[ RadSafe ] NW report Inconsistent with Better Controlled USA NSWS

[howard long]

Is bigger better? Not when less well controlled.
 
Cameron, one of the USA NSWS science advisors, found many SD greater longevity.
"B Cohen calc - 2.8 years added life for extra 0.5 rem compared with otherwise identical controls"   J Cuttler to DDP 7/28/02, approx. The only confounder I see is that  assignment to identical work may have selected healthier workers, knowing they might be exposed.
 
If Cameron's paper, "Is Radiation an Essential Trace Energy?" attached, does not come through the list, I will reattach it to individual requests..
 
Howard Long  

"Muckerheide, James" <jimm at WPI.EDU> wrote:
st1\:*{behavior:url(#default#ieooui) }


NUCLEONICS WEEK JULY 7, 2005

Nuclear workers study supports linear no-threshold risk model

 

The largest epidemiological study of low-dose radiation risk so far supports existing radiation protection standards that assume even tiny doses can cause cancer but adjust risk estimates downward for low doses, the study's authors said last week.

 

The "International Collaborative Study" of cancer risk for 407,391 nuclear industry workers in 15 countries showed a slight but statistically significant excess risk of cancer even at the low doses and dose rates typically received by the workers, the scientists said.

 

They estimated that 1%-2% of the cancer deaths recorded among the population studied were attributable to radiation exposure.

 

The study is the first to provide solid evidence of a linear no-threshold (LNT) model for low-dose radiation risk. That model, based on the high doses and dose rates sustained by Japanese atomic bomb survivors, holds that the risk should be extrapolated linearly down to zero dose. The LNT model is contested by some scientists worldwide who contend that doses below 100 milliSievert (mSv) or even 200 mSv are harmless and may even be beneficial. But there are also those who argue that low doses and low dose rates are actually more dangerous than higher ones.

 

The results of the new study were published June 29 in the British Medical Journal ( http://bmj.bmjjournals.com/onlinefirst_date.shtml ). 

 

The team of scientists was led by Elisabeth Cardis, head of the radiation group at the International Agency for Research on Cancer (IARC) ( http://www.iarc.fr/ENG/Units/RCAa1.html ) in Lyon, France.

 

On June 29, the U.S. National Academies' National Research Council published a study on low-dose health effects that also supported the LNT model (NW, 30 June, 12).

 

An earlier IARC collaborative study with a much smaller cohort covering only the U.S., U.K. and Canada, published in 1994, had not found a statistically significant excess risk for all cancers other than leukemia (NW, 27 Oct. '94, 1). 

 

For leukemia, that study found an excess relative risk per Sievert of 2.2, lower than the 3.7/Sv extrapolated linearly from the survivors of atomic bombings at Hiroshima and Nagasaki, the main source of estimates of radiation risk.

 

At the time, Cardis had noted that the follow-up study was needed to draw more solid conclusions, since most members of the earlier cohort were still alive and solid cancers have a much longer latency period than leukemia.

 

The new study, which has been eagerly awaited but took over a year to publish, found an excess relative risk for cancers other than leukemia of 0.97 per Sv, with a confidence interval of 0.14 to 1.97 (i.e., excluding an absence of risk).

 

The excess relative risk for leukemia (excluding chronic lymphocytic leukemia, CLL, which is considered not radiation-related) was 1.93 per Sv, with a confidence level of 0 to 8.47.

 

Those estimates indicate that for a dose of 100 milliSievert (mSv), currently the five-year limit for occupational radiation exposure in international rad protection standards, the risk of dying from solid cancer is 9.7% higher than that for natural cancer mortality, and the risk of dying of leukemia is 19% higher.

 

The latest study covered workers in the nuclear industry, except uranium mining, between 1944 and 2000 in the three original countries plus France, Japan, South Korea, Sweden, Belgium and seven other European countries. 

 

The overall average cumulative recorded dose was 19.4 mSv, but the distribution of recorded dose was greatly skewed. Ninety percent of workers in the cohort received cumulative doses of under 50 mSv, less than 5% received doses of 100 mSv over their career, and less than 0.1% received cumulative doses above 500 mSv. Most of the higher doses were received in the early years of the nuclear industry "when protection standards were less stringent than today," the scientists wrote.

 

A total of 24,158 people in the cohort were known to have died during the study period; 6,519 of them died from cancers other than leukemia and 196 from leukemia excluding CLL.

 

Based on their findings, the scientists estimated that about 1%-2% of the deaths from cancer among workers in the larger cohort "may be attributable to radiation," the scientists wrote.

 

The central risk estimates for all cancers (excluding leukemia) are between two and three times higher than the linear extrapolation from atomic bomb survivors, the scientists wrote, but because the confidence intervals are wide, their findings are "statistically compatible with the current bases for radiological protection."

 

Compatible risk estimate

 

As for leukemia, the scientists said, the central risk estimate from the new study-which is similar to estimates found in earlier large-scale nuclear worker studies-is compatible with the current approach of radiation protection, i.e., a linear dose-response model with no threshold but dividing risk estimates by two to allow for assumed reduced effect at low dose rates. However, they said "the confidence interval is wide" and the findings are also compatible with no reduction at low doses and dose rates, as well as with greater reduction of risk at low doses. In fact, they said, their central risk estimate for leukemia is halfway between estimates obtained by fitting a linear dose-response model and a linear-quadratic model to data on men exposed to the atomic bomb at age 20-60.

 

In an interview, Cardis noted that compared to the previous collaborative study of three countries' workers, the new study includes many more individuals with lower doses, primarily because there are more recent data-the early doses were the highest-and the cohort is much larger. A graph in the paper shows more than 250,000 of the workers at doses under 10 mSv, most of the others under 50 mSv, and only isolated individuals with doses up to 1.6 Sv.

 

Cardis said the research team had excluded all individuals judged to have received 10% or more of their dose from internal contamination or neutrons, which eliminated some of the higher doses. The team also analyzed risk by time periods to see the effect of dose levels. Cardis acknowledged that "the statistical significance of the results is driven by the higher-dose workers." If doses higher than 200 milliGray (200 mSv) were eliminated from the calculations, she said, the risk estimate "was similar but no longer statistically significant."

 

Risk comparisons based on time period and excluding the higher-dose workers will be included in a second, more detailed publication of the study results, she said.

 

The fact that the estimated risk from lung cancer was particularly high among the cohort suggests that smoking may have been a confounding risk factor in the study.

 

Cardis acknowledges that the researchers did not have access to clear enough data about the workers' smoking habits and that smoking may well be responsible for part of the cancer risk.

 

However, she said analysis had shown a "much lower" risk of smoking-related cancers excluding lung cancer, and said all the risk estimates-for mortality from all groups of cancers, both related and unrelated to smoking-were consistently higher than the risk estimate for solid cancers from the atomic bomb analysis. That, she said, suggests that "although smoking is playing a role (in the deaths), it is not the only factor."

 

Cardis said that as a next step, IARC is starting case-control studies of lung cancer and of leukemia, nested within cohorts of workers who have been employed in uranium and plutonium processing facilities.

 

Those studies, she said, "will allow us to better quantify the risk of these malignancies related to both internal and external exposures and to take into account the possible confounding effect of smoking and of other occupational carcinogens."

 

She said IARC would "also like to extend the follow-up of the cohorts included in the 15-country study."

 

-Ann MacLachlan, Paris

=============================================