ML20058N587
| ML20058N587 | |
| Person / Time | |
|---|---|
| Issue date: | 08/01/1990 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| References | |
| REF-10CFR9.7 NUDOCS 9008140215 | |
| Download: ML20058N587 (197) | |
Text
...
1 i
i h
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMIS SION-i~
{*[N j '
BRIEFING ON DEVELOPMENT OF RADIATION PROTECTION STANDAPSS S
LOC 3 tion l ROCKVILLE, MARYLAND h&(&l AUGUST 1, 1990 1
l 230&$l 94 PAGES NEALR.GROSSANDC0.,INC, COURT REPORTERS AND TRANSCRIBERS l
1323 Rhode Island Avenue, Northwest Washington, D.C.
20005 (202) 234-4433 oosic,;,; g * * 'i pf0 rnn
- w. c h gg i
m.7 qs m
+.
.$v
-em,,,.,..
9
4 DISCLAIMER This is an unofficial transcript of a meeting of the United States Nuclear Regulatory Commission held on August 1, 1990, in the Com:nis tion 's of fice. at One White Flint North, Rockville, Maryland.
The meeting was open to public attendance and observation.
This transcript has not been reviewed, corrected or edited, and it may contain inaccuracies.
The transcript is intended solely for general informational purposes.
As provided by 10 CFR 9.103, it is not part of the formal or informal record of decision of 4
the matters discussed.
Expressions of opinion in this transcript do not necessarily reflect final determination or beliefs.
No pleading or other paper may be filed with the Commission in any proceeding as the result of, or addressed to, any statement or argument contained herein, except as the Commission may authorize.
HEAL R. GROSS court REpoRTtt$ AND TRANSCRit$R$
1333 RHoot 18 LAND AYtMUE, N.W.
(202) 734-4433 WASHIN0 ton. DA 20005 (202) 232 6600 l
l
I i
UNITED STATES OF AMERICA NUCLEAR REOULATORY COMMISSION BRIEFING ON DEVELOPMENT OF RADIATION PROTECTION STANDARDS PUBLIC MEETING Nuclear Regulatory Commission One White Flint North Rockville, Maryland Wednesday, August 1, 1990 The Commission met in open
- session, pursuant to notice, at 10:00 a.m.,
Kenneth M.
- Carr, Chairman, presiding.
COMMISSIONERS PRESENT:
KENNETH M.
CARR, Chairman of the Commission KENNETH C.
ROGERS, Commissioner JAMES R. CURTISS. Commissioner FORREST J.
REMICK, Commissioner i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
\\\\
2 i
STAFF AND PRESENTERS SEATED AT THE OL.J.:41SSION TABLE:
w -
4 SAMUEL J.
CHILK, Secretary MARTIN MALSCH, Deputy General Counsel DR. ARTHUR UPTON, BEIR V Committee DR.
WARREN SINCLAIR, President, National Council on Radiation Protection and Measurements DR. WILLIAM ELLETT, BEIR V Committee e
CHARLES
- MEINHOLD, Division
- Director, Radiological Sciences Division, Brookhaven National Laboratory DR.
BILL
- MORRIS, Director, Division of Reg.
Applications, RES 1
r __
u._.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
3 I
1 P-R-0-C-E-E-D-I-N-G-S 2
10:00 a.m.
3 CHAIRMAN CARR:
Good morning, ladies and 4
gentlemen.
5 This morning the Commission will be G
briefed about the development of radiation protection 7
standards by representatives of the organizations that H
helped shape these standards.
The purpose of the 9
briefing is to describe the process by which radiation 10 protection standards are developed based on estimates 11 of the risks associated with ionizing radiation and to 12 highlight the special precautions that are implemented 13 throughout this process to ensure protection of the 14 public health and safety and the environment.
15 This topic is especially relevant at this 16 time given the Commission's recent approval of the 17 final radiation production standards in 10 CFR Part 20 18 which should be published in the near future.
19 Joining us today for the briefing are 20 Doctor Arthur Upton and Doctor William Ellett from the 21 National Research Council's Committee on-the 22 Biological Effects of Ionizing Radiation, the BEIR V 23 Committeel Doctor Warren Sinclair and Mr.
Charlie 24 Meinhold from the International Commission on 25 Radiological Protection and the National Council on i
NEAL R.
GROSS i
i 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 L
(202) 234-4433
l 4
r-1 Radiation Protection and Measurements; and Doctor Bill u -
2 Morris from the NRC staff.
3 Welcome, gentlemen.
4 Do any of my fellow Commissioners have any 5
opening comments?
G If not, please proceed, Doctor Upton.
7 DOCTOR UPTON:
Thank you, Mr. Chairman.
8 (Slide)
May I start with the first slide, 9
please?
10 The first slide provides the membership, 11 lists the membership of the so-called BEIR V
12 Committee.
The BEIR V Committee.
B-E-1-R stands for 13 Biological Effects of lonizing Radiation.
The 14 Committee represents the fifth incarnation, so to 15
. speak, of an expert group that convened under the IG auspices of the National Academy of Sciences National 17 Research Council to review the status of our knowledge 18 of the health effects of low-level ionizing radiation.
19 The forerunner of the BEIR V Committee was 20 the BEIR III Committee which provided a comprehensive 21 review published in 1980.
Concern about the problem.
22 of course, goes back to the post-war era when the 23 testing of nuclear weapons in the atmosphere led to 24 rising levels of global fallout.
It was the 25 geneticists then who suggested that this increase in t
NEAL R. GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
1 5
r-1 the background level above and beyond natural 2
background posed a threat to future generations.
But 3
we didn't think at that time there were threshold 4
doses for mutagenic ef fects, that any level of 5
background radiation would carry some genetic risk.
6 More
- recently, it seemed that the 7
carcinogenic
- effects, if
- anything, might be more B
important.
In the decade since the BEIR III Committee 9
and the BEIR V committee, reassessment of the A-bomb 10 dosimetry in Hiroshima and Nagasaki, as I'll point out i
11 in a moment, suggested that the risks might be larger 12 than had been thought before.
There was another 13 decade of follow-up of A-b omb sarvivors.and other 14 irradiated populations.
So, it waa time to reassess
~
15 the subject.
16 (Slide)
If I could have the next slide, 17 please.
18 This brings out that in Hiroshima the dose 19 equivalent shown in the lower right-hand corner with a 20 new dosimetry, the DS86 dosimetry shown on the right 21 was only about two-thirds of what was
.hought to be 22 true in 1980, principally because of t: a much smaller 23' neutron component in the radiations.
In the upper 24 left, we see the gamma ray component in the gray bar 25 and the neutron component in the black bar.
As you i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
r 6
i I
can see, with the new dosimetry, the DS86 dosimetry, 2
the neutron component was much, much smaller.
3 Historically, the risks to the populatton 4
at Hiroshima for a given distance or a given dose 5
seems substantially higher in Iliroshima and that had G
been attributed to the contribution of the neutrons to 7
the dose there, the neutrons being known to be more 8
effective than gamma rays.
With the greatly reduced 9
neutron component and the reduced dose equivalent 10 shown in the lower right-hand corner, for a given 11 total dose the effects now seem subs entially larger.
12 (Slide)
Next, please.
13 This set of bar graphs is the counterpart l
14 for Nagasaki.
Again, in the upper left, the neutron 15 component la substantially smaller but it was always i
16 relatively small.
Finally in the lower
- right, 17 weighting the neutron component with an appropriate l
18 quality or RBE factor, the dose equivalent today in 19 thought to be substantially smaller.
So, these l
20 factors led to some expansion of risks needed 21 evaluation at this time.
22 (Slide)
Next, please.
23 Another factor is that ac the populations L,
24 in Hiroshima have been followed, the magnitude of the 25 radiation-induced excess cancer risk has grown with 1'
I l
- t. -
NEAL R.
GROSS l
1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
l 7
p__
1 attained age.
This is a figure from the BEIR III 2
report and at the top of the figure you'll see what 3
are the two alternative models that were presumed to 4
he applicable.
The so-called absolute risk model on S
the left held that following a latent period, the 6
cancer risk would appear and then would essentially 7
parallel the natural baseline risk shown in the dashed 8
line.
Projected over a lifetime then, there would be 9
a given additional number of cancers each year 10 attributable to a given dose.
11 On the right-hand side, at -the top, you 12 see the so-called relative risk model or 13 multiplicative risk model where after the elapse of r-14 the latent period and the appearance of a radiation 15 induced cancer excess, cancer excess grows each year 16 with attained age, so that the separation between the 17 radiation exposed populatian on the top and the 18 baseline population shown in the dashed line, that 19 separation grows.
The excess then is a
given 20 percentage or a
given additional fraction of the 21 natural incidence.
22 It would appear for the solid tumors in 23 Japan that the relative risk model or the 24 multiplicative risk model fits the data somewhat 25 better over the last decade.
Historically in the BEIR
(
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
8 m
1 III
- report, that relative risk model led to 2
substantially higher lifetime risk-estimates than did 3
the absolute risk model or additive risk model.
4 That's another reason why the riska seen higher today 5
than they seemed before.
6 For leukemias, shown in the lower figures, 7
it was thought at the time of BEIR III and still R
appears to be the case that the risk makes its 9
appearance after a latent period and then decays with 10 time so that uitimately the risk is much smaller than 11 it was during the peak period of excess leukemia 12 incidence.
13 (Slide)
Next, please.
14 Another factor.
At the time of the BEIR 15 III report, it wasn't clear which of the hypothetical 16 models relating incidence to dose was
- 1. deed the 17 correct model.
This is a figure from the BEIR III 18 report showing in the upper right hand figure-the so-19 called linear non-threshold model with increasing dose 20 plotted on the horizontal.
The risk of cancer rises 21 in proportion to dose.
22 The so-called quadratic model is shown at 23 the lower left where the risk goes up as the square. of 24 the dose.
At the lower right, one has the linear 25 quadratic model, risk rising linearly with dose in the i
u-NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 it (202) 234-4433
9 1
1 low dose
- range, low to intermediate, and thin the 2
curve bends upward as the square term or quadratic 3
term takes hold, and then in the upper left linear 4
quadratic model with a cell k i l l i rig term to account 5
for the saturation of high doses.
6 In the case of the BEIR V Committee, the 7
linear quadratic model would appear to fit the 8
leukemin data best, as it seemed to in BEIR III.
Most 9
of the leukemias have occurred, we're over the peak 10 now.
That's not surprising.
Looking at the solid 11 tumors, the committee thought that the linear model in 12 the upper right fit the data better than the others.
13 So, the use of the linear model contrasts with the
' ~
14 model that was used in BEIR III.
BEIR III used the 15 linear quadratic both for leukemia and solid tumors.
16 Linear models for solid tumors is another factor for 17 an increased risk projection, as Doctor Ellett will 18 bring out.
19 (Slide)
Next, please.
May we have the 20 next slide, please?
21 We're dealing now with many cancers, not 22 just leukemia.
Leukemia is shown at the top.
23 Relative risk of leukemia is highest, but cancers of 24 many sites and the number of radiation-induced 25 sites the number of sitee at which the excess has i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
T I.Ill I I I ll IIJ llII lil i I P'
I I I
I I
II I I i
+
1 l
10 r---
1 appeared continues to grow with time as one follows 2
the population.
Suffice it to say not all forms of 3
cancer are increased in frequency.
4 (S1ide)
Next, please.
5 Doctor Ellett will review the models that 6
were used for estimating the cancer risks.
Let me sey 7
just a word or two about the genetic risk estimates.
8 There's been little change over t.he past 9
20 years in genetic r i s!. estimates.
They've been 10-based largely on the information coming out of 11 experiments with laboratory animals, principally mice.
12 We do n e,w have substantial data from the children of 13 the A-bomb survivors in which no genetic detriment has 14 been observable, despite substantial efforts to 1S identify it.
So, we're still making estimates largely 16 on the basis of the mouse data and you'll notice in 17 the BEIR III estimates the largest category of genetic 18 detriment attributed to radiation was the so-called 19 regularly inherited diseases.
Some 20 to 200 20 additional cases per million live born per rem per 21 generation were thought to be attributable to the 22 radiation.
23 (Slide)
In BEIR V -- next slide, 24 please -- there was no effort to estimate the genetic 25 detriment as expressed in the regularly inherited i
i NEAL R. GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
~
i I
11 F
1 diseases principally because we simply don't know to 2
what extent this type of disease is attributable to 3
newly produced mutations.
Most of us carry 4
detrimental genes.
Most of the common diseases of old 5
age, arthritis, heart disease, cancer, are thought to 6
have some genetic component, presumably for the most 7
part in the regularly inherited category.
8 But if we eliminate those, as the REIR V 9
Committee
- did, there really is no substantial 10 difference between the new genetic estimates and those 11 that we've had in the past.
12 (Slide)
Next, please.
13 Another category of radiation-induced harm 14 was prominent in the review of the BEIR V Committee 15 and that is depicted here.
This is a curve relating 16 the incidence or prevalence of severe mental 17 retardation to dose plotted on the horizontal scale.
18 One can see at the top, the line at the top, that in 19 those A-bomb survivors who were irradiated between the l
20 8th and the 15th week of gestation, the frequency of 21 severe mental retardation increased steeply with dose, 22 amounting to some 40 percent of the dose of the gray.
