, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

The most important thing to retain from the recent three country study of nuclear workers is that it supports the no safe dose-linear no threshold model (LNT) for ionizing radiation. Increased radiation exposure is increased risk. It shows association between protracted very low dose exposure to ionising radiation and cancer. Although high dose rate exposures were presumed by many “to be more dangerous than low dose rate exposures, the risk per unit of radiation dose for cancer among radiation workers was similar to estimates derived from studies of Japanese atomic bomb survivors.” (Richardson et. al., BMJ, Oct 2015).

The radiation exposures over an entire career, in this INWORKS study, are generally very low cumulative doses, even compared to what is considered “acceptable” radiation exposures for the general public from nuclear facilities and nuclear waste. However, the researchers appear to false the results by choosing to use the arithmetic average (mean) exposure of 20.9 mSv rather than the more appropriate median average of 4.1 mSv, thus reducing apparent cancer risk by a factor of 5.1 simply by this choice. Nonetheless, regardless of interpretation of this study – and there are several – it shows that excess cancer deaths, and thus cancer rates, are worse than even predicted by the BEIR VII (2006) report. How much worse ranges from around three times worse than BEIR VII to 26 times worse, (i.e. from per 100 people there would be 29 extra cancers per Sievert – which is approximately BEIR’s upper bound- to 262 or more extra cancers per Sv in a population of 100, meaning that 100% of the population having excess cancers would kick in at around 400 mSv. The middle estimate interpretation is around 152 extra cancers per 100 people per Sv meaning that a 100% excess cancer rate kicks in at around 700 mSv. Cancer for everyone would occur at lower doses due to already high cancer rates. Using estimates, a good computer program, accounting for radionuclide half-lives, could probably show that most of the existing cancers came from a combination of over 70 years of nuclear emissions, including above ground weapons testing, over-use of x-rays and medical isotopes, (and the earlier the radium craze). [Update note: The radium craze lasted for much of the 20th century in the form of radium watches-clocks, which are still found in homes and very hot, e.g. 70 mSv per year if worn 24/7]. Much of this radiation is still in the environment. At the beginning of the 20th century there was little cancer and surprisingly good documentation on cancer deaths in the US. According to the BEIR VII estimates, the approximately half of those who die from cancer due to radiation will have had their lives shortened by 14 to 15 years. Considering the recent US NRC proposal to raise the public exposure to 100 mSv per year we are forced to speak of the very real possibility of exposure from nuclear facilities, which would lead to 100% cancer rate for Americans within a few decades. The country would implode under the social and economic consequences. At the currently allowed exposure levels, radiation induced cancers will continue to slip under the radar, as they incrementally increase due to continuous exposure and build-up in the environment.

Returning to the Richardson et.al. study abstract: “The estimated rate of mortality from all cancers excluding leukaemia increased with cumulative dose by 48% per Gy (90% confidence interval 20% to 79%), lagged by 10 years.” In this context of external radiation exposure, Gy is equivalent to Sv (1000 mSv). Hereafter we will refer to Sv and mSv instead of Gy and mGy, unless it is a direct quote, because for external low LET gamma radiation examined in this study they are considered the same. In contrast, internal alpha emitters ( hi-LET) are weighted by 20 to convert from Gy to Sv, so that 1 Gy of these would be 20 mSv of biological impact.

What is mortality rate? “Mortality rate, or death rate, is a measure of the number of deaths (in general, or due to a specific cause) in a particular population, scaled to the size of that population, per unit of time.https://en.wikipedia.org/wiki/Mortality_rate

Initially we understood “The estimated rate of mortality from all cancers excluding leukaemia increased with cumulative dose by 48% per Gy” to mean 48 extra cancer deaths per 100 (48%) excluding leukemia, per Gy (Sv; 1000 mSv). The figure is .58 (58%) with a 15 year lag. Cancer rates are roughly double, meaning 116%. However, further down in the study they speak of excess relative rate as .48. If we assume that this is the same as excess relative risk (ERR), as defined by BEIR VII (see note at bottom of our post) – which seems to be (almost) the case based on the press release – then we arrive at a lower number. Or do we?

