4 ms·

> Perhaps we'll find a mass grave somewhere full of radioactive bodies of dead workers but barring that, most data based consensus puts the total death toll fro

> Perhaps we'll find a mass grave somewhere full of radioactive bodies of dead workers but barring that, most data based consensus puts the total death toll from all radiation effects at less than 1,000 and usually less than 100. The problem is that to this day we struggle to fully understand the effects of long-term low-level radiation exposure because figuring those out usually requires long-term epidemiological studies, which are expensive and complex [0] [0] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2757021/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2757021/

ChuckMcM 7y ago
Okay. I'm trying to understand the "low level radiation" part though. What is disappointing about the study you reference is that they don't characterize what the level of radiation is compared to normal background radiation, they don't account for changes in population or diet, and they don't account for changes in industry. As the study itself states: "These conclusions are regrettable because low statistical significance only means that chance has not been excluded as an explanation, assuming no bias and no real effect. In more detail, p values -that is, the probabilities that observed effects may be due to chance - are affected by both the magnitude of the effect and the size of the study. This means the results of statistical tests must be interpreted with caution [26]." Meta studies based on statistical analysis is inherently challenged. This one is no better (and no worse) than the others.
- freeflight 7y ago
  It's a commentary on the study, it also states: "Whatever the final explanation for the increases, the KiKK study and its implications raise many questions, including whether vulnerable people, such as pregnant women and women of child-bearing age, should be advised on possible risks of living near nuclear power stations." and: "It is recommended that US regulatory agencies should establish a KiKK-style epidemiological study of cancer incidences near all US nuclear power stations with precise distances being measured between cancer cases and nuclear reactors." The point here is that our understanding of these impacts is obviously not as complete as many people like to pretend.
  - ChuckMcM 7y ago
    I don't disagree, and that same text is also in various meta studies about Cell phone towers causing cancer. And it is the problem in general with statistics based meta studies which where lampooned in XKCD[1] and lamented by the American Council on Science and Health[2]. The summary bit might be "These studies point to areas that should be looked at but offer neither hypothesis or prediction and thus not 'science' so much as they are an early investigative technique." If nuclear plants caused childhood cancers, then they ALL would cause childhood cancers, so a scientific study would say "This mechanism, when present, has this effect. Here is our experiment to prove that, here are the results of the proof. We predict that if you look anywhere else where this mechanism is present you will see the same results." If low level radiation did cause childhood leukemia then there would be clusters in Arvada Colorado and Moab Utah. But there aren't. So why not? When the science is done, and the other variables eliminated, to date there is no study that links low level ionizing radiation to any harmful effect on humans. Even using that term is difficult because there is alpha, beta, and gamma radiation based on electrons, protons, and neutrons. Only Gamma radiation has any sort of penetrating capability at all. So without the science, this study means nothing. Now if they can follow up with core samples, air samples, water samples, and food products to see if there are any nuclear fission byproducts getting into the area around the plant that would be good science. [1] https://xkcd.com/882/ https://xkcd.com/882/ [2] https://www.acsh.org/news/2018/06/29/problem-p-values-13130 https://www.acsh.org/news/2018/06/29/problem-p-values-13130
    - effie 7y ago
      > there is no study that links low level ionizing radiation to any harmful effect on humans This does not mean much. The direct destruction effect of strong radiation on cells is cumulative, the longer you're exposed, the greater harm done. Given the physics and biology knowledge, our best bet so far is that effect of weak but extraordinary radiation is cumulative as well. It will increase probability of cancer and leukaemia. Alpha radiation is lethal too if you get its source into your body.
      - vonmoltke 7y ago
arcticbull 7y ago
Not really; there’s always constant low level radiation exposure in our day to day from background radiation. It goes up tremendously when you eat a banana a day. It goes up another few orders of magnitude when you get on a plane. We’ve got a huge corpus of constantly irradiated people (flight attendants and pilots). We’ve got a pretty good idea of how this works. [1] https://xkcd.com/radiation/ https://xkcd.com/radiation/
- effie 7y ago
  You miss the point. Studies on low level radiation effects study, for example, deviations from the linear model, but are so far inconclusive, because it is hard to prescribe and control the level of radiation over such a long time. It could be done in lab, but who would subject themselves willingly to regular blasts of radiation? Radiation intensity in planes varies a lot, depending on the height, time of flight, geographical location, etc., so little can be inferred. It would take decades and lots of volunteers willing to get harmed in the name of science to do such studies reliably. Do you know of any such study?