23 The data don't indicate clearly whether 24 there's a threshold in the range of 10 to 20 gray.
25 That can't be excluded.
But neither did the data rule I
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
i 12 i
1 out the possibility of some linear slope in the low 2
dose domain.
3 A
smaller excess was seen in children 4
trradiated between the 16th and the 25th week.
But 5
children irrndiated enrlier than the eighth
- week, 6
early in the first trimester or after the 25th week of 7
development, that is the third trimester, there was no 8
demonstrable effect on the frequency of this severe 9
developmental disturbance.
10 (Slide)
The ef fect on developing brain--
11 next, please -- is also manifested in a decreased IQ 12 score, 10 plotted on the vertical.
One sees again in 13 that middle age group, the 8 to 15 week group, a very U
14 pronounced dose-dependent decrease in IQ scores.
Seen 15 also in the 16 to 25 week group, not evident in the 0 16 to 7 or in the 26 plus.
17 (Slide)
So, I
think we have evidence 18 reinforced by the next slide, please, which is school 19 achievement test scores for the high sensitivity of 9
20 the developing brain, especially at a critical stage 21 in fetal development.
These are school achievement 22 scores, again showing in the sensitive age groups, the 23 8 to 15, 16 to 25 week age groups dose-dependent 24 decrements.
25 End of slides, please.
i NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
13 5
1 I think we don't know from these data that 2
there is any substantial risk down in the millisievert 3
range, a few millirem, but clearly as one gets up on 4
the dose curve, one needs to be concerned about damage 5
to the developing brain at this critical stage in 6
organogenesis.
7 In summary then, I think, before turning 8
the floor over to Doctor Ellett, we've seen increases 9
in risk estimates for cancer, increases in the risk 10 estimates for the developing brain at substantial
- q 11 doses of radiation, no evidence for an increase in the 12 risk of genetic detriment.
13 I think I'd stop and let Doctor Ellett 14 explain the cancer risk estimates, if I may, please,
' ~
15 Mr. Chairman.
16 CHAIRMAN CARR:
Please.
17 DOCTOR ELLETT:
(Slide)
If I could have 18 the first slide, please.
19 I'd like to look a
little bit more 20 carefully at -- that's lovely but it's upside down.
I 21 brought in my slides
- late, so I
was asking for 22 problems.
23 What we're going to see eventually is the 24 curve for dose response for leukemia.
I'd 25 particularly like you to notice that the curve seems I
1 NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
14-i 1
to be very linear up to about 300 rem.
The., there's a s -
2 break that indicates that there may be a quadratic--
3 there is a quadratic component.
You can show this-4 with a high degree of statistical certainty at high S
doses.
Then there's a decrease at higher doses.
- Now, 6
this is in complete contrast to what we saw for solid 7
cancers where the curve just goes op straight linear 8
and then just levels off at high doser.
9 What the committee did was restrict their 10 analysis to doses below 400 rem so that they were Just 11 looking at the part of the curve before it flattened 12 out.
13 I
think we'll just have to hold it a
14 minute.
Ah, there we are.
15 It's really very linear.up to about 300 16 rem.
It isn't true that we're just looking at doses 17 at 100 rada or so.
This is all high dose effects.
18 The points on that curve are significant up to above 19 30 rad and it's certainly linear down to about ten rad 20 or so.
We know nothing at doses lower than that.
21 (Slide)
There's been a slow increase of 22 solid cancers over time.
If we could look at the'next 23 slide, please.
This is historical data up to 1982 24 from the A-bomb survivors.
Estimated risk is shown on 25 the horizontal slide at the top.
This is time in i
u_
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
I 15 t-1 years going down vertically.
You can see the leukeria 2
has fallen off very rapidly with time, but it's still 3
statistically significant and it is continuing on a 4
much lower rate in * 'oomb survivors.
The relative 5
risk for solid cancers is slowly decreasing as the 6
population ages.
7 The BEIR Committee was, I think, very -- I 8
must say I'm not u member of the BEIR Committee.
I 9
was the Academy study director on this, but I was 10 impressed by the way the Committee looked after the 11 uncertainties of the risk estimates.
They aren't 12 coming out with just magic numbers, point to risk 13 estimates and say, "Well, this many people are going 14 to die per rem."
They looked at the whole uncertainty
'~
15 and the data they were using and the models they were 16 using and got a
pretty good picture of what the 17 distribution of risks were for a given dose.
18 (Slide)
All this is shown in the next 19 slide.
This is the histogram for solid cancers.
This 20 is for acute dose of ten ren to the whole populations, 21 males on top, females on the bottom.
The 90 percent 22 confidence interval is about a factor of 2 on each 23 side of the mean, which is 800 cases, early deaths or 24 excess deaths per million person rem.
So, the 25 Committee makes no claim that their risk estimates are i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
16 I
1 exact, but they do claim that they have looked at the s
2 sources of error in the risk estimates, including the 3
errors that are inherent in the aesumptions they made.
4 (Slide)
Cancer risks show a pretty strong effect of age of exposure, as shown in the next slide.
G This is for females.
You can see the risks are 7
particularly high for females under 30 years of age.
8 Except for lung cancer, they've pretty much decreased 9
by mid-life, for exposures occurring in mid-life and 10 later.
Respiratory cancer is a late-occurring cancer.
I1
- Now, the curve for men looks about the same as this 12 except digestive cancers are half as
- large, 13 respiratory cancer is twice as large 14 Notice that breast cancer does not seem to
-~
15 be too important with the latest data where it used to 16 he rather dominant for female cancer risk before.
17 (Slide)
Looking at the area under these 18 curves, total risk is a function of age of exposure.
19 as shown on the next slide.
This is age of exposure, 20 5,
15, 25, 35.
Forty-five is the same am 35 21 essentially.
You car, see that for children, the riska 22 seem to be about three times higher than for adults, 23 but there's still a pretty goot risk up until age 30 24 and then it starts to fall off.
I think this has some 25 implications tor how you consider exposures for r
u -
1 NBAL R.
GROSS j
1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
1.1 17 1 a l_
occupational purposes.
2 I think if the models are~ held constant, a-3 linear-response, a
relative ri k
- model, that's 4
assumptions made by the BEIR V Committee on, I think, 5
rather strong evidence, riskLestimates haven't changed 6
too much over the years.
They've been_ fairly stable, 7
as shown in the next slide.
8 (Slide)
BEIR I model was 690 cancers per 9
million person rem.
BEIR III was 500.
BEIR V, it's 10 up to about 790.
This is the average for males and 11 females.
There was a
difference in the way-of 12 accounting for deaths in BEIR V that makes comparisons 13 a
little bLt hard.
BEIR V
looked at the excess 14
- deaths, BEIR Committees I
and III look at early 15 doaths, people that would have died of cancer in all 16 probability later, but they died early due to cancer.
17
.That increased the risk estimates by about'25. percent.
18 (Slide)
Finally, I'd like toclook at_the 19 last slide where I-show the range of risk estimates by 20 varioua BIER Committees in the sense they're becoming 21 more precise.
BEIR I and BEIR III had no idea of what 22 the uncertainty was in their risk estimates.
They 23 just provided different models.
As you can see for 24 BEIR III, risks per million person rem varied anywhere 25 from 10 cancers to 500 cancers.
- Now, I maintain i
1 NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
ll
}/
1 l'
18' I
I that's a
range so wide that it isn't really very-2 useful for metting radiation protection standards.
' 3:
BEIR V,
I think, had a lot more _ dat a to 4
work with.
Perhaps the data was better analyzed.
5 Error intervals are about a facto" of two on either 6
sido of the point estimate of 800 cases.
But I think 7
that gives us an idea of what kind of uncerEn>e. v 8
we're dealing with when we're talking about radiation 9
risks from acute exp res.
10 Something that I
think everyone has to 11 bear in mind is that these risks are the risks. for
. 12 acute exposures, risks that are thought to be lower at
'13 low dose rates.
How much lower is something that the g-14 Committee decided they could not specify except to say 15, perhaps it wae as much as a factor of two or more.
16-There is very little scientific information-on what 17.
number you should pick for a-dose rate effectiveness 18-factor, how much you should reduce the risk at low 19 doses'and-low dose rates.
My own view.is this varies 20 with different cancers and there is no magic number.
21 COMMISSIONER ROGERS:
What was your point-22 for definin.
acute?
That was greater than what 23 amount?
24 00CTOR ELLETT:
Why did - they say that it 25 was r
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
1 19 1
1 COMMISSIONER ROGERS:
- Well, h o w -- d i d you
-2 define acute?
What was your definition of acute?
3 DOCTOR ELLETT:
Oh,
- Really, I think if 1
you do it operationally it's in terms of the' exposures 5
that were received by the Japanese which are really.
6 very high dose rate exposures over a --
7 COMMISSIONER ROGERS:
Well,-
I mean= in 8
terms of numbers, rem or something of this sort.
9 DOCTOR ELLETT:
Well, acute would mean it 10 occurred over a very -- this is, I think, an important 11 distinction this Committee made.
They-maintained matte:S how small, even a rad at very high 12 doses, no 13 dcse rates, would be an acute exposure.
A small' dose m
14 isn'.t n e c e s t, a r i l y going to be less damaging.
If a 15 dose is distributed over
- time, and they used the 16 example of a year, then the effects become less.
17 COMMISSIONER ROGERS:
What is a-small 18 time?
How small is small?
19 DOCTOR - ELLETT:
I'll go to my radiation
.20 pathology expert for that.
21 COMMISSIONER. ' ROGERS:
I mean are we
.2 talking about --
23 DOCTOR ELLETT:
I would guess --
24 COMMISSIONER ROGERS:
-- seconds, micro 25 seconds, weeks,-days --
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
YQ 20 I.
1 DOCTOR ELLETT:~ Days.
months?
I< mean 2
COMMISSIONER ROGERS:
3 what's small here?
4 DOC TO R - IIPTON :
I don't think the Committee 5
really got into this discussion.
Perhaps it should G
haves.
I think from my point of view one would want to:
7 approach this in terms of microdosimetry, = energy 8
desposition per sensitive target per unit time.
And if 9
one does this, I think one has different time scales 10 for different kinds of radiation probably, different 11 distributions.
12.
I think you're leading into a
set of 13 questions which the BEIR V Committee really didn't 14 address systematically and I think it brings out the 15 importance of the ICRP and NCRP which must confront; 16 these issues of dose rate
- effects, spacial and 17 temporal distribution of dose critically if they're 18 going to.come up-with recommendations that are.useful 19 and practical spheres.
20-COMMISSIONER ROGERS:
- Well, Just to get-21 some feeling about where you begin to move from one 22 regime-into the other.
When you're talking about 23-
- acute, w'i n t 's not acute?
If you're talking about 24 short time, what is a short time versus a long time 25 and just get a layman's grasp of what we' re ' talking i
u NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
y 21
-I
'I about in terms of scales here.
2 DOCTOR UPTON:
Commissioner Rogers, as I 3'
- say, I don't think the Committee - wrestled with this 4
issue systematically and I would venture to say that 5
different radiobiologists' would provide different 6
answers.
I'm not sure how helpful I can be.
7 --
COMMISSIONER ROGERS:
Okay.
8 CHAIRMAN CARR:
That completes your 9
presentation?
10 DOCTOR ELLETT:
.Thank you, 11 COMMISSIONER REMICK:
Jue t a question on 12 administrative effectiveness factor.
You mentioned--
13 I' don't know if you said recommended two or more.
In
- ~'
14 the report itself I see figures like 2 to-10.
Are 15 those the same or should I'--
.16 DOCTOR ELLETT:
They made
'no 17 reccamendation on this en purpose because they did not-18 think there was a scientific basis for picking a 19 particular number.
I think the executive summary has 20 words'to perhaps as much as a factor of two or more.
21 Later in the report, I believe it says between two and-22 ten.
23-COMMISSIONER REMICK:
Okay.
24 DOCTOR UPTON:
Coming back to the issue of 25 acute versus less acute, sub-acute, chronic, it may I
i NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
f 22 r
1 depend on the cell which you'.re thinking about or the 2
target
- organ, the rate at which lesions can be 3
repaired.
I think you're putting your finger on a 4
very important point that I think needs more attention 5
systematically in research.
I don't think we've G
really addressed this critically yet.
7 00" f 0R ELLETT:
I-think the committee was 8
a little bi, surprised when they looked at breast 9
cancer.
'chey thought they would see a dose rate 10 effect because the Japanese received the dose in 11 milliseconds to a second or so.
People that had 12 ~
received radiation in their tuberculosis treatment, 13 small doses over a long period of time, the difference u
14-in cancer risk in these two groups was insignificant.
15 There was no dos'e rate effect and this makes people 16 pause a bit.
You can argue' for various reasons why 17 this would be so, but.I think on animal data you would 18 certainly have expected a larger dose rate effect.
19 CRAIRMAN CARR:
Okay.
Let's proceed.
20-DOCTOR SINCLAIR:
Thank you, gentlemen.
21 (Slide)
Could I have the first-slide, 22 plehse?
23 Gentlemen, I'm going to be talking about 24 estimates of cancer risk and detriment as the basis of 25 ICRP and NCRP recommendations.
Inevitably, I'll be r--
L NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
t 23 p.
l
' covering some points that my colleagues from the BEIR 2
Committee have already made to you, but I expect to do 3
it from a different perspective,.that of someone who's 4
looking toward the recommendations that ICRP and NCRP 5
will have to make, which Charlie Meinhold is-going to 6
talk about after me.
