Uh, You Should Think Twice Before Opting Out of a CT Scan Because You're Worried About Cancer.
Sign up for the Slatest to get the most insightful analysis, criticism, and advice out there, delivered to your inbox daily.
When you get an X-ray, it can be a bit unnerving: The radiology technician puts on a lead gown or ducks behind a barrier before running the scan. Which might make you wonder: If the radiation is so bad that they need protective equipment, why are you getting exposed to it?
There is a simple answer. You are getting only one dose of radiation from your scan, but the tech might be doing a couple dozen of those same scans a day. Radiation doses add up, so people who regularly work with the machines need protection so that they can lower their risk of getting cancer from their job.
But a new study questions whether scans—specifically computed tomography scans —are causing more cancer than we thought, specifically among the patients who receive them. CTs are given to people in emergency departments who have had serious trauma or are having a suspected stroke . Other CTs are used to identify and treat cancer, rule out serious neurological problems, and more. According to the news stories , CT scans could be causing a remarkably high number of cancers each year and could eventually be responsible for 1 in 20 cancers in the United States. Yikes.
There are, however, some major issues with the media reports. Although CT scans can and sometimes do cause cancer in patients, the reality is that you probably aren't in much danger when you go in for a recommended scan.
The study itself is basically a modeling exercise. The authors took an estimate of the number of CT scans done every year in the US, based on research from 2023. Next, they estimated the dose of radiation that each CT would give the patient, based on the body part that was exposed in the scan and evidence from studies of radiation dosage. Finally, they used an online calculator, called RadRAT, to estimate the number of cancers that the estimated radiation dosage would cause over the lifetimes of all the people getting the scans.
The authors found that in 2023 there were 93 million CTs done on 62 million patients. The total estimated lifetime number of cancers caused by the radiation from these CTs in their model came out to about 100,000. So, for the 62 million people who got CTs over the course of 2023, there would eventually be an additional 100,000 cancers.
This sounds like a big number, but is it, really, compared with the overall amount of cancer in the population? If you ask the authors, they say that cancer from CT scans could eventually add up to 5 percent of all cancers diagnosed in the United States. That seems pretty significant! But this is assuming that the number of yearly cancers stays the same between now and 2075 (about the time the kids included in the study would be developing cancer from the CT scans they received in 2023). That's very unlikely—the total number of cancers annually will probably increase by then due to population growth alone. In any case, it's a figure based on an iffy assumption.
I think a better way to look at this is by considering that about 40 percent of people will get cancer in their lifetime. If 62 million people are getting CTs, we'd estimate that 25 million of them would get cancer without the additional CT risk. The extra cancers from the CTs therefore account for an additional 0.4 percent, or one extra lifetime cancer per 250 people.
In context, that number is not scary at all. The authors of the study compare CT scans with alcohol consumption, which is another risk factor for cancer. But people rarely get CT scans for the fun of it. One of the most common reasons for having a CT in the first place is because you either probably or definitely have cancer already, and the imaging can help doctors know exactly what they're dealing with so they can do their best to get rid of it. Most people are likely fine with accepting a 1-in-250 risk of getting an additional cancer if it means that their current cancer can be treated appropriately rather than having a surgeon simply guess where the tumor is located. No serious medical treatment is risk-free; we get treatment anyway because the downsides of not doing anything are worse. In this case, they can be much worse.
The other issue with the estimate of 100,000 cancers is that it's almost certainly too high. RadRAT, the calculator that the researchers used, is designed to estimate the lifetime risk of cancer for people who are generally healthy . Specifically, when calculating lifetime risk, the tool assumes that people who are exposed to radiation are going to live as long as the average person of that age and sex who was alive in the US in 2019.
But we know that CTs are not generally given to otherwise healthy people. If you're not in great shape to start, you're less likely to live as long as the average person, and you have less time to develop cancer in the first place. It's quite likely that the lifetime number of cancers is dramatically overestimated by the published paper.
There is a useful bit of truth in the lesson news stories are taking from the paper: You shouldn't get a CT scan unless you actually need one. We do have a problem with doing too many CT scans. Studies looking at whether CT scans were appropriately ordered suggest that somewhere between 8 and 36 percent of them are not needed. Given that these scans definitely do cause cancer—albeit quite rarely—we should instead be trying to ensure that we perform them only when absolutely necessary.
That being said, the idea that 1 in 20 cancers is—or will be— caused by CTs is very unlikely to be true. The actual figure is presumably quite a bit lower. And while we should be careful to use those high-radiation scans only when necessary, in most cases if a doctor tells you to get a CT, you should probably do it.
Sign up for Slate's evening newsletter.