Actually, if we use the more appropriate median average of 4.1 mSv (half above and half below) rather than the 20.9 mSv arithmetic average (mean), which they use, the risk would be 5.1 times higher. ERR, according to BEIR, is exposed population risk divided by unexposed population risk minus 1. Thus, in their example of 0.48 the ERR would actually be 2.45. Assuming BEIR’s lifetime risk estimate of dying of cancer (except leukemia) for men of 22,100 per 100,000 people, then one arrives at a higher number than our initial interpretation. There would be 54 extra deaths from cancer out of 100 (54%) due to cumulative 1 Gy (Sv) exposure. Baseline lifetime cancer rate of 45.5 per hundred (excluding leukemia) means 111 extra cancers (excluding leukemia) per 100 due to cumulative 1 Gy (Sv) exposure or 10 years of the 100 mSv annual radiation exposure recently proposed by the US NRC for the general public. This excludes excess leukemia, (as they insist on doing). Using Richardson et. al’s preferred 10 year lag scenario, excess cancer deaths including leukemia for men (lifetime estimate) would be 59 per 100 for 1000 mSv (1 Sv) cumulative exposure and excess cancers of 120 per 100 for 1 Sv (1000 mSv) cumulative exposure (Baseline cancer rate for men is 46.3 per 100 including leukemia). Using the 15 year lag all excess cancer deaths for men (lifetime estimate) would be 68 per 100 for 1000 mSv (1 Sv) cumulative exposure and excess cancer rate of 137 per 100 for 1000 mSv cumulative exposure. This is above and beyond the already high risk of getting cancer and dying from it. See Table 12-13 p. 291 “BEIR VII Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII – Phase 2” for cancer rates and their LAR (Lifetime Risk) estimates for 100 mSv exposure (Available online for free).

From Richardson et. al., 2015:
Table A1 showing 5, 10, 15 yr time lag; Richardson et. al. 2015 BMJ INWORKS
Richardson et. al., 2015, BMJ.
According to the press release, for Richardson et. al., about 1 in 100 (1%) of the workers who died of cancer, other than leukemia, died due to their work related radiation exposure. It says that the risk of death from solid cancers increased by about 5% per 100 mGy. This appears based on use of the arithmetic average of 20.9 mSv cumulative dose over a career, whereas the median of 4.1 mSv cumulative dose appears most appropriate in this (and most) contexts due to extreme exposure outliers. See press release here: https://miningawareness.wordpress.com/2015/10/22/nuclear-worker-study-affirms-that-low-doses-of-radiation-are-deadly-increased-cancer-risk-much-worse-than-previously-believed/

Richardson et. al. 2015 BMJ INWORKS p. 4 showing data skew to low dose
Richardson et. al. 2015-BMJ, p. 4

Even Richardson et. al. note that the distribution of cumulative dose estimates was skewed with the median 4.1 mSv, mean 20.9 mSv, 90th percentile 53.4 mSv, maximum 1331.7 mSv. (It is important to take note that the ICRP and US NRC find 1 mSv per year exposure to the general public from routine emissions from nuclear reactors and other nuclear facilities acceptable, and the US EPA .25 mSv.)

Dividing the 1% (0.01) by 20.9 mSv and multiplying by 1000 to get 1000 mSv (1 Sv) yields 0.48 (48%), the figure which they call Excess Relative Rate per Sv However, dividing the 1% (0.01) by 4.1 mSv and multiplying by 1000 yields a figure which is 5.1 times higher, i.e. 2.44 (244%) for the Excess Relative Rate.