7 (Slide)
First of-all, what are the--
8 next slide, please.
9 What are the concerns in low dose 10 radiation protection?
- Well, they're primari'ly.
11 stochastic effects.
No threshold is
- assumed, the 12 magnitude of the effect is the same at all doses.-the 13 frequency is proportional to dose at low doses.
Those-m 14 are our assumptions.
The principal effects under that-
-15 category are hereditary effects and the induction of 16 cancer.
Hereditary effects, as we've just heard, have 17
'not changed recently, so we're going to concentrate on 18 the induction of cancer'.
19-There are two special problems that should 20 be mentioned,_ but I'll not deal with them here' and 21' that's the - risk of mental retardation in the fetus, 22 which Doctor Upton has discussed, and deterministic 23 effects.
Direct effects and damage to tissue are not 24~
an issue in low dose radiation protection, broadly 25 speaking that is, because the doses we expect we're NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
a 24 r---
-m 1-dealing with are below the thresholds for those 2
effects.
They would occur in accidents, but not
. 3 elsewhere.
4 CHAIRMAN CARR:
Let me - ask you, are your 5
low doses the same as the BEIR V's Icw doses?
I
~
6 understood BEIR V's low doses were down to ten rad and 7
they:didn't really get into lower than that.
Is tha' 8
where your cutoff is or do you start --
9 DOCTOR SINCLAIR:
Well, no.
We 10 consider -- in assuming no threshold, we consider that 11 our low doses are down in the region of one rad or 12 less.
The exposures that people might get 13 occupationally in particular and some exposure --
14 CHAIRMAN CARR:
I'm trying to figure out 15 there where you say deterministic effects are not a 16 concern because the limits are-below thresholds, I
17 don't know what limits we're talking about. -
18 DOCTOR SINCLAIR:
- Well, the limits for 19 occupational presently at 5 rems a year and even for 20 the public at 100 millirems a year for manmade sources 21 are way below the 50 to 200 rad threshold levels of 22 effects like cataract --
23 CHAIRMAN CARR:
Okay.
24 DOCTOR SINCLAIR:
and so on.
So, 25 they're not our concern.
They would be in an u._
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
e i
25
.i-1 accident, but they're not in low dose.
2 (Slide)
- how, there are two points I'd 3'
like to make before we. go to - some of the numbers.
4 First-of
- all, on the~ next-slide,- we can see the 0
exposed populations for risk estimation.
The Japanese 6
survivnr. eme first and they always head ' the list.
2 7
.But I wanted to ake the point that we have lots of 8
information from other sources as well,.
mostly 9
medical, and some of the studies-are quite important 10 and back up what we believe we know from the Japanese 11 survivors.
There's quite a list of them there, as you 12 can see.
.13 (Slide)
The next slide, please.-
14 This demonstrates another point that I
~-
15 think one needs to bear in mind - in considering the 16 induction of cancer by radiation.
If a population 17 were irradiated at time zero on that
- graph, we 18 wouldn't see anything, wouldn't see any leukemias for 19 two years because there's a latent period at least
-20' that long.
But-then it rises rather rapidly' for 21 leukemia, as you can-see, to a peak at about 5 or 7 it 22 years and then falls off almost to zero after 23 shouldn't go quite to zero on that thing 24 about 30 years.
25 Solid tumors, on the other hand, have a i
i NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
.n
1 26 I
1 latent period of about ten years and then they rise 2
more or less proportionally to the way s pc;it an e ous 3
cancers occur in the population as we age.
4=
CHAIRMAN CARR:
Now,-this is incidents?-
5-DOCTOR SINCLAIR:
It would-be incidents.
a 6
CHAIRMAN CARR:
And you-were dealing. in 7
' deaths, Doctor Upton?
8 DOCTOR UPTON:
Primarily, yes.
9 DOCTOR SINCLAIR:
Shape weuldn't be very-10 much the same, wouldn't be much different if it were 11 not.
The
- leukemia, for
- example, incidents -and 12 mortality are essentially the same, 13
. CHAIRMAN CARR:
Okay.
u_
14 DOCTOR SINCLAIR:
But some of the ~ other 15 cancer-is not.
An important point to note, beyond'40 16 years we don't yet know precisely how that curve-will 17 go.
There's some evidence-that it's beginning to drop.
18
- off, but we just don't know as yet because the 19-Japanese population which we follow the.most closely 20 hasn't reached there.
Furthermore, the Japanese 21 population, as we know, is.at this-point in time
~
22 somewhat less than 40 percent of them have died.
So 23 we have to project our lifetime risks to the others-24 and that's where these projections by relative risk 25 and multiplicative models come in.
u _1 NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
{
27-1
^ -
1 (Slide)
Next slide, please.
2 Back in 1977, the ICDP and not only the 3
ICRP - but ~ UNSCE AR-and the preferred model. at the time-4 for the BEIR C'ommittee and NCRP all considered that' Sc th'e risk from induced cancer was about 10-2 That's 6
one percent per sievert and that was an average value
- 7 of one and a half for females and one for males and 8
then rounded off and that's the number we used for a 9
long time for radiaticn protection.
10 (Slide)
What's happened since then?- -on 11 the.next slide we see that the ept '.em i o l o g i c a l 12 information since
- 1977, of course,-
includes
-an 13 extensive update in the Japanese A-bomb survivors,--but 14 update in all the other clinical studies.too which are 15 still being followed,-
particularly the ankylosing 16 spondy11 tics in the United
- Kingdom,
-an important-17 source of information; the International Cervix 18
- Series, which has become available. since 1977; and' 19 other clinical updates in particular organs such as 20 the breast and the thyroid.
21 (Slide)
The next one, please.
22 The Japanese survivors, as we've seen, are 23 the most important for risk estimation.
They provide 24 the largest sample and over the broadest dose range.
25 So, we have to concentrate on them and we have three l
'~
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433 4
b 7
28 I
new cycles of Japanese data since the
-1977 2
evaluations.
The increase in the solid tumor database 3
is f rom - not much more than 100 to a little bit less 4
than 300 between a factor of two and three.-
That's a S
big increase in --
G CHAIRMAN CARR:
Per what?
7
. DOCTOR SINCLAIR:
- Well, these are the 8
excess tumors.
9 CHAIRMAN CARR:
Per --
10 DOCTOR SINCLAIR:
Not per anything, the 11 absolute number of excess tumors at this point in 12 time.
13 CHAIRMAN CARR:
That's got to be over more 14-than~one person.
15 DOCTOR SINCLAIR:
Well, at this point in 16 time, I.think something like in the Japanese it would 17-have been between 5 and 6,000 cancer deaths in this 18 population.
~About 300 of those are at t ributable - to-19 solid tumor excess exposure,'all exposures.
20 There's more information on
.the.
age 21 dependence because even the youngest - age groups.. are 22 now beginning to reach the age when they are more 23 likely to get cancer.
There's more information on the 24 time course, of course, because we have 11 more years 25 and we've heard about the revisions in the survivor u-NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
-- g -
1 dosimetry.
We believe the DS86 system is a much more 2
sophisticated one than its_ predecessor and-it has of.the order of one-and a 3
increased the risks by 4
half-to two
- times, depending vhich organ you're 5
talking about.
6 (Slide)
Amongst tho evaluations of risk, 7
the next one, please, the Radiation Effects Research' 8
Foundation do their own evaluations and they did this 9
in 1988 and came to the conclusion that the total risk 10 in a
population of all ages for high dose rate 11 exposure was about 12 percent per sievert, the third 12 number down there, 11 percent.of its solid tumors, one 13 percent leukemia.
If one were to divide by a dose 14-rate effectiveness factor and they didn't opine on 15-that, they simply said other people have used about 16 two and a half, so if we did use that then we'd get 17 about five percent per sievert and if we'd applied 18 this to an adult population without'the_ younger, more 19 sensitive people in it,-
we would have got three 20 percent per sievert.
That's, of course, substantially 21 higher than our normal risk up till that time.
22 (Slide)
The next one, please.
23 And the UNSCEAR Committee-looked at all 24 the human sources of exposure that I listed in the 25 second slide, decided that the Japanese cample was so r--
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
2 30
. s- -
.x.
auch superior to the others that they would-2 concentrate on it for their. number generating. but.
3 they would
- use, for
- example, the ankylouing 4
spondylitics and the International Cervix Series to 5
support.that informatton.-
6 They used two projection models to get to 7
this lifetime risk, the additive model not favored so 8
much today and the multiplicative model, which is, and 9
obtained about 11 percent per sievert as' the number 10 for high dose, high dose rate in a population of all 11 ages.
For a working population, the same_ number is 12 eight percent per stovert.
And for low dose,' low dose 13
- rate, they didn' t actually do an evaluation ' of the 14 situation, they simply quoted others. and mainly the 15
.NCRP, as a matter of fact, and said, "We should. divide 16 by something.
We think it's between-two and ten 17 because that range has been used by others."
~ 18 CHAIRMAN CARR:- Before you leave that one,
'19 let me get the high dose rate and low dose -- Icsean
.4-20 high dose and low dose separated.
I understand the 21 rate at the bomb was instantaneously nearly.
What's 22 the high dose there you're talking about?
23' DOCTOR SINCLAIR:
- W911, you're talking 24 about if it's 400
- rad, 40) rad point in 25 Hiroshima / Nagasaki was delivered in a second or-less.
i u
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
~..
31 Jr-1 CHAIRMAN CARR:
Yes.
And'the low dose ---
2 DOCTOR SINCLAIR:
Lose dose, lose dose 3
rate we would consider to be the low doses that we get' 4
-in occupational and otherwise, which are of-the order 5
of fractions of a rem per year usually.
6 CHAIRMAN CARR:
So you're talking millired 1
7 ttore, huh?
8 DOCTOR SINCLAIR:
Yes.
9 CHAIRMAN CARR:
Okay.
10 DOCTOR SINCLAIR:
Five rems a year is the 11 limit.
We'd still consider that a low dose rate.
.12 (Slide)
The next one, please.
13 Then we have the BEIR Committee gave us t
14 three particular numbers in the top range there which-
~~
15
. we reinterpret to compare with UNSCEAR and one-can't 16 do this exactly for some of the reasons that Doctors
'17 Upton and'Ellett have already mentioned, but it comes 18 out at about 9 percent per sievert for a-population of 19 all ages and it would have been lese than that, about 20 seven percent per sievert for the working population.
21 The BEIR Committee did look at the question of dose.
22 rate effectiveness factors, decided they couldn't come 23 up with a number, again for reasons which have been 24 mentioned, and their advice was two or more.
25
- Well, this is a rather important number i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
32
- i 1
and.it's important in. radiation protection because we 2
do need nominal values throughout..
The advice that 3
UNSCEAR and BEIR have offered'us isn't sufficient for 4
us
-to do that.
A decision has to. he made by 5
protection people.
6 (Slide)
On the next slide, I've.tried to 7
show you what it is we're talking about in this dose 8
rate effectiveness factor.
The top-line, if we have 9
data from high dose rate sources, it would be up in 10 the top region of that top curve and we'd draw a 11 straight line down to zero and delta would be the risk 12 coefficient for that high
- dose, high dose rate-
.13 circumstance.
14 As we get information at lower and lower 15 doses and particularly from the laboratory in animals 16 and cell studies, we see that the shape of the curve 17 is: not a
simple linear.
It's most often linear 18.
quadratic, although it differs in different biological 19 systems.
So, we believe that the linear quadratic.
20 curve there, which ultimately gives you this slope 21-alpha, would represent the situation better.
Then the 22 dose rate effectiveness factor is the ratio between 23 delta and alpha, high dose rate information over the 24' expected low dose rate information.
25 As
- say, we get quite a bit of that NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005
-(202) 234-4433
l 33 1
information-from the laboratory and that's the primary 2
s'ource of information there.
One of the major 3
evaluations of that was done by the NCRP in 1980.
4 (Slide)
The next slide, please.
5 The next slide lists what people have 9.-
6 actually used, what other organizations have actually 7
used in the past for this dose rate effectivenessi 8
factor, starting with UNSCEAR in.1977 used two and a 9
half.
The DIER Committee, in using a linear quadratic 10 effectively in 1980, the BEIR III Committee was 2.25.
11 NCRP did this comprehensive evaluation in 1980 and 12 came to the conclusion that the range covered. by 13 animal and cellular data was between 2 and 10 and that
' ~'
14 the 2s were really different from the 10s-and depended 15 more on the
- system, so that simple averaging was 16 really not advisable.
17
.Now, there's one point about thet.
The 18 animal data and the cellular data normally cover.a 19 substantially broader-done range than the human 20 information.
Since the dose rate effectiveness factor 21 itself depends on
- dose, one might expect to get 22 somewhat higher values'in these animal systems than in 23 human ones.
That indeed turns out to be the case 24 because our human experience, if you look down a bit 25 on the slide, for breast and thyroid in studies made
=i NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
34 i
is _
1
~ about a decade or so ago, there was no effect of-2 fractionation in those studies whatsoever.
Now, they' 3
were not terribly precise studies-because they can't 4,
be in humans, but one could not have assumed from.that-5 a dose rate effectiveness. factor of more than one.
