5 ms·

Wait, how did you arrive at that conclusion? All nuclear accidents so far have resulted in very low doses to the average individual. Of course this doesn't appl

by simloo 6y ago

Wait, how did you arrive at that conclusion? All nuclear accidents so far have resulted in very low doses to the average individual. Of course this doesn't apply to plant workers in the direct vicinity of e.g. the Chernobyl accident, but since those are very few compared to the population of a country, the number of estimated cancer cases should go down.

Retric 6y ago
It’s the linear model which suggests those low doses are mostly safe, it may be a more accurate model means average adults are at say 1/2 the risk of linear models. But, say children are at 10x risk, babies are at 100x risk, and Fedus are at 1,000x higher risk than those linear models suggest. Thus because of a high risk to a significant population it’s even more dangerous. It’s very easy to find some way a specific model is wrong in one direction, but hard to say if that’s the only adjustment you need to make. Ex: Counting better cancer treatments making nuclear safer while ignoring longer lifespans making it more dangerous.
- simloo 6y ago
  That's true, but I think most other models assume there's a lowest dosage which is totally safe? For very low doses, even if LNT predicts a very low rate, it still a lot of cancer cases for a population of several million. Even if there would be very vulnerable groups, it's hard to imagine a model that predicts more cases than LNT for very low doses.
  - Retric 6y ago
    There are a huge range of models, it’s not clear which ones are correct. Nuclear proponents tend to push threshold models suggesting there are safe doses. On the other hand: “Approximately 38.4% of men and women will be diagnosed with cancer at some point during their lifetimes (based on 2013–2015 data).“ That makes it very hard to validate small changes and any reasonable study size is going to fall below the noise floor. In other words people pushing those models lack any direct evidence to support them.
    - simloo 6y ago
      Ok, fair enough.
pfdietz 6y ago
Let's consider the risk of cancer from radiation as a function of the radiation dose. This is some function, call it f. If I am receiving a dose r, and then get an additional does dr, then the additional risk is f(r+dr)-f(r). Or, this is f'(r)dr, where f' is the slope of f at r. This is simple calculus. For LNT, f(r) is assumed to be a linear function of r, so the slope is the same everywhere. In this case, we don't need to know the actual dose someone gets, just the increment; an increment of dr in the dose causes the same risk of cancer. This allows the risk to a population to just be computed from the total population incremental dose. But if f is NOT linear, the slope will be higher in some places than LNT would imply (this again is a simple theorem from calculus). For example, suppose f'(r) is higher at very low dose, but less than LNT above some threshold dose level (that is, the curve us concave downward). In this case, the risk of cancer from some extra dose dr will be higher down in that low dose area. Is this plausible? I don't think it's ruled out. Radiation apologists would have us believe in radiation hormesis, in which radiation induces repair mechanisms. Under that hypothesis, the slope of the curve below a threshold at which this induction occurs could be steeper than LNT would imply.
- roenxi 6y ago
  > But if f is NOT linear, the slope will be higher in some places than LNT would imply (this again is a simple theorem from calculus). What does that even mean? That is junk maths, I don't believe there is such a theory. A line can have any gradient and I can always come up with an arbitrary curve that has a lower or equal gradient at every point that matches the line. A model that matches LNT exactly to some threshold then the risk drops to 0, for example. Not higher than the LNT anywhere in any negative sense.
  - Retric 6y ago
    Suppose a vending machine gives 20 soda for 20$. If someone walks up and inserts 1$ and can get 0 soda, but if they insert an extra 19$ they get 20 soda. If that’s true then for some amount between 1$ and 20$ inserting an extra 1$ must let you get more than 1 soda in order for it to be handing out 20 soda at 20$. PS: Back to cancer, if it’s 19 cancer at 20x then you can keep a 1:1 relationship at 2+x and 0 cancer at 1x. Then again if you can’t tell if it’s 19 cancer or 20 cancer then it might just be 21 cancer.

by simloo 6y ago

Retric 6y ago

It’s the linear model which suggests those low doses are mostly safe, it may be a more accurate model means average adults are at say 1/2 the risk of linear models. But, say children are at 10x risk, babies are at 100x risk, and Fedus are at 1,000x higher risk than those linear models suggest. Thus because of a high risk to a significant population it’s even more dangerous. It’s very easy to find some way a specific model is wrong in one direction, but hard to say if that’s the only adjustment you need to make. Ex: Counting better cancer treatments making nuclear safer while ignoring longer lifespans making it more dangerous.

simloo 6y ago
That's true, but I think most other models assume there's a lowest dosage which is totally safe? For very low doses, even if LNT predicts a very low rate, it still a lot of cancer cases for a population of several million. Even if there would be very vulnerable groups, it's hard to imagine a model that predicts more cases than LNT for very low doses.
- Retric 6y ago
  There are a huge range of models, it’s not clear which ones are correct. Nuclear proponents tend to push threshold models suggesting there are safe doses. On the other hand: “Approximately 38.4% of men and women will be diagnosed with cancer at some point during their lifetimes (based on 2013–2015 data).“ That makes it very hard to validate small changes and any reasonable study size is going to fall below the noise floor. In other words people pushing those models lack any direct evidence to support them.
  - simloo 6y ago
    Ok, fair enough.

pfdietz 6y ago

Let's consider the risk of cancer from radiation as a function of the radiation dose. This is some function, call it f. If I am receiving a dose r, and then get an additional does dr, then the additional risk is f(r+dr)-f(r). Or, this is f'(r)dr, where f' is the slope of f at r. This is simple calculus. For LNT, f(r) is assumed to be a linear function of r, so the slope is the same everywhere. In this case, we don't need to know the actual dose someone gets, just the increment; an increment of dr in the dose causes the same risk of cancer. This allows the risk to a population to just be computed from the total population incremental dose. But if f is NOT linear, the slope will be higher in some places than LNT would imply (this again is a simple theorem from calculus). For example, suppose f'(r) is higher at very low dose, but less than LNT above some threshold dose level (that is, the curve us concave downward). In this case, the risk of cancer from some extra dose dr will be higher down in that low dose area. Is this plausible? I don't think it's ruled out. Radiation apologists would have us believe in radiation hormesis, in which radiation induces repair mechanisms. Under that hypothesis, the slope of the curve below a threshold at which this induction occurs could be steeper than LNT would imply.

roenxi 6y ago
> But if f is NOT linear, the slope will be higher in some places than LNT would imply (this again is a simple theorem from calculus). What does that even mean? That is junk maths, I don't believe there is such a theory. A line can have any gradient and I can always come up with an arbitrary curve that has a lower or equal gradient at every point that matches the line. A model that matches LNT exactly to some threshold then the risk drops to 0, for example. Not higher than the LNT anywhere in any negative sense.
- Retric 6y ago
  Suppose a vending machine gives 20 soda for 20$. If someone walks up and inserts 1$ and can get 0 soda, but if they insert an extra 19$ they get 20 soda. If that’s true then for some amount between 1$ and 20$ inserting an extra 1$ must let you get more than 1 soda in order for it to be handing out 20 soda at 20$. PS: Back to cancer, if it’s 19 cancer at 20x then you can keep a 1:1 relationship at 2+x and 0 cancer at 1x. Then again if you can’t tell if it’s 19 cancer or 20 cancer then it might just be 21 cancer.