If one uses the 20.9 mSv as they do and assume the overall lifetime cancer death rate given by BEIR for males of 22.1 (excluding leukemia) per 100, one would get 10.6 extra deaths per 100 for 1000 mSv cumulative radiation exposure. Lifetime cancer risk for males, excluding leukemia is 45.5 per 100. Thus, it would be 22 extra cancer cases per 100 people (22%) if one uses the 20.9 mSv, as they do. (Note that they also have an upper bound number for the 0.48 figure of 0.79, which leads to a still higher cancer incidence and death rate.)

However, if one uses the more appropriate 4.1 mSv one arrives at a figure of 2.45 rather than 0.48, and extra lifetime cancer risk, excluding leukemia, of 111 cancer cases out of a male population of 100 per 1000 mSv (1 Sv). For excess cancer deaths it would be 54 extra deaths, out of a male population of 100 (54%), excluding leukemia. In other words, due to cumulative 1000 mSv of ionizing radiation, alone, everyone would get cancer, and around half would die. This excludes all other exposures; it excludes accumulation in the environment.

What the press release discusses is numbers who died due to exposure (choose 20.9 mSv or 4.1 mSv) divided by those who would die otherwise plus those who died due to exposure, which is not exactly excess relative risk and gives a lower number than excess relative risk. Is this what they mean by excess relative rate? Or, is the press release intentionally low-balling?
(Excess Relative Risk is risk of those exposed divided by those not exposed (RR) minus one.) Choosing to use 20.9 mSv instead of 4.1 mSv also gives a five times lower risk.

Perhaps there are alternative readings to this study, because there was disagreement between the 13 authors as to the conclusion? Who wrote the abstract? Was it written by someone who believed that 4.1 mSv was the appropriate number and wanted to give everyone that hint by writing this sentence which could be interpreted in two different ways: “The estimated rate of mortality from all cancers excluding leukaemia increased with cumulative dose by 48% per Gy…“?

Their excess relative rate seems based upon a relative rate best fit line, which may not necessarily to be the best fit – based on the graph itself.
Richardson et. al. 2015, Figure S1.  Relative rate of all cancer other than leukemia by categories of cumulative colon dose less than 100 mGy, lagged 10 years in INWORKS.
Richardson et. al. 2015, Figure S1. Relative rate of all cancer other than leukemia by categories of cumulative colon dose less than 100 mGy, lagged 10 years in INWORKS.
Adjusted Richardson et. al. 2015, Figure S1.  Relative rate of all cancer other than leukemia by categories of cumulative colon  dose less than 100 mGy, lagged 10 years in INWORKS.
Adjusted-Modified from Richardson et. al. 2015, Figure S1. Relative rate of all cancer other than leukemia by categories of cumulative colon dose less than 100 mGy, lagged 10 years in INWORKS. Note that RR of 1.10 for 100 mSv. It would be Excess Relative Risk (ERR) of 0.10 per 100 mSv and ERR of 1 (100%) per 1000 mSv. This excludes leukemia and uses the 10 year lag, meaning that the ERR should be even higher if leukemia were included and the 15 year lag used.

If one compares another best fit line from the BEIR VII report (p. 16) one can see that BEIR’s best fit appears done in a manner more similar to ours than Richardson et. al.’s line is done. Our best fit line would result in a Relative Risk of 1.97 to 2 per 1000 mSv, which appears roughly in line with Cardis’ 15 country study of nuclear workers. Both exclude leukemia. Thus, one would arrive at excess relative risk of around 100% (RR-1) or approximately 45.5 additional cancers per 100 (excluding leukemia), which along with the existing 45.5 cancers per 100 people leads to 91 additional cancers per 100 people – of which 45.5 would be due to cumulative 1000 mSv exposure. This is based on BEIR VII-2, Table 12-13, p. 291, lifetime cancer risk estimates. If you use the 4.1 mSv median rather than the 20.9 mean, then logically this number would also be 5.1 times higher or 232 excess cases per 100 people per 1000 mSv or 116 cases per 500 mSv, still excluding leukemia.

