4 ms·

These special studies are good for a scientific consumption, but they don't do much to make the results comprehensible to the general public. In this specific

by maratc 3y ago

These special studies are good for a scientific consumption, but they don't do much to make the results comprehensible to the general public. In this specific case, what's missing is the bigger perspective, like this: "If we look at the 100,000 average dead people, we'll find that 30,000 of them died from cancer. Now, we are looking at 100,000 dead low dose nuclear workers, and we have previously thought that 30,010 of these would have died from cancer, but now we've found that number to be closer to 30,020." This really puts "2x worse" into a proper context, which is: there are low-risk occupational hazards in nuclear industry.

csours 3y ago
Other numbers are more meaningful than "died of cancer" - Total time with cancer - including time without symptoms - Total time affected by cancer as a disease - time with symptoms - Total lifespan lost - how young did a person die of the disease There's probably no way to really know #1, #2 is hard to gather information on, #3 is more do-able.
- landemva 3y ago
  Why are 1 and 2 more meaningful?
- maratc 3y ago
  The relevant measurement is called QALY, or quality adjusted life years. It is meant to be a meaningful representation of a disease burden. As an example, thyroid cancer is hardly curable, but a common remedy is to remove the thyroid and to rely on medications going forward. Somebody in this condition can have a long quality life otherwise. Pancreatic cancer, on the other way, has very low survival rate, so it cuts a lot of QALYs out of a person’s life. It would be great if the authors would quantify their findings in QALY, as in e.g. “while we have previously believed that the low dose radiation results in a loss of x days (QALY wise), we now believe that number to be 2x days”.
omgJustTest 3y ago
Thank you for making this point, I made it earlier in the day in a less convincing way... 2x is not even claimed in the article.
godelski 3y ago
There's also some interesting things about the paper that set off some flags for me. I've studied radiation extensively, but not medicine fwiw. - I'm always wary of anyone that will do something like write 2000 keV instead of 2MeV. They also are constantly saying "photon radiation" instead of "gamma," which just reads weird. Neither of these are "incorrect" though. - They seem to not be accounting for neutrons which is quite a common source if we're talking about reactor workers or Hiroshima. The neutrons get absorbed into the atoms and then it sheds energy through gamma radiation, but a new neutron results in potentially a different chemistry in addition to the ionization energies. Actually reactors depend on this: 235U + 1n -> 236U -> 89Kr + 114Ba + 3n + 2gamma. You just don't often see the 236U discussed, but it is unstable and causes the splitting. - They use Japanese atomic bomb survivors as comparisons but that's not great and there are many studies looking at radiation workers (both these get neutron doses). There's two big aspects here that matter. First, acute dosage is very different from chronic exposure, and atomic bomb dosages are going to be acute (not just Japanese, but also soldiers). Most workers are going to be chronic, but some may be acute which you need to account for. They don't seem to. There are also many studies on places in the world where the natural background radiation is over the 100mGy that they are saying doubles the cancer rate, but people from these places don't show increased cancer rates (this is mostly radon too, so alpha radiation, and of course this is chronic and not acute). I don't feel qualified to invalidate the work btw, but these seemed points of note. I know there are some users here with some more expertise, especially in radiation physics, so I was hoping someone could help me understand.
DoreenMichele 3y ago
Thank you for this. This helped me enormously.
zosima 3y ago
There is no increased mortality in the paper. The paper only looks at risk of death from solid tumors. I'd be surprised if the authors haven't done the calculation of total mortality and decided that there is no discernible radiation dose-dependent effect on total mortality or total QALY. In a better world, that would also be a very good paper, but it isn't in ours. That's why they are looking for solid tumor mortality.
- maratc 3y ago
  “Out of 100,000 living people, x will die from solid tumors in one year; out of 100,000 nuclear workers we believe x+2y will. This differs from the x+1y figure that was the previous estimation.” QALY would be best in any case.

by maratc 3y ago

csours 3y ago

Other numbers are more meaningful than "died of cancer" - Total time with cancer - including time without symptoms - Total time affected by cancer as a disease - time with symptoms - Total lifespan lost - how young did a person die of the disease There's probably no way to really know #1, #2 is hard to gather information on, #3 is more do-able.

landemva 3y ago
Why are 1 and 2 more meaningful?
maratc 3y ago
The relevant measurement is called QALY, or quality adjusted life years. It is meant to be a meaningful representation of a disease burden. As an example, thyroid cancer is hardly curable, but a common remedy is to remove the thyroid and to rely on medications going forward. Somebody in this condition can have a long quality life otherwise. Pancreatic cancer, on the other way, has very low survival rate, so it cuts a lot of QALYs out of a person’s life. It would be great if the authors would quantify their findings in QALY, as in e.g. “while we have previously believed that the low dose radiation results in a loss of x days (QALY wise), we now believe that number to be 2x days”.

omgJustTest 3y ago

Thank you for making this point, I made it earlier in the day in a less convincing way... 2x is not even claimed in the article.

godelski 3y ago

There's also some interesting things about the paper that set off some flags for me. I've studied radiation extensively, but not medicine fwiw. - I'm always wary of anyone that will do something like write 2000 keV instead of 2MeV. They also are constantly saying "photon radiation" instead of "gamma," which just reads weird. Neither of these are "incorrect" though. - They seem to not be accounting for neutrons which is quite a common source if we're talking about reactor workers or Hiroshima. The neutrons get absorbed into the atoms and then it sheds energy through gamma radiation, but a new neutron results in potentially a different chemistry in addition to the ionization energies. Actually reactors depend on this: 235U + 1n -> 236U -> 89Kr + 114Ba + 3n + 2gamma. You just don't often see the 236U discussed, but it is unstable and causes the splitting. - They use Japanese atomic bomb survivors as comparisons but that's not great and there are many studies looking at radiation workers (both these get neutron doses). There's two big aspects here that matter. First, acute dosage is very different from chronic exposure, and atomic bomb dosages are going to be acute (not just Japanese, but also soldiers). Most workers are going to be chronic, but some may be acute which you need to account for. They don't seem to. There are also many studies on places in the world where the natural background radiation is over the 100mGy that they are saying doubles the cancer rate, but people from these places don't show increased cancer rates (this is mostly radon too, so alpha radiation, and of course this is chronic and not acute). I don't feel qualified to invalidate the work btw, but these seemed points of note. I know there are some users here with some more expertise, especially in radiation physics, so I was hoping someone could help me understand.

DoreenMichele 3y ago

Thank you for this. This helped me enormously.

zosima 3y ago

There is no increased mortality in the paper. The paper only looks at risk of death from solid tumors. I'd be surprised if the authors haven't done the calculation of total mortality and decided that there is no discernible radiation dose-dependent effect on total mortality or total QALY. In a better world, that would also be a very good paper, but it isn't in ours. That's why they are looking for solid tumor mortality.

maratc 3y ago
“Out of 100,000 living people, x will die from solid tumors in one year; out of 100,000 nuclear workers we believe x+2y will. This differs from the x+1y figure that was the previous estimation.” QALY would be best in any case.