6 More recent
- studies, one on the breast 7
dose rate, does indicate possibly a factor of-three, A
and studies with I131 and comparing it'with external 9
radiation,'which may not be entirely due to dose rate 10 but that would be a factor in it, have shown about 11 four.
But one has to see that:the human experience is 12 down near that end.
13 Then for Hiroshima / Nagasaki
- itself, as 14 you've already heard, the leukemia fits quite well to-15 a dose rate effectiveness factor which would come'down 16 at about two.
The b'e s t fit for the solid tumors =is 17 still linear, which would 'give us about one.
.Now, 18 both of those, of course, are based on data which is 19 statistically not very precise.
So you could stretch 20 the solid tumors to two and the leukemias to a bit 21 more than two.
could I 22 (Slide)
Considering all that 23 have the next slide, please?
Considering *he fact 24 that in the animal data they tend to be a broader dose 25 range which might give you higher values, considering r- -
u _.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433 i
35 1
that-the human data don' t give you much - -although 2
not very precise -- don't give you much more than two.
3:
The ICRP decided that they would use 2,
4 and this is their estimate et-ttrdr-est-iwate o f the 5
risks for low dose, low-dose rate.
They start with---
G-by averaging UNSCEAR and PEIR, although that.also 7
could be ~done a number of ways, and say that the high 8
dose, high dose rate for a population of all ages the 9
risk is about 10 percent per
- sievert, and for a
10 working population it's about 8 percent per sievert.
11 And I
think it's worth mentioning that 12 probably the working population number is a little -
13 better known than the total population, simply because 14 included in the total population are the younger age 15 groups.
They have to be projected the most and 16 they're the least certain. -Then ICRP decided that the 17.
dose rate ef'ectiveness factor they would use was-2.
18 And therefore, for a population of all' ages 5 percent 19 per sievert, for the working population 4 percent per-20 sievert.
21 (Slide)
But that isn't all that you need 22 for a working protection system.
ICRP had to go 23-further and examine the details of the various 24 different organs, which is shown on the next-slide.
25 They determined these by comparing transfet models NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
i 36 r-I from the Japanese population to.
various
- others, 2
including the United
- States, the United
- Kingdom, 3
Puerto Rico, and China, and _ averaged the results of 4-their determinations in order to get what they 5
consider to be best numbers for the organ risks.
They o
6 add up in the right-hand - column there_ to' 510,000,-
7 which is the 5 percent per sievert for the whole 8
population of.all ages.
You'll notice most of them 9
are higher.
Bone marrow is higher.
Lung is higher.
10 But not all of them.
Bone surraces are about the 11
.same.
Breast is about the same.
Thyroid.is about the 12 same.
So in dealing with particular organs one needs l'3 -
'to look specifically at what the best values seem to
.14 be.
15 Well, in order to get a weighting system 16.
out of that, ICRP' realizes.of course that-these-organ 17 risks are not very precise, not as precise - ot-not as 18 well known as the total risk.
19-(Slide).
Could I -have the next
- slide, 20 please?
21 So for weighting factors, they decided 1to 22 form only four groups of weights and these are grouped 23 in.01 for the bone surfaces and the skin,.05 for the 24
- bladder, breast,-. liver, esophagus, thyroid and all 25 remainder tissues, and
.12 for bone marrow,
- colon, r--
i_.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
37 I
1 lung and stomach, and.20 for the gonada.
Now you 2
have to take the total detriment into account in 3
deriving that, and so far I've talked only about the 4
fatal cancer risk.
5 (Slidei if we a over to the last slide, 6
we see that the I C R P ' <,
estimate of the detriment 7
includes 1 percent for serious hereditary disease in a 8
l population of all
- ages, 5
percent for cancer 9
m o r t. a l i t y,
as we've
- seen, 1.5 percent for concer 10 morbidity by using a formulation which seems to have 11 been fairly acceptable to make an estimate of the 12 measure of detriment that results from a cancer that 13 is cured but was caused by radiation for a total of
~~'
14 7.5 percent for the de,riment for a population of all 15 ages.
F' o r the working population the corresponding 16 numbers, all a bit less, add up to about 5.8 percent.
17 But these numbers, of course, are some 18 three or four times higher than the nominal detriment 19 of 1.65 percent that ICRP used in 1977.
There are 20 some different things in them of course, in addition 21 to the increase in enncer risk.
But they did behoove 22 both protection
- bodies, both ICRP and
- NCRP, to 73 consider what should be done niaut this with respect 24 to recommendations, and I'll leave that to Mr.
25 Meinhold to talk about.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
38 i
a _
1 COMMISSIONER ROGERS:
Before you-leave 2
-that graph, do you really believe-these numbers to.
3 something like si tenth of a percent?
4 DOCTOR SINCLAIR:
Well, I'm tempted to sny 5.
of course not, but the formulation that one uses in 6
the cancer morbidity, for example, is one that simply 7
gives you that cort of number.
It's not easy to round.
8 it down to 1 percent or up to 2.
=9 COMMISSIONER - ROGERS:
- Well, it just sort 10 oflooks like it's a mix of numbers that are known to 11 different -- or believed to different extents.
12 LOCTOR SINCLAIR:
Well, I think that's the 13 one that is the weakest from that point of view, 14 because it appears to be the most precise.
And in 15
- fact, of course, it's based entirely on the way. in
,16 which-you judge the -- the risk of death, actually,
-17 from the given cancer is the -deciding factor and it 18 varies a great deal of course between leukemia with a 19 lethality ratio of about
.99, the thyroid with-a-
20 lethality of about
.10, and the skin with a lethality 21 of less then i percent.
And we try to correct for 22 that, but it appears more precise than it is.
23 CHAIRMAN CARR:
Thank you.
24 Doctor Meinhold?
25 MR. MEINHOLD:
(Slide)
Could I have the r-L NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
l Washington, D.C.
20005 (202) 234-4433
1 1
39' i
1-first. slide, please?
2 I'd like to begin by complimenting-the 3
Commission on adoption of. revised ' Part 20.
I think 4
that will do a great deal and I'm very happy to see 1
5 that and I don't want anything that I'll - be saying.
6 about what --
7 CHAIRMAN CARR You've seen it?
J 8
MR. MEINHOLD:~
Yes, I have.
9 CHAIRMAN CARR:
Oh, okay.
10 MR.
MEINHOLD:
I don't want to say 11
.anything that anything we're going to be saying.
12 here to suggest that we would disagree with that 13 decision at this time, because it's -- I think it's.an-14 excellent approach.
l 15 In. view of tho' fact that the time is 16 short, I won'.t be talking about the second through the l
17 sixth ~. slide, which really - were the -basis. for. the 18 ICRP's 1977 recommendations.
You k'now all of those,-
19 because that's basically what.went-into your revision 20 of Part. 20 and also into 'the Presidential guidance yt.
21 that EPA put out ini1987.
So I'm really going to use 22 that as a base time.
The 1977 ICRP publication is the 23 one that you've been dealing with recently in terms of 24 your new guidance, but we're
- asking, then, what's 25 happened since then, what's going on.
I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
h
'I C
40
.r.
L:-
l=
(Slide)
I've listed in the next slide, 2
please, the four major issues that we f ac<s in terms.of 3
radiation protection standards and whett.er or not we 4
need - to change our basic recommendations.
And the 5
first of those is that industry is becoming safer, and-6 is you recall we - used-safe industry as our criteria.
7 And of
- course, you've heard a
lot about the new 8
-Japanese survivor data..
You've henrd something about 9-the projection models and the change of emphasis that 10 ICRP and NCRP would use.
And of course, you've heard 11 something of the fetal risk concern.
12 (Slide)
The most important of these, if 13 we could go to the next
- slide, please, the most 14 important perhaps that the public saw even before we 15 got the Japanese data is this question of accidents.
16 frequencies going down in' the. safe industries, and 17 I've just given you a rough-cut from. the National 18 Safety Council kind of data showing you that all 'of 19 these are decreasing at a rate of about 2 percent, 2 20 or 3 percent-per year.
21 (Slide)
And of course this curve, if we 22 go to the next-slide as well, this distribution merely
.3 shows you two things, that they're all going down-in-24
' terms =of their likelihood of death in those industries 25 and secondly that the safe industries are still I
s.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Y Washington, D.C.
20005 (202) 234-4433
I
'41
.1-1 roughly in the range-of lx10-4, but with a trend-down.
2 CHAIRMAN CARR:
Those numbers are per 3
million people?
4 MR. MEINHOLD:
Yes.
-5 CHAIRMAN CARR:
Per year?
6 MR. METNHOLD:
Hight.
So if you look at 7-the first one would have been -- for instance, service-8 would have been 1.14x10-4, which is the kind of number 9
you're probably used to hearing in terms of the fatal 10 acr : dents.
11 (Slide)
If we could go to the next slide, 12 the-NCRP issued Report 91 in 1987 in which -- and-if in which the 13 we could go to the next slide, please 14 NCRP was beginning to be aware, of course, of changes 15 but felt that it was time for us to revise our 16 recommendations.
And in that one, we did adopt for a 17 nominal. risk, the risk of 10-*
per rem, which was the 18 1977 ICRP and UNSCEAR evaluatiua.
We also noted that s
19 the annual dose equivalent to monitored workers was in 20 the order of 230 millires, and so we felt that it was 21 still-rensonable to stay at that time with 5 rem per 22 year cr 50 millisieverts.
23.
(Slide)
The next slide I think indicates 24 some of the concerns that we have, and that of course 25 was that the risk estimates were likely to increase.
-i NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
1 r
42
- I x -
1 We ' knew that-th'e new data was in Japan.
We Just 2,
couldn't;get a quantitative handle on the_ numbers.
3' (Slide)
And we' also knew about the safe 4
industries, if I could go.to the next slide, please.
5 So we took three steps, some of which you've taken in 6
your revision to cart 20, of course, discontinuing the 7
age minus 18 recommendation times 5.
But - secondly, 8
and we think-this is terribly important and I believe 9-
. you' ve s t ressed this as well in your division, the 10 upper boundary nature of the dose limit, that it's 11 simply an inappropriate quantity to use for desig.4 12
- criteria, that sort of thing, that it's a boundary 13 condition to whatever approach you take to your design w
14 and that it's really trying to point out that the dose 15 limit itself is the edge.
It. isn't a desire or a 16-goal.
17 (Slide)
And then the next slide is one in-18 which we probably could have made our job a. lot easier 19 today.
In fact, I believe that if in 1977 we had made 20 this discussion about keeping individuals below their i
c 21 dose
- limit, below,their age in turma of tens of 22' millisievert or in-
- rem, if we had made this 23 recommendation at that time.to apply to the i n divi'du al y
24 we probably would be in -- wouldn't have to revise in 25
'91 at this time bacause it probably takes care of the r-I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-443C
43-
.I 1
problem.
2 The committee I of NCRP is reviewing ~this 3
data right now to see what we should do about it, and 4
1 can't even tell you any more than that.
But at this.
S time we really thought we'd like to do it,.but we got 6
some static from our own counsel in terms-of whether 7
it was necessary at that time.
8 But we're saying here that one of the ways 9
in which you can assure that the lifetime risk to the 10
-individual is controlled is to put essentially a limit in effect assures a lifetime limit 11 on him that.is 12 of somothing in the order of 65 or 70 rem.
We felt 13 this was preferable to a set limit of 100 rem, which 14 would mean that at some age he gets-to 100 rem and.
~
15 he'd be out.of work.
This way, God willing, he's 16 always going to have:next year when he gets a year he would have, 17 older and he would be available 18 then, another rem available in terms of his working 19 lifetime limit.
But we did not make that a specific 20 limit on the worker but only in terms of how an 21 operation should be run.
I suspect that we'll move 22 perhaps strongly in that regard.
23 With regard to protection of the embryo 24 and fetus, in 1987 -hen we wrote -- when this document 25 was written there was a new concern about the fetal I
.J NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
44 I
- a 1
risk in terms of the bra *n. r, ental retardation problem, 2
and in fact the suggestion that it might have been n 3
threshold effect, a deterministic threshold effect.
4 (Slide)
And as a renult, if we could go 5
to the next slide, we did recommend two things, one of G
which was that the total to the fetus should be no 7
more than 500 millirem.
But a more important one and 8
perhaps more restrictive was that after diagnosis it O
should be limited in terms of 50 millirem in any <> n e 10 month.
11 (Slide)
The next slide and actually what I'd 12 the next two but the next slide shows 13 like to point out are some, if you like, secondary 14 considerations that were given in NCRp
'91.
But I
15 think it's terribly important that the NCRP considers 16 that we ought to be looking at this case of over-17 exposure not in terms of a worker and his health risk 18 but in terms of the operation and the way in which it t
19 resulted in his over-exposure.
20 So we're really saying.that an exposure in 21 excess of the limit probably doesn't affect his 22 lifetime risk at all.
If he gets 6 rem in one year, 23 it's certainly going to be offset by the less than 6 24 rem he gets in other years.
But more importantly it 25 suggests an inadequate system of protection and that NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
)
[
45 1
I his return to work should be almost entirely based on 2
improved control of the work
- place, unless the 3
exposures are accidental and very serious.
4 (Clide)
The next slide also is a--
5 constitutes one of the sections in NCRP
'91, which i G
tnink we need to give special consideration to as we l
7 think about lovering dose limite.
In 1987 we said 8
that we based our 5 rem on safe industries and we snid 9
that that's normally easily obtainable.
