Cumulative backscatter X-ray risk

January 11, 2010 2:39 PM

How many additional people would die per year because of exposure to radiation if every flyer in the US was examined with a backscatter X-ray machine?

I am aware that the additional radiation exposure in such machines is extremely small and not relevant for individual flyers ("equivalent of two minutes of flying on an airplane").

However, as stated by a doctor in this Router's article, "If very large numbers of people would be exposed to a small risk, then you've got a population problem." How big would this problem be?

Even if backscatter X-rays give 1/1000 the dose of a chest X-ray, I wonder if there could still be a significant cumulative effect. In particular, how would this number of deaths compare to the number of deaths from airplane related terrorism that it might hypothetically prevent?

posted by zxcv to health & fitness (12 answers total)

Well, for comparison, how many people have died in airplane-related terrorism incidents since 2001?
posted by spaceman_spiff at 2:43 PM on January 11, 2010

The rate of cancer death from a single x-ray has been estimated at about 8 in 10,000.

The rate of deaths due to terrorism on planes is currently 1 in 10.4 million.

There is no way to know how many deaths x-ray could prevent unless you know future rates of terrorism and also what percentage of attempted terrorism is being foiled by current methods that might be scrapped in favor of this new one. But I'm going to go out on a limb and say that it's highly unlikely that x-rays at airports would save more than they kill.
posted by decathecting at 2:44 PM on January 11, 2010

Sorry, I read the x-ray numbers wrong. Looks like an x-ray is 3+ times the risk I originally indicated. Point still stands.
posted by decathecting at 2:47 PM on January 11, 2010

You might find this NY Times article helpful. Apparently there is some disagreement on this question. But, I found this quote interesting:

According to a blog published by the Transportation Security Administration, the radiation dose from the scanner is about the same amount as an average American receives from natural background sources in four minutes on the ground.
posted by found missing at 2:54 PM on January 11, 2010

From Wikipedia
The radiation dose from a backskatter xray machine (designed to ANSI N43.17) is to be less than 25 micro Rontgen per examination. (0.25 micro Sievert or 2.5x10^-7 Sievert) The radiation dose for a chest xray according to the 8 in 10,000 link from decathecting is 3.2x10^-5 Sievert. So it's quite a bit less than even the mildest medical xray.

Now, another line in the wikipedia article, even more relevant:
Twenty-five micro Rem is equal to the amount of background radiation every human is exposed to (from the air and soil) at sea level every 1.5 hours and is also equal to the radiation exposure from cosmic rays when traveling in an airplane at altitude for 2 minutes.

Which is saying that the additional radiation you are exposed to from the backscatter xray is negligible.

Don't forget, a backskatter xray does not need to penetrate your body (like a medical device). The Wikipedia article has more details.
posted by defcom1 at 3:03 PM on January 11, 2010 [1 favorite]

According to a blog published by the Transportation Security Administration, the radiation dose from the scanner is about the same amount as an average American receives from natural background sources in four minutes on the ground..

This is terribly bad scientific thinking, ignorant at best and designed to confuse people at worst. Probably they're right and there's no real danger (I am in no way qualified to assess that), but so much of chemistry and biology relies on threshold levels below which nothing happens at all, and above which something good/bad starts happening. 4 minutes worth of radiation at once is in no way comparable to 4 minutes of lower level radiation.
posted by Dr.Enormous at 3:05 PM on January 11, 2010

I'm not really sure how traceable it would be anyway. Even if someone did die as a result of a backscatter scan, it wouldn't be so immediate that a coroner could say "the cause of death was clearly that airport security machine." On the contrary, it would probably manifest itself as cancer over several months or years and be attributed to a number of different factors. A backscatter machine walked through six months ago wouldn't even be on the coroner's radar as a possible trigger.