Overall cancer rates seem to be slightly higher in the US than in the UK, perhaps due to US above ground weapons testing. Thus, the excess case (morbidity) and death (mortality) numbers would be slightly lower.

Excluding leukemia low-balls the number. Choosing a time-lag of 10 years, rather than 15 years, also leads to a lower risk assumption. If we include leukemia and use the 15 year lag time the excess relative rate number jumps from 0.48 to 0.58 per 1000 mSv, with an upper bound of 0.92. For males the estimated lifetime cancer cases, excluding this particular exposure, is 46.3 per 100. Thus, the excess cases would be 26.8 of 100 men. The upper bound number of 0.92 would mean 43 cases per 100 men. However, assuming 4.1 mSv (median) exposure for the workers, rather than 20.9 exposure (mean) means that one must multiply the 0.58 by 5.1, leading to 137 cases per 100 men, meaning that 100% cancer rate would be arrived at less than 1000 mSv.

It is important to note that the study focuses on male nuclear workers, with a median average age of 58. Median age of diagnosis for all cancers in US is 66 yrs. http://www.cancer.org/acs/groups/content/@research/documents/document/acspc-042801.pdf

Women Excess Cancer Risk is 1.5 times higher than for men

According to BEIR VII, excess cancer risk for women is around 1.5 times higher than for men, though the baseline of cancers is less. Women cancer lifetime risk x .58 x excess cancer risk compared to men is 37490 per 100,000 x .58 is 21744 x 1.5 for excess cases compared to men is 32616 (32.6 per 100 per Sv of exposure) x 5.1 is 166,342 per 100,000 for 1000 mSv of exposure (166 per 100). If their graph is wrong, and if the RR is 2 and then ERR is 1, it would be 37,490 for 100,0000 women x 1.5 is 56,235 x 5.1 is 286,799 suggesting that around 100% of women would get cancer after an exposure of only 350 mSv (100 per 100). Thus, after only 3 1/2 years of the 100 mSv per year, recently proposed by the US NRC, there would be enough extra radiation exposure for all women to get 1 extra cancer. Around half will die, with their lives shorted by 14 to 15 years, according to BEIR VII. For men, the number is 46.3 x 0.58 x 5.1 is 137 and if their graph is wrong it would be 236. In this last scenario, it would take around 5 years of 100 mSv for all men to have an extra cancer. Since lifetime chances of getting cancer are already around 37.5 per 100 for women and 46.3 for men, everyone would have cancer even sooner. Note that women have a 1.6 higher excess risk for cancer, excluding leukemia, compared to men and multiply accordingly.

BEIR VII (2006), which did not get to consider either Richardson et. al. or fully consider Cardis’ 2005 work (see BEIR Appendix E) when arriving at their estimates, gave a high number estimate for men and women of 28000/100,000 cancer cases per Sv (Table 12-3), p. 291. This is largely in line with a low assumption based on Richardson et. al., 2015. If we assume that the 0.58 is increased cancer rate, and account for the 1.5 greater risk for women and adjust for the US sex ratio of 0.97 men to women, assuming cancer rates given by BEIR, we end up with 29000/100,000 cancer cases (29 per 100) per Sv. This is 2.9 per 100 for 100 mSv, roughly three times more than BEIR’s estimate of 1 out of 100 people getting cancer in a lifetime due to exposure to 100 mSv (Thus, 10 per 100 due to 1000 mSv). However, if we use the more appropriate 4.1 mSv rather than 20.9 mSv then we arrive at a figure which is around 15 times higher than the BEIR average estimate: 15.2 out of 100 will get cancer in a lifetime due to 100 mSv exposure and 152 out of 100 for 1000 mSv. Although people can get more than one cancer in a lifetime, this suggests that 100% excess cancer rate kicks in at about 700 mSv. The upper bound for Richardson et. al. is 0.92 or around 47.3 extra cancers per 100 people for 1000 mSv. Once again, this appears based upon the 20.9 mSv and would presumably be 5.1 times higher or 241 extra cancers per 100 people for 1000 mSv or 108 per 450 mSv, if they used the 4.1 mSv median average.