That's one of 10 the value judgments that's made is if we have safe 11 industries we ought to be able to accomplish that.
12 And we saw even then that it wasn't possible with 13 space flights, and in fact NCRP has done some work for 14 NAS/.
to develop different limits for astronauts 15 primarily on this basis.
16 And so we're saying that you could have 17 new limits bemed o ri informed consent of the workers 18 )
and a demons f rt.ted need.
There's nothing magic about 19 this as long as we say that safe industry is our t
r qL baels, and co i n f a<: t if you don't want to have a safe-20 21 industry criteria you could have something else which 22 you aight have for a eiven situation.
In fact, under 23 some conditions it might be reactor maintenance
~
24 workers that can't live by the dose 114tte end some 25 special requirements would be required, again focusing i
NEAL R.,
OROSS 1323 Rhode Island Avenue, N.W.
Washingtoni D.C 20005 (202) 234 4433
t-46 1
on the lifetice risk when you do this.
2 I'd like to shift now to the -ICRP 1990 3
draft, which Warren alluded to in his discussion of 4
the risks.
And I
ought to say that it's very 5
important whenever we consider radiation protection i
6 recommendations that we begin as we did in this 7
discussion with the biology.
You have to know what we 8
know about the risk before we can begin.
9 But in looking at this, in 1977 the ICRP 10 said that they were going to do this comparison with 11 safe industries.
They get concerned about two aspects 12 of this.
One of them was that, particularly for the 13
- ICRP, they're not uniform worldwide.
And secondly, 14 they arrn't unif3rm in time, as we've just seen.
And 15 they also were concerned about the fact that the 16 mortality data applies to the average, that is the 17 risk number, whereas the done limit applies to the 18 individual.
19 Now there is a little bit of a circular 20 reasoning on that, of course, because in that safe 21 industry there are also individuals who are at the 22 high end of the risk.
But the fact is we've set our 23 limit in publication 26 based on what we expect the 24 average to be, and in fact the Commission is concerned 25 about that.
i u
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
47
-r 1
(Slide)
So if we look at the next slide, 2
we see that in 1977 the limit was,udged ngainst the 3
average, but that in 1990 we also want to look at the 4
maximum.
That is, what is the what would the 5
worker -- what would his reaction be to the range?
i G
Obviously, if the average is going to be 7
safe industry there will be an, average, an individual 8
who will be, say, at some factor above that, perhaps 5 9
or so.
And when the Commission looked at this. they 10 looked primarily at studies done in the United Kingdom 11 in which they looked at perhaps some different words.
12 The English language has a strange set of
+
13 words with tolerable, intolerable, acceptable and 14 unacceptable.
They don't work well.
And so we really 15 see t h a t. unacceptable is probably right at the edge of 16 the unsafe industries, that is the deep sea divers, 17 the deep sea fishermen, and some.of the farming 18 industries.
Toloreble is probably closer to what we 19 called acceptable in the past, which is in the order 20 of 10-4 because of course even in the 10-4 industry 21 we're working all the time to reduce the accidents and 22 to improve safety.
So it's clearly not acceptable, 23 but tolerable in the sense that people will go to work 24 without -- yes sir?
25 COMMISSIONER REMICK:
Question.
Is that i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433 1
I 48 i
I 10-4 per year?
s 2
MR. MEINHOLD:
Yes.
3 COMMISSIONER REMICK:
Okay.
4 MR. MEINHOLD:
Yes, it is.
And acceptable 5
is probably c factor of t es n below that, 10-5, when G
people don't even think
- r. bout whether or not there's 7
any need to improve it.
8 CHAIRMAN CARR:
Did you also look at non-9 occupatirenal risks?
10 MR. MEINHutD:
Yes.
Yes.
And of course, 11 the driving back and forth to work is one of the 12 contrelling risks.
It's about 10-4 by itself.
13 (Slide)
If I could go to the next slide, 14 one of the things that comes out of the new risk 15 projection models is that it has to change a little 16 bit how we look at the basis for our limit.
If you 17 look at the lifetime
- risk, we see that using the and again, I take -- I 18 additive model it would 19 agree that we've got too-many significant figures--
20 but if we look at the 5.66 against the 8.56, we see 21 that, yes, the multiplicative model has increased our 22 expectation of lifetime -- of a probability of cancer.
to the 23 When we come down to the third 24 second
- line, I'm
- sorry, we see that the loss of 25 lifetime goes from 20 years under the additive model I
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
b 49 1
to 13 years under the multiplicative model.
And of 2
- course, it's because the relative model essentially 3
says you'll get the cancer at the same time you'd get 4
it naturally, which would be at older ages.
5 And so if you look at the next two, you-G see that loss of life expectancy, the difference 7
between the 1.12 and the 20 of course is that that.is 8
the average for everybody that gets exposed.
The 20 9
is for those people that die from the exposure.
So 10 there's a the 20 years is loss of lifetime if you 11 die from the attributable cancer.
So it's an 12 important thing to remember that maybe we can't look 13 just at the first line, but we have to look at some of
~
14 these other considerations since it's a multifaceted 15 problem.
IG (Slide)
So we look at the next niide, and 17 what the Commission did, the International Commission, 18 was look at these in terms of some suggested or trial 19 limits..
And among other things -- this isn't the only 20 data that was used, because of course we used the 21 morbidity data and some other considerations but 22 primarily they said let's look at what we get for 23 those different values and compare those with what we 24 had in 1977, compare them with safe industries and all 25 the rest of it.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
r s
i l
50 t
1 And so you see that there's a distribution c
2 there.
Obviously the loss of lifetime isn't going to 3
be very sensitive because that's going to happen when 4
you get the cancer, and the same is true of the 5
attributable age.
But the loss of life expet:tancy and G
the lifetime risk, of course, are going to change.
7
- Well, given those, the Commission said, R
well, our best judgement in terms of the value that 9
taken into account all of these attributes in given in 10 the next slide.
11 CHAIRMAN CARR:
Let me look, before you go-12 on to that one.
I'm not sure I understand.
What's 13 the unite on loss of lifettwe?
Years?
14 MR. MEINHOLD:
Yes.
15 CHAIRMAN CARR:
Okay.
And the loss of 16 life expectancy?
17 MR. MEINHOLD:
Is years.
18 CHAIRMAN CARR:
Is also years?
19 MR. MEINHOLD:
Right.
And the difference the loss of 20 between those, of course, the point 21 life expectnney would be if everyone who is exposed 22 the average weald be
.23, whereas the 13 would be, if who 23 you like the average, those who die from the 24 have the attributable death.
So it would -- the 13 is 25 the people that actually suffer the attributable r---
u-NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
i 51 F
1 death.
The.23 would be everybody.
2 CHAIRMAN CARR:
When I read it, is what 3
you're telling me what you intended to tell me?
I 4
shouldn't make any differentiation between the 10 and 5
the 50?
6 MR. MEINHOLD:
Well, on that basis, on the 7
basis of loss of lifetime and the most probably age, 8
but remember that the lifetime risk is greater, of 9
course, between the 10 and the 50, so for that one 10 you'd have to put some weight to it.
For the loss of 11 life expectancy, you
- see, it's about five times 12
-greater.
So it depends upon how you would weight --
13 CHAIRMAN CARR:
If I'm most probably going 14 to die at 78 anyway - -
- ~
15 MR. MEINHOLD:
Right.
I don't know whether I 1G CHAIRMAN CARR:
17 worry about it or not.
18 MR. MEINHOLD:
That is 19 CHAIRMAN CARR:
That's what you're trying 20 to tell me?
21 MR. MEINHOLD:
That's one of the things 22 that need to be considered when you're trying to pick that's my point.
- That, 23 out which of these values 24 among others, is what ynu would have to consider.
It 25 isn't just the risk.
It's also all of these i
NEAL R. GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
52
.i I
considerations.
2 CHAIRMAN CARR:
That's the most probably 3
age that I might die from cancer?
4 MR. MEINHOLD:
Right..
5.
CHAIRMAN CARR:
Okay.
6 MR. MEINHOLD:
Okay.
7 (Slide)
The next slide, then, gets us
- t. o 8
the values that the ICRp has recommended.
They 9;
recommended essentially 10 rem in 5 years, and 5 rem 10 in any 1 year.
The othur values, the annual. dose I
11 equivalent for t h e limbs,
the skin and the hands and 12 feet are based primarily on the deterministic effects 13 which Doctor Sinclair decided not to tell you about.
14 (Slide)
The next is the occupational 15 exposure of women, which is a very difficult problem 10 in terms of the social impacts as well as the fetus.
17 And the Commission has decided primarily that if 18 they're not -- for women that are
.w t pregnant they 19 will have the same as men.
There is no restriction on 20 what we used to call women of reproductive age or 21 capacity or
- whatever, and so because the 22 consideration is that the probability of a
fetus 23 surviving an exposure which would be detrimental in 24 that first two months or so probably isn't going to be 25 an impact on the woman.
c NEAL R. GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
p l<
I
$3 I
4 1
The second is that when the pregnancy has 2
been, if you like, diagnosed ared declared, then limit 3
the dose equivalent to the mother's abdomen to 200 4
millirem and tho fetus to -- it implies also the fetus 5
to something in the order of 100 millirem.
6 (Slide)
I've given on the next slide--
7 Just a quick one, although we haven't discussed the 8
basis the public dose
- limits, and one of the 9
reasons is that those haven't changed because we did 10 lower those.
Both the NCRP and ICRP have lowered 11 those to 100 millirem sometime in the last five or six 12 years.
The only difference for the ICRP is that it 13 allowed excursions to 500 with an idea that you could
~~
14 average it over the lifetime, and now we're saying 15 it's got to be restricted in terms of the time, that 16 you can average it to only five years.
17 (Slide)
If I could go to the r.e x t slide, 18 perhaps we could talk a little bit about the system of 19 protection.
I've tried to diagram here sort of the 20' conceptual way the Commission looks at exposures.
And 21 primarily we look at the fact that there is the source 22 of exposure; there is the environment through which 23 that exposure is received and the individual who 24 receives it.
25 Now if we look at those sources such as--
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
54 1
I've used accelerators, but it could be teletherapy 2
units or it could be steam generators or whatever if we look at that kind of a source, we say 3
else 4
that we can control that at the source.
We can 5
control it in the environment.
That's really, if you 6
- like, our occupational case where we can put more 7
shielding around it.
We can ventilate.
We can do 8
whatever else.
And we can control the worker.
We can 9
control his access, his training, and all the rest of 10 it.
11 If we come to th top one, we're talking 12 there about essentially remediation.
That is, the 13 source already exists.
We can't control the source.
14 It's in the ground or in the soil.
We can work with 15 the home.
We can improve ventilttion in the basement.
16 We can seal up the cracks and that sort of thing.
17 And we have very little control over the 18 public.
That is, we don't control who comes and goes 19 in the house and the rest of it.
So we have much less 20 control over the public.
In
- fact, as a
general 21 principle, the Commission believes that almost always 22 you need to control public exposure at the source, 23 whereas occupational exposures you can control all the 24 way through.
25 If we look at waste disposal, the reason I i
1 NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
4 55 I
1 put ths.t on here is because we have a new category of 2
what we call " potential exposures," that is exposures 3
which may not happen.
That could be a vault that s 4
been controlled by an interlock system and yoi.'re 5
asking about the failure of the interlock system.
G Here, you're talking about the failure of the waste 7
disposal system, that is intrusion or explosion or 8
whatever else, and there you again want to control-D that to the degree that you can at the source.
And 10 you can control it perhaps in the environment and at 11 the worker.
But
- again, as with the redon
- case, 12 whenever it's n public exposure it's much better done 13 at the source.
14 (Slide)
Now we have divided these up, if 15 I go to the next slide we _have divided them up in a 16 number of ways.
Clearly we've stayed with our 17 occupational, medical, and public exposure which is 18 the traditional way to divide up exposures.
But you 19 also see that we've
'done it in terms of these 20 classifications I've just discussed:
21 Practices or planned exposures, which are 22 the occupational or even the public exposure in which 23 you determine that you're going to release radioactive 24 material at a given rate from a facility; 25 Then there are the potential exposures, f
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
56 I
1 and I think these are very important areas where we're 2
looking not at what is being received but what the 3-probability of the release times the probability of 4
the detriment, and that would be a waste disposal 5
- facility, interlock
- facility, or in your case a
6 reactor accident of some kind; 7
The last one is the intervention, and here 8
is a
'n e w concept that the Commission is trying to 9
introduce in order to eliminate some of the confusion 10 in for instance Chernobyl or other places in which 11 radioactivity is in the environment.
It wasn't 12 planned.
It's not a potential exposure.
It's there 13 like the radon case in the home.
And so intervention 14 is a different matter than the first two.
practices 15 and potential exposure can be handled in terms of the 16 normal operating situation, whereas intervention is a 17 different beast, and that's how you need to be looking 18 at recovery operations, if you will.
19 (Slide)
If we go to the next slide -- the 20 slide I should have begun with if I was a loyal ICRP is to start with the three tenets, which is 21 person 22 justification, optimization of protection, and the 23 dose limits.
But these are the basic tenets of 24 radiation protection, optimization derived primarily 25 from the fact that we have adopted an assumption of a r-E.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202).234-4433
L i
57 I
1 linear response and therefore any exposure causes some 2
detriment.
