, its probably a legacy setpoint/procedure/dose level that no one has thought to update.


chris m

Newer CT Scanner have more sensitive detectors and require less radiation to create images. Also newer scanner have all types of dose saving methods for delivering the needed dose to create images.

Yes it is true that many routine CT scans are ordered when other exams could similarly provide results that do the same. Also it is true that some doctors seem to rely on Radiologist multiple times to practice medicine for them.

One must keep in mind that Its easy to be critical of the negative aspects of CT but in critical and life threatening cases such as a bleed in the head, rupture of aorta, trauma from an MVA, pulmonary embolism, and appendecitis - CT is currently the only way to go.

If one is in doubt they should talk to the hospital surgeons, orthopedic surgeons, Cardiologist, and vascular surgeons on the value of CT scans.

MRI is a great imaging modality but currently has to many contraindications to allow it to surpass CT has the modality of choice for emergancy trauma or patients that are clausterphobic and cases that are time sensitive.



I had a ct scan and the machine kept spinning for a long while after the ct scan of the chest was completed. I was told the machine keeps spinning for cool down and that no radiation was emitted during the fast spinning of the machine. I have never seen that before during any of the other CT scans that I have had. What is the correct answer to that concern?