Shockingly, almost 2 months after this study was released, the authors have made no effort to make an understandable, useful, version for the general public. Rather, two of the authors in a Press Release and in a Reuters interview have further muddied the waters by speaking of CT scans, which have nothing to do with the nuclear industry, the topic of the study. In theory, the person undergoing CT scan gets some benefit from it (though the educated know to ask for MRIs), whereas the general public gets no benefit from being poisoned by the nuclear industry. Any alternative source of energy is safer. And better to do without power, as humans did for most of their history, than to die grisly deaths from cancer. Cataracts, heart disease, and other diseases caused by exposure to ionizing radiation may be discussed at a future date by Richardson, et. al. According to the US FDA, and others, “More recent studies suggest that the lowest cataractogenic dose in people is much less than these values, is statistically compatible with no threshold at all, and that increasing dose is associated with an increasing prevalence of cataracts.” (USFDA)

While we no longer accept comments, it would be most appropriate for Richardson et. al. to issue a statement of some sort clarifying what their results mean in terms of public and worker exposure to the lethal radionuclides legally emitted from nuclear reactors and waste on a routine basis. This is all the more true since Richardson is chair of an EPA committee on this topic, and, in the context of the recent 100 mSv per year public exposure to nuclear emissions proposal by the USNRC. However, at this point we are too skeptical of the authors of Richardson et. al. to trust anything which they say, because there appears to be an intentional downplaying of the results. One can but hope this is for good intent and not bad intent, but several of the researchers have clear conflicts of interest.

It is important to note that the majority of nuclear workers, over their entire career (cumulative), were exposed to much lower doses than the nuclear industry-nuclear labs are allowed to expose the general public to, excluding accumulation in the environment, (and excluding bioaccumulation). The US NRC and ICRP allow a maximum exposure for the general public of 1 mSv per year. The US EPA is supposed to try to hold radiation exposure down to around 1/3rd of their 0.25 mSv maximum due to ALARA. In only 4 years at 1 mSv per year the public would reach the 4.1 mSv median average that the workers were exposed to over their entire career.

The majority of workers’ exposures were concentrated in very low cumulative dose rates. At the lower end of interpretation of estimates given by this study, we find that BEIR needs to do away with the low dose DDREF reduction factor of 1.5 and the ICRP to do away with its reduction factor of 2 for low doses. We arrive at a figure which is in the range BEIR’s upper number. At the upper end of interpretation, we arrive with a figure of over 200 cancers per 100 people for a cumulative exposure of 1000 mSv.

We have chosen 0.58, using the 15 year lag for all cancers (as opposed to the 10 year lag excluding leukemia). How they are using lag time in this study is unclear. Are they including all cancers which occur after 10 years or after 15 years? This appears unlikely as their excess relative rate is higher at 15 years than at 10 years. Latency for cancer between exposure and cancer development is around 4 to 30 years or more, depending on the individual and cancer type. Then deaths must occur. Many die in less than 5 years. More die in less than 10 years. 37% die in less than 5 years. 59% die in less than 10 years: http://www.cancer.org/acs/groups/content/@research/documents/document/acspc-042801.pdf. Others hang on longer. The Richardson et. al. study is based on death records. The only thing that we do know is that, according to Richardson study, the excess cancer rate jumps at 15 years, compared to 10 years. They need to include all cancers from 4 years after exposure to the present. The study actually cuts off around a decade ago, as well.