But you don't want to spend unnecessary 3
resources and so you try to make sure that you have i
4 optimized protection looking at the economic factors, 5
the social factors and whatever else.
G (Slide)-
If I could go to the next slide, 7
I did want to show the difference between the way in i
8 which one looks at justification and optimization in 9
the intervention case.
And here the justification 1
10 isn't that I'm going to introduce a new source and is
]
11 it worthwhile to do it.
The radiation exists, and so 12 all the justification is is am I going to do more harm 13 than good excuse me, do more good than
- harm, 14 because anything else is a spurious question.
I menn,
' ~ ~
15 that's the only issue.
Can I help?
If I can help, I 16 do it.
17 And the second one, the optimization, is 18 merely that I do the best way I can.
It's-kind of a' 19 different concept.
And we don't believe that dose 20 limits apply, that new specific intervention levels 21 need to be derived.
Just as the radon levels in homes 22 in the United States, the intervention levels would be 23 much higher than any limit that we'd put on 24 controllable -- or the practices that would result in 25 people being exposed due to somebody else's i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
l l
1 58 I
1 activitles.
1 2
Well, I think that's pretty much the level I
3 of what I wanted to get into in the time we have 4
available.
5 CHAIRMAN CARR:
Thank you very much,
~
G Questions, Commissioner Rogers?
7 COMMISSIONER ROGERS:
On this last subject 8
of J us t i fi ca t i c;.,
which is something we've wrestled 9
with e great deal and I'm very interested in, how t
10 would you classify -- you've indicated how you'd deal 11 with the justification question -in the intervention 12 classification, but how about practices?
Ilow would 13 you deal with the issue of justification?
Ilow would 14 you approach that?
15 MR. MEINHOLD:
Well, let me say first that 16 it's not just a radiation protection issue.
It's much 17 broader than that.
And it really says that the public 18 good is going to be greater than the harm.
That's 19 basically what but that harm is all of the costs.
4 20 It's the cost of producing whatever it is you're going 21 to do.
It's the cost of protection.
It's the cost of
.=
22 the detriment.
You
- see, there are many costs 23 involved.
24 I think a good example that I've always 25 used for this is baggage x -ray machines in airports.
- r--
- i. -_
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
59 1
1 What is the justification?
Well, the justification is i
2 made by FAA and a few others and government who say 3
that it's very important to reduce the threat of 4
hijacking for either perceived
- reasons, for actual G
reasons, for the cost of airplanes.
As a protection i
G person, I don't need to know what their values are for 7
those.
All I
want to make sure is that they 8
understand if you like the detriment and what we enn s
9 do to reduce it.
10 Now once they decide that given a
11 reasonable estimate of what the exposures would be 12 that they're willing to accept that detriment along P
13 with all the other costs of introducing baggage x-ray, I
14 then I say to them now you must optimize it.
You must 15 now say, well, what's the total amount of collective 16 dose I'm going to get from this baggage x-rny and do I 17 add a flying spot x-ray machine which is going to cost 18 me $10 million to develop.
And we say,
- yes, you 19 should.
And that's the optimization part of it.
And 20 the dose limits part would still come in to be sure the 21 that no individual member of the public 22 constraint on that would be that no individual member 23 of the public, no matter how justified it is, should 24 exceed 100 millirem per year.
So that's the system, 25 if you like, applied to a very specific case.
I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
60 r--
1 COMMISSIONER ROGERS:
Yes, but there f..t e 2
of course many other possibilities that one might have 3
to consider.
And the question of justification 4
nometimes becomes really a very difficult debate as to 5
who's going to make that decision und on what banis 6
that practice might be Justified.
7 MR.
MEINHOLD:
I suppose it's the 8
licensing authority.
9 COMMISSIONER ROGERS:
I beg your pardon?
10 MR.
M E I N H01.D :
I suppose it's the 11 licensing authority.
12 COMMISSIONER ROGERS:
But on what basis?
13' MR. MEINHOLD:
All of the above.
14 COMMISSIONER ROGERS:
As soon as you start 15 to move out of a purely technical basis for decision-16
- making, you get into a
wide open arena that a
t 17 technical person is no more qualified to make that 18 decision than anybody else in some ways.
And it is a 19 very sticky issue and it is one that we havy wrestled 20 with.
21 MR. MEINHOLD:
And of course the question 22 is, is there a sort of trivial tae of radiation that 23 we would discourage even under your BRC philosophy.
24 And the one I
always use for my own 25 purposes is if a manufacturer decided he would paint I
u-NEAL R. GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
l 61 1
the eyeballs of dolls with tritium to make they glow 2
at night.
How would I buy that?
And some how or 3
other, no matter what the dose is, it seems to me to 4
be kind of a dopey thing to do.
And those are the 5
kinds of decisions I guess you're saying.
It's a 6
dopey thing to do, rather than I can calculate the 7
done and I can calculate the detriment and I'm going 8
to say what that is.
9 But I
think those are the kinds of 10 judgments that are in justification.
And as I said, 11 they are not -- I don't think they're at all radiation 12 protection alone issues.
They're much broader than 13 that.
m 14 COMMISSIONER ROGERS:
Well, we could talk
--~
15 on it all day.
IG MR. MEINHOLD:
Yes.
17 COMMISSIONER ROGERS:
I'm sure it is a 18 sticky issue.
19 You didn't discuss it here, but in reading 20 the transcript of your presentation to the
- ACNW, 21 Doctor Upton, I know at the end of that you really put 22 your finger on a serious problem with respect to who's 23 going to be around to do BEIR XV when the time comes 24 and what the sources of qualified people will be to 25 supply that need.
And I know that's not what we're i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
G2 r---
1 really principally concerned about here today, but I 2
wonder if you could say Just a little bit about how 3
you think that problem might be solved?
4 DOCTOR UPTON:
Well, Commissioner Rogern, 5
l'm not sure I know how the problem could be solved.
G I
think that we are seeing the disappearance of a 7
generation of people who've come into the radiation 8
research area following the World War II impetus from 9
atomic energy.
And unlike the situation in other 10 branches of science uhere there has been systematic 11 effort on the part of Unie Sam to support training to 12 produce people, I don't think that same intensity of 13 offort has 'een allocated to the problem of radiation
- r. _
14 protection, assessment of radiation risks, and the 15 development of alternatives and standards.
16-So my own view is that there really ought 17 to be a national study of the need for such people, 18 not just to rely on-Art Upton's top of the head view 19 of things but a systematic examination of where we 20 stand as a country.
How many people are in the 21 pipeline?
Are they enough?
What is the anticipated 22 need for such people?
And then, how do we produce 23 them if we do need them?
24 I think that just as after World War II 25 people came into this field because it was a new I
u _.
NEAL R. GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
63 I
F-~'
1 field, there was obvious need for effort, I think if 2
the nation identifies a national need, develops the 3
training programs, the career opportunities that would 4
be necessary, people will come into it.
5 CHAIRMAN CARR:
Before we run into too G
much of Doctor Morris' time, let's get Doctor Morris i
7 on the record here.
8 DOCTOR MORRIS:
Thank you.
9 The purpose of my pret entation today would 10 he to provide an overview of the process that's 11 followed in developing regulatory standards for 4
12 radiation protection and to provide a description of 13 the NRC involvement in that process at all stages.
r, -
14 The key message I want to leave with you
~
15 today I
think is evident here from what's been 16 discussed at the table, that this to a careful and 17 deliberate process that not only involves the 18 generation and development of basic scientific and 19 technical information but involves the development of 20 an international and national consensus on what kinds 21 of recommendations are appropriate for radiation 22 protection based on this information.
And then 23
- finally, it involves the development of regulatory 24 criteria that are reasonable, practical to implement 25 in the field.
I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
64 I
1 And so as we've looked at the evolution of 2
the information from BEIR V and grappled with the 4
3 implications of that for the BRC policy statement and
]
4 10 CFR Part 20, that mesonge I
think has to be I
5 recalled, that there are many steps before we would G
now take actions involving use of BEIR V
type 7
information in our regulations.
I think I'll try to q
e J
8 lay out some of those steps now.
9 I want to point out that one part of this 10 process is the development of basic scientific nnd 11 technical information.
The staff is involved in 12 monitoring that process at various stages.
For 13 instance, we were aware through the staff review of 14 technical information of the new dosimetry from the 15 bomb survivors and of the developments of 16 recommendations by UNSCEAR and the BEIR V and were 17 able to provide the Commission with recor.mendations on 18 how to deal with those early on.
We were able to 19 incorporate some of that thinking into the proposals 20 in the BRC policy statement.
21-But in addition to monitoring these 2?
developments, the Agency supports development of 23 scientific and technical information through research 24 programs, for instance.
We have programs who look at 25 the way -- how you would go about calculating the dose r-a NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washi.igton, D.C.
20005 (202) 234-4433
i I
i-t 65 f
I to the embryo / fetus based on the intak" to the 2
pregnant woman.
We have studies going on in 3
Brookhaven on the effectiveness of various specific 4
measures to achieve doses that are ALARA throughout-5 the world, and that's very useful research we think.
6 So the Agency is involved at the very underpinnings of 7
the development of this information.
8 In addition, as we look at.the development
?
9 of recommendutions and guidance by atandards writing E
10 committees such as ICRP and NCRP and others, the staff 11 monitors those developments carefully at the 12 professional staff technically and in the managerial 13 level to do that.
And in addition, we support those r.-
14 developments through individuals' participation in the
"~
15 various committees and through, in some instances, 16 funding of those committees.
17 For example, we had Dick Cunningham, who 18 works on the committee on applications of ICRP 19 recommendations.
We occasionally provide funding to 20 the ICRP.
There are eight different staff 21 participants in working groups from the NCRP, although 22 we have no member of the council
- itself, and we 23 provide general funding amounting to $150,000.00 per 24 year approximately to the NCRP to support its work.
y 25 We have had staff members participating in i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
66 l
t 1
the development of International Atomic Energy Agency 2
recommendations.
Important among those 3
recommendations were the safety guide '89 used as a i
4 reference point for developing the BRC policy.
We're 5
now staff is now working on a
rulemaking to G
incorporate into Part 71 the recommendations from the 7
IAEA on transportation.
8 We participate in the Committee on 9
Interngency Radiation Research and Policy 10 Coordination.
About ten individuals participate in-11 various committees or working groups of that 12 orgnnization.
he support CIRRPC at about $100,000.00 13 a year.
And I also mention here, although this isn't 1
- ~
14 something that the staff involves itself in, the work 15 of the Environmental Protection Agency developing r
16 generally applicable environmental standards that also 17 are adopted as part of our regulations.
18 The final point is that the step that is 19 taken historien11y before we would,
- say, adopt the 20 guidance of these _ committees is to develop federal 21-
- guidance, signed by the President.
That is done 22 involving various agencies working in committees 23 chaired by the Environmental Protection Agency.
24 Currently we are participating in efforts to develop 25 guidance on the protection of the general public.
I i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202)-234-4433
67 r-s -
I would imagine that at some point in the future, the P
2 EPA would convene a committee to deal with the newer i
3 recommendations of the ICRP and the NCRP.
4 One point I wanted to mention too is that 5
the standards that we're talking about are not just G
Part
.20.
There are a
number of places in our 7
regulat.ons where we would have what would be 8
considered radintion protection atandards, Pnrts 34, 9
35, 39, and in a number of regulatory guides.
So, all 10 of that is part of the pattern that we're talking 11 about.
12 (Slide) ao on to the next viewgraph.
13 A key part of what we do in developtag our r.,
14 regulatory standards and regulations is the evaluation 15 of operating experience.
One example of that is the 16 radiation exposure information reporting system that 17 gives us insights about the degree to which the 18 licensees are achieving doses that are ALARA and this 19 provides insight on the impacts of potential "O
reductions, for instance, in done limits.
21
- Also, for
- instance, our examination of 22 operating experience was able to focus on the issue of 23 hot particles and how we might want to look at 24 revising the skin dose limits in Part 20 in view of 25 that experience.
M NEAL R.
GROSS 1323 Rhode T-1and Avenue, N.W.
Washingicw, 9.C.
20005 (2 r;i 3 A-4433
n l
68 r--"
1 Based on all these elements, the basic 2
scientific and technical information, the 3
recommendations of various groups and the evaluation 4
of operating experience, then the agency identi fies 5
those areas where it believes it needs to modify ita G
regulations and that enn come either from petitions, 7
from directions from the commission or from 8
recommendations from the EDO to the Commission.
Ilu t 9
in each
- case, thero's a
deliberate step where we 10 decide that we do need to do something with 11 regulations to incorporate this evolving guidance.
12 When we have made that decision, we go 13 into the rulemaking process, of course, and there we 14 are involved in analyzing benefits and innnets as 15 exemplified by the regulatory impact analysis that we 16 develop for every regulation and regulatory guide.
We 17 try to develop these regulations through the public 18 comment process, sometimes involving public workshops, 19 to show that the requirements are reasonable.
20 inspectable, practical to implement.
In certain 21 instances, when we're trying to provide additional 22 guidance for broadly applicable criteria that will be 23 consistent with the dose
- limits, we will include 24 margins to address uncertainties that are involved in 25 trying to envelope a
large range of operating i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
69 I
situations.
