Radiation dose chart...

[erictabellion]

Hi,

I just saw this excellent radiation dose chart: http://xkcd.com/radiation/

This is a very good chart!

I was shocked to see that "radiation workers" limit is as much as half the dose clearly linked to increased cancer risk (I would have though it was much less). Does this mean radiation workers have a 0.5 probability of increased cancer risk ?

Cheers,
Eric

[Malle]

In general, no. Without knowing more about the data that has been gathered, it's really hard to say.

What I suspect has been done is a study to determine the lowest dose that leads to a statistically significant increase in cancer rates (in this case, 100 mSv) with some certain standard deviation and then used a safety factor of 2 for radiation workers (thus giving the 50 mSv limit). As far as I know this is acceptable practice, assuming that the studies have at least a resolution (in mSv in this case) that is not greater than the difference given from the safety factor (so the studies here should have a resolution of at least 50 mSv).

[Ulc]

Another very good chart can be found here

http://upload.wikimedia.org/wikipedia/en/f/fd/Fukushima_map.png

As can be seen, the amount of radiation that the fission plant workers are exposed to do carry a slight increase in cancer risk.. but really, 1:20000 compared to normal cancer rates of 1:25 doesn't matter in the least.

Basically, it's not enough to know when there can be observed increased risk. It's also important to know how large a difference in contained in that "increased risk" statement.

[Tass]

Does this mean radiation workers have a 0.5 probability of increased cancer risk ?

How could you have a certain probability of increased risk? A risk is a probability. 0.5 probability of increased risk would just be the risk increased by half as much.

[Soralin]

Another very good chart can be found here

http://upload.wikimedia.org/wikipedia/en/f/fd/Fukushima_map.png

As can be seen, the amount of radiation that the fission plant workers are exposed to do carry a slight increase in cancer risk.. but really, 1:20000 compared to normal cancer rates of 1:25 doesn't matter in the least.

Basically, it's not enough to know when there can be observed increased risk. It's also important to know how large a difference in contained in that "increased risk" statement.

That chart doesn't seem particularly good (at least the bar on the left), it has it labeled in mSv/hour, but then has places marked on it for things like radiation poisoning or cancer risk. Which doesn't make any sense, unless you know how long the exposure is. Is it that amount for an hour, for a second, for a year, indefinitely? It seems like those things match up with an hour of exposure at that level, although the other chart listed has 100mSv as lowest one-year dose clearly linked to increased cancer risk. If that's the case, then is the bar there just assuming the linear no-threshold model? http://en.wikipedia.org/wiki/Linear_no-threshold_model It seems like it is, since 1/100th the radiation shown there is listed as having 1/100th the cancer risk. Although if that's not the case, then the same dose of radiation over a shorter period of time would have an increased risk, and less of a risk if it happened over a long period of time.

[Zamfir]

As people above said, 100 mSv is not the level where you certainly get cancer, it's the lowest level where we have statistically significant evidence that it increases cancer .

Under normal situations, radiation workers never get to 50 mSv in a year. It's more the limit after which the government starts serious action against the plant, if they expose a worker to even more.

In most countries, 20mSv or so is the normal yearly limit, and you should give good reasons why you expose someone to the full extent of it (like a repair that cannot be done any other way). There is a principle called ALARA, 'as low as reasonably achievable', meaning you cannot expose people to a dose just because it falls within the limits.

In parctice, plants tend to keep workers below 5mSv a year, often aiming more to keep everyone below 2 or so, which is comparable to background radiation. Under normal circumstances, that's doable.

[Robo Connery]

The idea behind the 100mSv number being the lowest dose in which we can observe an increase in risk is either that; below 100mSv the risk is so small it is difficult to tell statistically (say it increases the chance by 1 in 100,000 and if you only have a million people studied then a change of 10 extra cases isn't noticeable) or, as most people believe, the body can easily deal with doses <~ 100 mSv without any increase in risk of cancer so, it isn't just a statistical thing, there is actually no harmful effects from a dose that small.

Either way the radiation workers have nothing to fear.

[KingofMadCows]

Smoking a pack of cigarettes a day will get you exposed to about 8 to 10 mSv per year.

[Moose Hole]

Smoking a pack of cigarettes a day will get you exposed to about 8 to 10 mSv per year.

So you're saying radiation is healthy?

[Tass]

Smoking a pack of cigarettes a day will get you exposed to about 8 to 10 mSv per year.

...in addition to the much worse effects of the nonradioactive carcinogens and toxins.

[Waterice man]

I've just seen this data used (probably) by no less than the BBC

http://www.bbc.co.uk/news/magazine-15288975

[Minerva]

If every single cumulative nanosievert of ionising radiation dose really was responsible for adverse health effects and increased morbidity and mortality, as per the "linear non-threshold hypothesis" that anti-nuclearists love to carry around etched on their stone tablets from the mount, then we would be able to see that increased morbidity and mortality - through careful, well-controlled, quantitative scientific epidemiology - in every single medical radiologist and nuclear medicine technologist, every experimental physicist and every biomedical researcher who uses radionuclides. We would see it in nuclear energy workers who work around ionising radiation.

We would see the radiation-related morbidity and mortality from Chernobyl fulfuling the linear-non-threshold based predictions of the anti-nuclear activists, instead of the real-world epidemiological outcomes which turned out, to the surprise of many scientists, to be nowhere as bad as some predicted.

We would see it epidemiologically in every person who lives in Ramsar, or Kerala, or Denver or Cornwall, compared to populations living in other places that do not have those unusually high natural background ionising radiation doses.

And of course we would see increased radiation-related morbidity and mortality in all commercial airline pilots and crews, who have the greatest occupational ionising radiation dose of any people working in any occupation.

But we do not see any of these things, the predictions of what we would empirically see.
That's how we do science. You make the hypothesis, you determine what the consequences will be that you will be able to see empirically, and then you go and look at the empirical data.

It doesn't matter who made the hypothesis, or what her name is, or how smart they are or how highly regarded they are - if it doesn't agree with experiment then it's wrong.

The hypothesis that every single cumulative nanosievert of ionising radiation dose is responsible for adverse health effects and increased morbidity and mortality has been a hypothesis for the last 50 years, and it has remained a hypothesis without any real body of empirical evidence coming along to support it.

The entire science of radiotherapy is based on known radiobiological behavior - namely that fractionated doses do not have a linearly cumulative response and that there dose thresholds below which tissue does not experience any real morbidity in practice.