We are left with numbers which range from near the upper end of the BEIR VII report of 2.8 extra cancers per 100 people (2.8%) per 100 mSv (28 per 1000 mSv) and an upper range of 100 to over 200 cancers per 1000 mSv (meaning 100% at around 500 mSv). Whichever figure is correct, this excludes medical exposures, it excludes accumulation in the environment – ongoing since the beginning of the nuclear age, and even before due to the radium craze. It excludes bioaccumulation. It also excludes the already high cancer risk, which may well be caused by radiation exposure, simply not this particular exposure. If these numbers are included 100% of the population having cancers kicks in much sooner.

The only thing which we can know for certain about this study is that it affirms that there is no safe dose of ionizing radiation. Any exposure at all can cause cancer: BEIR’s “nonzero chance”: “Whatever molecular mechanism is envisaged for radiation, at very low doses (e.g., 0– 5 mGy low LET), increases in dose simply increase the probability that a given single cell in the tissue will be intersected by an electron track which will have a nonzero probability of inducing a biological effect. Therefore, at these very low doses, a linearity of response is almost certain (Chapter 3). Second, given the intimate relationship established between DNA damage response, gene or chromosomal mutations, and cancer development, the form of the dose-response for mutation induction in single cells should be broadly informative for cancer initiation.” “Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII – Phase 2“, 2006, p. 245)

Furthermore, the recent Richardson et. al. study shows that exposure to very very low doses of ionizing radiation, over time, is at least as dangerous as a one-time exposure. This is important, and not a given.

While some, or even most, of the researchers are probably trying to be even-handed, which can also mean understating risk to be “conservative”, one or more may well be swayed by employment and/ or funding. Clearly due to tainted funding, the results are as bad as stated or worse. Whether intentional, or not, the misleading nature of the press release is breathtaking. Although the author of the press release is French, the Press Release reads the same in French. Press Release found here: https://miningawareness.wordpress.com/2015/10/22/nuclear-worker-study-affirms-that-low-doses-of-radiation-are-deadly-increased-cancer-risk-much-worse-than-previously-believed/

Funding, as listed in the study: Support from the US Centers for Disease Control and Prevention; Ministry of Health, Labour and Welfare of Japan; Institut de Radioprotection et de Sûreté Nucléaire (pro-nuclear); AREVA (French state owned nuclear utility with major interest in showing nuclear is good for you); Electricité de France (French state owned nuclear utility with major interest showing nuclear is good for you); US National Institute for Occupational Safety and Health; US Department of Energy (pro-nuclear with major interest in showing radiation is good for you due to need to clean-up nuclear sites); and Public Health England. Data are maintained and kept at the International Agency for Research on Cancer (UN-WHO entity in Lyon, chemical center in pro-nuclear France. The UN also often appears pro-nuclear).

After almost 2 months there has been little discussion of this very, very important and poorly written study. The poor writing may well be due to the adage of “too many cooks spoil the soup.” With the many researchers and the way in which it was written, one can come to suspect that the unclear writing came from possible discord as to what the results really are.

While we hope for a readable summary written so that the general public can better understand the risks of legal leakage of lethal radionuclides into the environment from the nuclear industry, at this point we probably wouldn’t believe the low-ball number of 29.3 additional cancers per 1000 mSv (2.9 per 100 mSv), because there are too many strange things about this study.

We still don’t know for certain if Richardson et. al.’s Excess Relative Rate is the same as Excess Relative Rate or not.

About Relative Risk (RR) and Excess Relative Risk (ERR):
Epidemiologists use the term “risk” in two different ways to describe the associations that are noted in data. Relative risk is the ratio of the rate of disease among groups having some risk factor, such as radiation, divided by the rate among a group not having that factor. Relative risk has no units (e.g., 75 deaths per 100,000 population per year ÷ 25 deaths per 100,000 per year = 3.0). Excess relative risk (ERR) is the relative risk minus 1.0 (e.g., 3.0 – 1.0 = 2.0). Absolute risk is the simple rate of disease among a population (e.g., 75 per 100,000 population per year among the exposed or 25 per 100,000 per year among the nonexposed). Absolute risk has the units of the rates being compared. Excess absolute risk (EAR) is the difference between two absolute risks (e.g., (75 per 100,000 per year) – (25 per 100,000 per year) = 50 per 100,000 per year). If the rates of disease differ in the exposed and unexposed groups, there is said to be an asso-ciation between exposure and disease. None of these mea-sures of risk is sufficient to infer causation. A second step in data analysis is necessary to assess whether or not the risk factor is simply a covariate of a more likely cause.” “Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII – Phase 2,” p. 132