2 So,
- again, as I say, this is a process 3
that we go through that involves a number of careful 4
steps and I
think that we have seen a kind of a 5
snapshot of where we are today with regard to the i
S ovolution of the response to the BEIR V,
UNSCEAR r
7 information and the development of recommendations by 8
these ICRp and NCRp.
9 CHAIRMAN CARR:
Thank you very much.
10 Go ahead, commissioner Rogers.
11 COMMISSIONER ROGERS:
I got started too 12 soon.
13 CHAIRMAN CARR:
- Well, I started you too 14 scon.
15 COMMISSIONER ROGERS:
- Well, I don't want 16 to take up too much time.
But it seemed to me in j
17 looking at some of the material in the BEIR V Report, 18 I
don't know whether it was the reporc itself or i
19 whether it was your presentation to ACNW, I seem to 20 recollect that you said something to the effect, and 21 please correct me if I'm wrong here, that some of the 22 data which you were able to extract with respect to 23 excess cancers you probably could never have seen if 24 you hadn't known there'd been a bomb dropped on that 25 population.
Those numbers were so small that if you NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433 i
)
i 70 i
1 hadn't known that there was an initiating event that 2
then gave you the impetus to make a study, that if I
3 you'd simply been studying that population and looking 4
at it, it wouldn't have told you that something had 5
happened.
i 0
DOCTOR ELLETT:
I think leukemia would
~
7 have.
The other things, that might be true.
8 Whnt do you think?
9 DOCTOR UPTON:
That's right.
I think as 10 Doctor Sinclair brought out, there have been roughly 11 G,000 deaths from cancer in the study population, 12 roughly 100,000 people in the two citien.
Using the 13 risk models that we've discussed, it has been 14 estimat'ed that there are perhaps 300 and some
'~
15 radiation-induced cancers out of that 6,000, of the 10 order of 5 percent perhaps.
17 Epidemiological tools are just not 18 sensitive enough to recognize that small an excess.
19 There's enough difference between different cities in 20 Japan so that an extra 300 out of 6,000 nouldn't 21 attract any attention except as Doctor Ellett points 22 out, that the excess of leukemia is large enough so 23 that one would wonder what was going on there.
But 24 the total cancer excess would not attract attention.
25 CHAIRMAN CARR:
Does the Japanese i
NPAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
I 71 r-1-
population risk of cancer
- deaths, normally is it 2
comparable to ours?
3 60CTOR UPTON:
Roughly comparable, yes.
4 Different uites, much higher risk of gastrointestinal 5
cancer, stomach, esophagus in Japan, higher risk of 6
colon and breast in this country.
7 CHAIRMAN CARR:
You had a comment?
8 DOCTOR S I NC I, A IR :
Well, I just wanted to 9
mention there's an additional point there.
It's not i
10 Just the total number of cancers, which is pretty 11 small, but it's highly dose related.
We know that the 12 distribution of those cancers across a done, and that 13 makes it almost incontrovertible that it was radiation l
l 14 that caused those feu.
I think it's helped us to see 15 it very much.
16 COMMISSIONER ROGERS:
Well, I'm not being 17 critical.
I'm just trying.to say we always have to 18 put this whole question, these health effect questions 19 in some broad context in which to understand them.
20 This whole question of when something cannot. be 21 measured in terms of its consequences, but one can 22 calculate it, how do you deal with that?
How do you 23 really practically deal with this kind of a situation, 24 which is what we're dealing with right now in things?
25 DOCTOR SINCLAIR:
Well, I think it's quite i
L. -
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
x,
- e 72
-.i 1
possible, Just to emphasize the point, that if there 2-had been a five percent increase in the population and 3
we didn't have any bomb to attribute It to, it would 4
be very difficult to see because-we don't know rates-
)
5-even in.dif.ferent parts of the' country that well.
But 6
1 think, as Doctor Silett's pointed out, the leukemia
]
W 7
rather speci fic and. the dose-related nature of the 8
other cancers does give us a great deal o f. con fidence l
9 that that's what they were due to.
}
l 10 CHAIRMAN CARR:
.Do you have that much I
11 confidence that you know what the dose rate was-for 12 those people?
13 l
DOCTOR SINCLAIR:
The dose rate?
14-CHAIRMAN CARR:
The doses.
ij l
15 DOCTOR SINCLAIR:
I have a fair amount-of f
16 confidence in the
- DS86, yes.
It has
- limits, of s
I 17 course.
But, you know, with the extra confirmation i
l 18 that we've
- well, first of
- all, DS86 is a
19 tremendously comprehensive evaluatioa.
- Now, there's 20 ten yecrs of dosimetry work. in there involving a j
21 considerable number of --
22 CHAIRMAN CARR:
But the exposure levels-23 those people were exposed to is kind of --
24 POCTJR SINCLAIR:
- Well, they range from 25 less than a rad to over 400.
We think we know those f
I NEAL R. GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
l b
73
- i l-numbers te the order of 25_to-30 percent now.
And we v__
2; have direct experimental confirmation, which we've 3
never had be fore, of the-gamma rays from measurements 4
of thermal luminescent materials that were-present at
-- S the site.
That was never available to us before 6
because the technique wasn't-sensitive enough.
It 7
confirms the calculations-which are fairly 8
sophisticated now to within about ten or 15 percent or 9
so.
10 CHAIRMAN CARR:
Excluding shielding that =
1l~
you don't know about.
12 DOCTOR SINCLAIR:
- Well, shielding is a 13 bigger uncertainty and it's a bigger uncertainty with s_
14 some-members of the sample that we haven' t l yet been 15 able to put into the sample.because we haven't: figured 16 out exactly how to have them reach the precision of 17 the dosimetry system -itself.
But we're doing that 18 slowly with various approximations.
19 CHAIRMAN CARR:
Do you want to comment on-20 that, Doctor Upton?
21 DOCTOR U PT0f. :
No, the area of dosimetry 22 in outside my field of expertise.
But I think, as has 23-been brought out, where we see the big effects are in 24 those small numbers of people who got the big doses in 25 close to ground zero.
There's no question about the i
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
g p
t 74 I
excess there.
'But mont'of the people, of course, were a.
2 out at greater distances and the ef.tect falls 1 off with 3-
' distance wi t h - dose, so that looking at the whole-4 population it's diluted out. 'You wouldn't see it, t.
= 5 But looking at the heavily irradiated s
.y.
G' population, there is a definite effect.
For leukemia, 7
many times' normal incidence, many times normal.
8 COMMISSIONER ROGERS:
I'll pass it.
9 CHAIRMAN CARR:
Commissioner Remick?
10 COMMISSIONER REMICK:
- First, I'd like to 11 add my; personal welcome to-such a distinguishe'd group.
12 and thank-you for coming in und spending time with us.
13 It's been extremely helpful.
y--
- 14
. !ho c t o r
- Meinhold, you touched upon 15 something I've been curious about-fr,r a.long time and 16 that is the dose to astronauts.
You point'out that' 17 they don't fit.the typ.: cal coccupttional mold.
Do you 18 know what a typical. case night be?
19' MR.
MEINHOLD:
- Well, I'll let Doctor he's been heavily involved in that 20 Sinclair 31 Committee activity, so-we'11'let htm.
22 COMMISSIONER REMIOK:
Okay.
23 DOCTOR SINCLAIR:
Well, we simply had to 24 have a different appronch to them because normally, of 25 course, we use for radiation protection occupationally NEAL R.
GROSS j
1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
p
.c
[..
k v
75 a
1 on the ground an annual limit to control the exposures 2
of the individuals.
It's not easy to. translate that
-3 straight into space for various different reasons.
4-They have missions rather than yeart.
In considering 5-
=the space
- station, for
- example, we wece already 6
informed that they were likely to have 90 day tou rs. of--
7 duty, as one example, i
8 We decided then that the only way to go 9
was to fix a career limit which would have a certain 10 percertage risk.
How do we get-at that?
We have -in 11 essence in worker populations three major groups:
12 shall we say the very safe industries working in
.13 offices and so on; the not so safe but very normal
. g_.
14.
o c c u p a t'i o n s ; and the least safe which are not normal-F 15 either, test pilots and all that sort of thing.
16 For various reasons, we considered that 17 the astronauts could not be put in the safest category 18 because considering the various risks of other kinds-19 that they had',
which were much larger, it would be 20 ridiculous -to limit them, absurdly.
On the other 21
- hand, you couldn't put them up in a
test. pilot 22 category for radiation either because they already did 23 have a big risk of a test pilot character going for-4 24 them.
So, we docided on an intermediate level which 25 worked out to be at that time three percent career e -
I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
76 I
I limit.
Then we decided as a function of age how much s
2 that would translate into a career limit and it worked 3
out at numbers from 100 rems upwards in terms of a i
4 i
career 1imit which, by the
- way, none of them have 5
approached yet.
They're not even close.
I think 6
around 10 rems or so is the highest --
7 CHAIRMAN CARR:
Do they wear dosimetry?
8 DOCTOR SINCLAIR:
ll e g your pardon?
9 CHAIRMAN CARR:
Do they take dosimetry on 10 the trips" 11 DOCTOR SINCLAIR:
Yes.
- Yes, they do.
12 We're facing considerations
- now, of course, of what 13 the exposures will be in a Mars mission.
That will 14 certainly be up there.
15 COMMISSIONER REMICK:
Has the NCRP done 16 anything with airline pilots, similarly looking at 17 risk and amount of exposure?
18 DOCTOR SINCLAIR:
We haven't.
We have a 19 study in prospect right now.
20 COMMISSIONER REMICK:
And one final 21 question.
What is the status of the ICRP draft 1990 22 report?
Where is that?
23 MR. MEINHOLD:
The expectation is that it 24 will be approved at the end of this year and probably 25 be ava;1able earlier next year.
I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
4 77
- r---
1
-COMMISSIONER REMICK:
Are you able to 2
predict any possible changes or do you think it will
'3 go as recommended?
'4 MR.
MEINHOLD:
1 think the-dose limits 5
will go pretty much as recommended and t'e reighting 6
factors as Doctor Sinclair, presented.
So, I think--
_7
'e way, I should point out this is the first time 8
ne ICRP has s e n t-its-documents out for 9
discussion.
Th,y went all over the.world, which had l
tremendous beneft which both Doctor Sincluir and'I l'
are very proud of because we beat pretty hard:an-the 12 Commission to allow that to happen.
13 The fact is most people who commented were 14 more concerned about clarity than they were about the 15 answers.
So, I think in general the final document 16-won't be terribly different in terms,of the numbers.
1.7 Hopefully it'll be a little clearer.
. 18 COMMISSIONER REMICK:
Thank you.
IS CHAIRMAN CARR:
Commissioner Curtiss?
20 COMMISSIONER CURTISS:
I just have a
21 couple of questions here.
22 Doctor Meinhold, your suggestion that we 23 take a look at occupational dose limit in the context' i
24 of the remaining age of the individual touched on a 25 concern, if I understand what you're saying, that I've E
u_
NEAL R. GROSS 132' 2hode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
I 78 i
I had in the past and that's the assumption that we as 2
regulators use in terms of the exposure of the 3
individual, in particular the concept of the maximal 4
(
exposed individual, full-time exposure, 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> a day 5
for a 70 year lifetime 6
I wonder if you could expand upon that i
7 i
concept and maybe touch on the notion of MEI, maximal 1
1; 8
y exposed individual, and what you see.
Is there an U
9 emerging consensus about the use of that conct pt or 10 the departure from MEI to some other more realistic 11 and accurate approach?
12 MR.
MEINHOLD:
I maybe don't follow you 13 exactly, but I think you're dealing with really what 14 the ICRP was concerned about in its old recommendation 15 in which it set a dose limit for the average -- based 16 on the average exposure, assuming a distribution of 17 exposed workers, and that that distribution would be 18 similar to safe industry distribution.
19 The new ICRP recommendation, and I think 20 the lifetime limit for ' ie NCRP are both aimed at 21 looking at that individual who might be exposed at the 22 maximum level over his working lifetime.
It's those 23 two considerations which lead you to what ICRP has now 24 said is a
limit of intolerability, which is th-
?S maximal exposed individual which would be reached at I
i NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
79
' r-
- l their consideration that at two rem per year, and what 2
NCRP-will be looking ~ at in terms of its present 3
guidance of 5 rem and to your age, which: again --Ethe 4
- 5. rem per year + allows the flexibility for any worker.
5 The-age limit' is this maximum exposed individual, G'
protected so.that over his working. life he-doesn't 7
exceed two or three percent-in te.ms of his lifetime 8
risk' of cancer.
9 So, I think they both are maximum exposed 10 individual concepts now.
11 COMMISSIONER CURTISS:
Maybe the question 12 that -I have goes to the more generic issue-of whether 13 the concept of the maximal exposed individual, seven
.w l
14 year exposure for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> a day from your perspective 15-is a reasonable one.
16 MR. MEINHOLD:
-I think it's unreasonable 17 but not -- that may be-wrong.-
I don't think it's very
_18 likely, but I think it's a possibility.
The thing 19 that I think that I would drive toward is that'for the 20 NCRP approach, that is some annual limit -- let us-use-
_21 5
since that's what's in
'91, that will normally 22 result in exposures far less than one rem to most of 23 the individuals.
I believe that there's normally a 24 distribution of exposures not just in the work force, e
'25 but in the worker.
That is, he will get most of the L. _.
NEAL R.