That said, as with all cancers, I don't know that you can ever pinpoint one event as the "reason" for cancer. Seems like it's always the cumulative result of a bunch of tiny things, any of which taken individually would be perfectly harmless. The point being that there's really no practical way to measure the harm of a backscatter X-ray machine. The harm may exist but it's unquantifiable.
posted by The Winsome Parker Lewis at 3:05 PM on January 11, 2010

There is an unspoken assumption in your question, which is the linear no-threshold model of radiation exposure. This is the theory that there is no minimum safe dose of radiation. It is not without controversy and there are other theories, such as the threshold model and radiation hormesis.

But, presuming the LNT model to be correct, we can make some calculations. (The unit here, the sievert is intended to approximate the biological effects of radiation exposure rather than just the energy involved). The linked article says the backscatter machines produce an exposure of .1µSv. The average natural radiation that a person receives just from living is about 10µSv per day. You get another 1.8µSv/day from manmade sources. So if you went through a backscatter machine once a day it would increase your radiation-induced cancer risk by about .85%.

But of course people don't go through such machines every day. There were about 704 million passenger emplanements in the US last year, which broadly suggests that Americans fly an average of once per year, round trip, for two exposures. That means an annual exposure increase of .0046%, more or less.

But of course, as I said, that's not a .0046% increase in the total cancer rate but rather the radiation-induced cancer rate. It's not well known exactly how many cancers are attributable to radiation, whether natural or artificial, so it's pretty hard to answer your question exactly. Nonetheless, let's press on with some estimates.

Since the radiation does not penetrate the skin, presumably the most likely cancer to be caused by this machine would be skin cancer, so let's make the (not totally absurd) assumption that all skin cancers are the result of radiation. About 11,500 people died of skin cancer in 2009 [pdf]. A .0046% increase would be about half a person per year. So, pretty much completely negligible.

So, if this security measure saves about one life per year on average, it will be effective in terms of lives (though not necessarily dollars). As others have pointed out, however, that's an open question.
posted by jedicus at 3:18 PM on January 11, 2010 [7 favorites]

Looks like decathecting and found missing's answers contain the key information:

Odds of traveler death due to terrorism on a flight (ignoring victims on the ground):
1 in 10.4 million

A (high) estimate of the number of additional deaths according to an expert in the New York Times:
1 in 100 million

This confirms jedicus's calculation that significantly more people die from terrorism than would from X-ray backscattering.

Thanks for the help, guys!
posted by zxcv at 3:23 PM on January 11, 2010

Since jedicus already covered the no-threshold problem, I'll skip that, but it's important to consider (some people consider low dose radiation to actually be helpful).

According to physics, you suffer a 1 in 1,000,000 chance of getting cancer if you're exposed to 10 mrem of radiation, or 2,000 backscatter x-rays. So for every 2 billion passenger-flights, there'd be one additional case of cancer. This is in line with jedicus' estimate.

The good news is that since you get an additional 5 mrem of radiation in a coast-to-coast flight, we'll prevent cancer deaths if 1 in every 1,000 passengers is deterred from flying because they don't want to be seen naked by a security screener.
posted by 0xFCAF at 3:25 PM on January 11, 2010 [1 favorite]

So it's quite a bit less than even the mildest medical xray.

And I must say, every orthopedic surgeon I've seen loves to do x-rays. Every check-up if possible.
posted by smackfu at 4:54 PM on January 11, 2010

The good news is that since you get an additional 5 mrem of radiation in a coast-to-coast flight, we'll prevent cancer deaths if 1 in every 1,000 passengers is deterred from flying because they don't want to be seen naked by a security screener.

Actually, this brings up another good point. If you're interested in measuring risk broadly, you have to take into account the fact that some people won't fly if they have to undergo an x-ray. Whether they're right or wrong about the dangers of radiation or squeamish about the idea that someone might see their naughty bits, some people won't fly. Many of them will drive instead. And the average number of fatalities per mile traveled is 37 times higher for driving than flying.
posted by decathecting at 8:08 PM on January 11, 2010