ChuckMcM 7y ago

Okay. I'm trying to understand the "low level radiation" part though. What is disappointing about the study you reference is that they don't characterize what the level of radiation is compared to normal background radiation, they don't account for changes in population or diet, and they don't account for changes in industry. As the study itself states: "These conclusions are regrettable because low statistical significance only means that chance has not been excluded as an explanation, assuming no bias and no real effect. In more detail, p values -that is, the probabilities that observed effects may be due to chance - are affected by both the magnitude of the effect and the size of the study. This means the results of statistical tests must be interpreted with caution [26]." Meta studies based on statistical analysis is inherently challenged. This one is no better (and no worse) than the others.

freeflight 7y ago
It's a commentary on the study, it also states: "Whatever the final explanation for the increases, the KiKK study and its implications raise many questions, including whether vulnerable people, such as pregnant women and women of child-bearing age, should be advised on possible risks of living near nuclear power stations." and: "It is recommended that US regulatory agencies should establish a KiKK-style epidemiological study of cancer incidences near all US nuclear power stations with precise distances being measured between cancer cases and nuclear reactors." The point here is that our understanding of these impacts is obviously not as complete as many people like to pretend.
- ChuckMcM 7y ago
  I don't disagree, and that same text is also in various meta studies about Cell phone towers causing cancer. And it is the problem in general with statistics based meta studies which where lampooned in XKCD[1] and lamented by the American Council on Science and Health[2]. The summary bit might be "These studies point to areas that should be looked at but offer neither hypothesis or prediction and thus not 'science' so much as they are an early investigative technique." If nuclear plants caused childhood cancers, then they ALL would cause childhood cancers, so a scientific study would say "This mechanism, when present, has this effect. Here is our experiment to prove that, here are the results of the proof. We predict that if you look anywhere else where this mechanism is present you will see the same results." If low level radiation did cause childhood leukemia then there would be clusters in Arvada Colorado and Moab Utah. But there aren't. So why not? When the science is done, and the other variables eliminated, to date there is no study that links low level ionizing radiation to any harmful effect on humans. Even using that term is difficult because there is alpha, beta, and gamma radiation based on electrons, protons, and neutrons. Only Gamma radiation has any sort of penetrating capability at all. So without the science, this study means nothing. Now if they can follow up with core samples, air samples, water samples, and food products to see if there are any nuclear fission byproducts getting into the area around the plant that would be good science. [1] https://xkcd.com/882/ https://xkcd.com/882/ [2] https://www.acsh.org/news/2018/06/29/problem-p-values-13130 https://www.acsh.org/news/2018/06/29/problem-p-values-13130
  - effie 7y ago
    > there is no study that links low level ionizing radiation to any harmful effect on humans This does not mean much. The direct destruction effect of strong radiation on cells is cumulative, the longer you're exposed, the greater harm done. Given the physics and biology knowledge, our best bet so far is that effect of weak but extraordinary radiation is cumulative as well. It will increase probability of cancer and leukaemia. Alpha radiation is lethal too if you get its source into your body.
    - vonmoltke 7y ago

arcticbull 7y ago

Not really; there’s always constant low level radiation exposure in our day to day from background radiation. It goes up tremendously when you eat a banana a day. It goes up another few orders of magnitude when you get on a plane. We’ve got a huge corpus of constantly irradiated people (flight attendants and pilots). We’ve got a pretty good idea of how this works. [1] https://xkcd.com/radiation/ https://xkcd.com/radiation/

effie 7y ago
You miss the point. Studies on low level radiation effects study, for example, deviations from the linear model, but are so far inconclusive, because it is hard to prescribe and control the level of radiation over such a long time. It could be done in lab, but who would subject themselves willingly to regular blasts of radiation? Radiation intensity in planes varies a lot, depending on the height, time of flight, geographical location, etc., so little can be inferred. It would take decades and lots of volunteers willing to get harmed in the name of science to do such studies reliably. Do you know of any such study?