Richardson study found here: https://miningawareness.wordpress.com/2015/10/21/new-study-of-us-uk-french-nuclear-workers-supports-linear-no-threshold-model-radiation-is-bad-for-you-increased-dose-is-increased-risk-hormesis-debunked-funding-from-pro-nuclear-govts-nuclea/
Richardson et. al: “Risk of cancer from occupational exposure to ionising radiation: retrospective cohort study of workers in France, the United Kingdom, and the United States (INWORKS)” BMJ 2015; 351 doi: http://dx.doi.org/10.1136/bmj.h5359 (Published 20 October 2015, CC- BY- NC: http://www.bmj.com/content/351/bmj.h5359

Click to access ricd026856_updated_data_supplement.pdf



Notes re Radiation and Eyes-Heart
Radiation, turns out cataracts in eyes, eyes are very sensitive to radiation. And it turns out that, it turns out that the stem cells, maybe you didn’t know this, your, the lens of your eyes is completely being, is being grown all the time. There’s new layers of cells being put down all the time and at the back of the posterior portion of the lens. And the stem cells that generate these clear cells for the lens are very sensitive to radiation. And so if they’re damaged as a result of the radiation they end up making opaque cells. And so it’s very easy to see this signal of radiation damage in the form of a radiation cataract and noticed in many of the atomic bomb survivors but also in people who work around x-ray machines in the medical field or in other, nuclear industry.” (Dr. Timothy Mousseau)

According to the US FDA, and others, More recent studies suggest that the lowest cataractogenic dose in people is much less than these values, is statistically compatible with no threshold at all, and that increasing dose is associated with an increasing prevalence of cataracts. (USFDA) https://miningawareness.wordpress.com/2015/06/22/no-dose-of-radiation-is-safe-for-the-eyes-any-dose-can-cause-cataracts-usnrc-comment-deadline-today-11-59-pm-ny-dc-et-one-minute-to-midnight/

The ICRP says: “… a ‘threshold dose’ of 0.5 Gy is proposed here for both cardiovascular disease and cerebrovascular disease on the basis that this dose might lead to approximately 1% of exposed individuals developing each disease in question. Nevertheless, there are notable uncertainties in determining risks of these diseases at this level of dose.” (p.292) https://miningawareness.wordpress.com/2015/06/20/heartless-us-nrc-ignores-icrp-concerns-re-radiation-risks-to-cardio-cerebrovascular-systems-comment-deadline-monday-22-june-1159-pm-one-minute-to-midnight-ny-dc-et/

The lens of the eye is one of the most radiosensitive tissues in the body (Brown, 1997; Ainsbury et al., 2009). When the radiosensitivity of various eye tissues is compared, detectable lens changes are noted at doses between 0.2 and 0.5 Gy, whereas other ocular pathologies in other tissues occur after acute or fractionated exposures of between 5 and 20 Gy” (ICRP, 118, p. 117) https://miningawareness.wordpress.com/2015/06/22/no-dose-of-radiation-is-safe-for-the-eyes-any-dose-can-cause-cataracts-usnrc-comment-deadline-today-11-59-pm-ny-dc-et-one-minute-to-midnight/

An approximate threshold dose of around 0.5 Gy has been proposed for acute and fractionated/protracted exposures, on the basis that this might lead to circulatory disease in only one to a few percent of exposed individuals, although the estimation of risk at this level of dose is particularly uncertain.” p. 304, ICRP 118.