GROSS 3
1323 Rhode Island Avenue, N.W.
l Washington, D.C.
20005 (202) 234-4433
80
,s.
1 exposure from-age 23 to 35 = and then - he becomes
.a-2-
commissioner or something and doesn't~ have to get 3
_ exposure.
4 COMMISSIONER ROGERS:
That's a fact.
'S MR.' MEINHOLD:
bot the fact is that-I 6
think there is that distribution but it doesn't mean 7
that there aren't-a few guys ' out there who like to 8
clean steam' generators and they get a lot of money for 9
it and that's their livelihood.
When they look to 10 other alternatives, they find the) :an't make anywhere 11-near as much money.
~
12 Is it possible?
Yeah, I guess it probably 13 is.
It may be possible for -- well, just if the dose for instance, the ICRP numbers of two rem 14-limits 15 per year under some of the new aircraft that are being.
16 discussed by the industry, it may - well be that the 17 pilots will be a problem in terms of dose-limits.
And 18 they' could do it over their careers, from the time 19 they're perhaps 30 to_66.
20 So, I
don't know the answer to your 21 question.
22 COMMISSIONER CURTISS:
Okay.
23 MR.
MEINHOLD:
I don't think it's
'24 expected.-but I think there could be individuals and 25 the fact is that I think the flexibility of the five
- i. j
. a NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 3
(202) 234-4433
.w i
D 81 r---
4 -.
- 1 rem per year and the age takes' care ~of-both of-these-2 comfortably..
It means' you probably aren't-going to
~3-have any difficulty with - the five rem as a normal 4
limit for most individuals and for that strange 5
individual we're worried about, he'll be protected by-G the lifetime limit.
7 COMMISSIONER CURTISS:
Doctor
- Upton, 3
8 you've described the process that you've gone through
.9 as one that's continuing and i
suspect with the 10 additional data that will come i r.
will cont inue - to 11 examine these questions.
I wonder if you could s'ay a 12' word or two about the significance of t he-C hernobyl-13 health effects data and in particular the value of-14-that data from the standpoint of filling-whatever gaps 15 in -the information we have now and.that over the.
16 cour4e of the next several years we might find to be
.17 beneficial.
18 DOCTOR UPTON:
Thank you,. Commissioner.
19 Curtiss.
I think that if it is possible to quantify.
20 the doses to individuals'in tho area around Chernobyl 21~
or individuals involved in the cleanup o f' the 22 accident, the fire fighting and so on, then the study, 23 the long-term study of that population would be 24 scientifically useful.
Radiation was received over a 25 matter of days or weeks or even longer, unlike the r--
w -.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
f 1
82 1
I situation in Hiroshima / Nagasaki.
I understand that 2
there were substantial doses to a substantial number 3_
o f - people.
Perhaps the total collective populatton 4
dose approaches-that that we had in 5
Hiroshima / Nagasaki.
6 So, epidemiologically, I think it deserves 7-careful consideration.
I'm not as close tot the 8
. situation as Doctor Sinclair is.
I believe he's been 9-over and visited the site, talked to-people there and 10 his, pinion-is probably better informed than mine.
11 Warren?
12 SOCTOR SINCLAIR:
- Well, I
t,
_at the 13 present time the situation is not clear about what we 9"
- 14 can expect in the way of epidemiological studies from 15 the Russians.
It st'arted off with what was called an 16 all union center -in Kiev which was. to be set 'up to 17 look at all the likely groups and register some at 18 least 200,000 people who might have had more than
'19 minor exposures.
20 But since the political changes-in the 21 Soviet Union, the republics have decided' to--do these 22 things by themselves and not as an all union affair.
23 The center at Kiev has become for the Ukraine only.
there are 04' Byelo Russia, which is one of the other 25 three republics with hot spots in them.-
Byelo Russia i
I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
i L
83 r
1-
~ and Russia are setting up separate studies to examine 3
2 the effects in their own people from their own hot 3
spots.
4 And at.this point in time, I'm not ~ very 5
clear, and I'm not sure if anybody-is, about what the j
6 exposed groups have-been exposed to.
We thought at 7
one stage that it was a.very nice epidemiological-8 package, i f you like, in the people who were evacuated 9
from Pripyat and around Chernobyl.
There were about 10 24,000 of those people who seemed to have- ' average 11 doses of about 45 rads.
Not a bad little peckage, 12 But I understand those doses have since been revised =
13 downwards and that in the hot spots that have been 14 found in Byelo Russia and Russia, which were the 15 result of' rainfall because they're quite distant from 16 Chernobyl relatively, I. don't know what the exposures 17 are and I haven't got a. good record of them.
18
-So perhaps somebody does know, but its 19 not me at this point in time, I'm afraid.
I don't.
20 have too much hope that we'll get a lot from those 21 studies.
- 2. 2 COMMISSIONER CURTISS:
That's all-I have.
23 CHAIRMAN CARR:
Doctor Upton, the BEIR V 24 Report states that no increase in the frequency of 25 cancer has been documented in populations residing in i
i _.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
t
-84 i f-~
3 1
areso_of high natural background radiation.
To a non-2' specialist', there appears to be a logical disconnect
'3 between this statement.and the downward' extrapolations
't 4
of-the risk-associated with the-higher - radiation 5
doses.
Why-aren't the.results of the ' background G
s tudies. given greater weight-in developing these risk 7
coefficients?
8 DOCTOR UPTON:
- Well, the
- report, Mr.
9
- Chairman, does stress that the epidemiological data 10:
don't exclude zero risk at levels of natural 11-
. background.
The problem, I think, is that as one goes 12 to lower and lower levels of exposure one gets down 13 into -the noise level, if you will, and it's simply not'
'14-possible to distinguish effects that are so small.
15-There.has been a study of thyroid nodules 16 in women residing in a region of China where the
-l 17 natural background levels are elevated.and there--dose 18 to 70 years is estimated to approach something'of the 19 order of 9 rada.
Nine rads given to a child, that 20 leads to an appreciable excess of thyroid nodules and 21 thyroid cancers.
If there is no demonstrable excess 22 in-women in China, argues that at that low level 23
- exposure, rate-of
- exposure, the 9
rads over a
24 lifetime, the radiation is very much less effective, t
25 That's a possibility, as it's been brought out.-
We n.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
n, s
f 85 i
4 I
really don' t _ know as we go down t o.' lowe r an'd lower 2
dose rates that the risks will be
_as large as 3.
suggested by'the linear-extrapolation.
4 CH AIRM AN_ _ C ARR:
- Well, did the Committee 5
consider the epidemiological follow-up studies of 6
patients who received up to 50 R for thyroid-therapy 7
'in diagnosis?
8 DOCTOR UPTON:
Yes.
Yes.
That 9
information w e,s considered-There is a re ference1 to 10 the work by Hoel.
I believe that's the study that 11 you're referring to.
I think it's been brought out 12 carlier today that there are some suggestions in the.
13 epidemiological literature that protracted irradiation 14
-is less effective.
I think Doctor Sinclair said 15 perhaps by a
factor
-of four irradiation fro,i 16 radiciodine over a long-period of time could be, over 17 a lifetime, much less effective.than a factor-of four, 18 There are uncertainties there and I think 19 that was one of the reasons why the committee argued 20 that for highly protracted irradiation one -should 21 assume that these risk estimates-are likely-. to' be high 22 by a factor of two or more.
23 CHAIRMAN CARR:
Doctor Ellett?
24 DOCTOR ELLETT:
Mr. Chairman, I think the 25 Committee really discussed that thyroid paper in some E
t __
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202).234-4433
4 5--
4 86
-l' detail and their _ conclusions on thyroid cancer ' was 2
that it was really something that had to be studied, 3
that there was conflicting evidence on both sides.
4-I must point out that that study in Sweden-
'5 is by no means complete.
People have been followed a
..?
6; long time.
There have been.other reports quite
'7 similar from studies in the United States and when the 8
people _have been followed for 30 or 40 years, the 9
excess appeared.
It takes a long time.
But I don't 10 think-it's too safe to look at an early_~
114 epidemiological study ~and conclude there's no effect.
12 CHAIRMAN CARR:
Okay.
On one of your you-know, the 13 slides,'and I may have mis-looked it 14 press release when the BEIR V Report came out said the
' ~
15 risk from radiation, ionizing radiation, was four 16
(
times greater than previously estimated.
But one of 17 your slides kind of indicated that
.the' BEIR V
~
18 Committee was in the range of all the rest of the-risk 19 estimates from the I, II, III, IV committees.
i' 20 DOCTOR ELLETT:
I'm glad you asked that
'21 question.
The BEIR III Committee came out with a 22 preferred risk model based on a
linear quadratic 23' response that gave -- that was.the one newspnpers and 24 regulatory agencies alike used as the BEIR III risk 25 estimates.
The BEIR V Committee's risk estimates are
' I NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 t
(202) 234-4433
is i
87 i
e=
l-about three' to four times higher than that preferred 2
one.
If you go to the same model, you get about the
.. 3 same answer.
But the BEIR III committee used a linear 4
quadratic model.
5
- Now, I
think ' i t 's interesting that that 6
'model
- had, in a
- sense, a dose rate effectiveness 7
factor of 2 and a half built into it, as I believe 8
Doctor Sinclair pointed out.
If'you apply.a dose rate 9
effectiveness factor of-two, if you will, to the BEIR 10 V Committee's risk estimates, you'll find that their 11 risk estimates are higher by about a factor of two.
12 That might have-been a better way to put it' in the 13 report.
I don't know.
The. committee did this both 14 ways really and finally decided you should really 15-compare it to the preferred model.
That was the ene 16 people knew, not the highest number in the BEIR III 17 Report.
18 CHAIRMAN CARR:
Okay.
I guess I's trying 19 to figure'out whose slides these were.
I guess these 20 were yours, Doctor Sinclair.
On the one on risks of 21 cancer after one rad of whole. body irradiation, that 22 doesn't have anything in it on rate, dose rate of that 23 one rad?
24 DOCTOR SINCLAIR:
No, it doesn't.
It has 25 numbers over on the left-hand side which I'm --
w -.
NEAL R.
CROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
88 T-1 CHAIRMAN CARR:
It says an ennual risk.
2 DOCTOR 5INCLAIR:
-- not at all proud ot 3
because it's an old slide based on the old numbers and n
4 h I haven't redrawn it yet.
So, it's the shape of the l'
curve and it's time relationship that's important 5
l l
6 rather than the absolute value of the risk.
7 CHAIRMAN CARR:
Okay.
I guess I
8 understand.
U 9 ]
And Doctor Upton, given the large amounts I
10 4
of uncertainties in the risk associated with the low 11 doses of ionizing radiation, what research would you 12 think the Committee would recommend that we pursue to 13 attempt to reduce those uncertainties?
Is there
~
14 something we can spend our money on to kind of 15 eliminate that uncertainty?
1G DOCTOR UPTON:
Well, it's been brought 17 out -- Mr. Chairman, it's been brought out a number of 18 times elsewhere that more than half of the atomic bomb 19 survivors who were put into the study in 1950 are 20 still surviving.
Much of the uncertainty in our 21 estimates really relates to what's going to happen as 22 those survivors age.
23 CHAIRMAN CARR:
We've got to wait for time 24 then.
j 25 DOCTOR UPTON:
And we need to look at I
=
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 I
(202) 234-4433
m l
89 i
e_-
I human populations where there-is good reason to think-2 we can harvest information.
Reference has been made I
3 to the Chernobyl population, that there are 4
uncertainties there.
- Clearly, I
- think, it's very 5
important to cor,tinue to follow the A-bomb survivors 6
' meticulously for another 20 years.
That will helpLa 1
i 7
whole lot in narrowing the uncertainties.
But again, j
i 8
as you point
- out, that's extrapolation from 9
instantaneous exposure, relatively high doses 10 CHAIRMAN CARR:
- Woll, has anybody taken i
11 these guys who love to clean steam generators that l
11 Doctor Meinhold talked about --
-.i l13 DOCTOR UPTON:
I.think it's important to 14 follow the --
]i 15 CHAIRMAN CARR:
-- and watching that'where 16 we've got a good record of how much they've been 17 exposed to?
18 DOCTOR UPTON:
I think it's very important 19 to follow the occupationally exposed populations.
I 20 myself doubt that we can refine risk estimates very
{
21 much that way, but we can help to set upper limits.
22 If, as some allege, the BEIR V estimates are too low, 23 then studying radiation worker populations and cooling 24 data from different countries, that analysis ought to 25 help reassure us that these estimates are not --
u-NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
90 I
1 CHAIhMAN CARR:
Certainly we've got better 2
data on those people than we do on Chernobyl ~for 3
instance --
4 DOCTOR UPTON:
Indeed.
or even the bomb-5 CHAIRMAN CARR:
6 survivors.
7 DOCTOR ELLETT:
- Well, this is a place 8-where I
think the Commission could really help 9
' science.
We have a committee at the Academy that's 10 looking now at the study of U.S.
utility. workers for 11' radiation effects.
One of the major di f ficulties in 12 performing a study like this is. the availability of 13 data for the contract workers who get most _of the 14 dose.
~ This data is held by the Nuclear Regulatory 15 Commission.
It's in the computer.
The people at NRC 16
'have been very good about coming in and briefing us 17 about it.
But the problem is Commission's regulations
-18 do not allow the release' of this data for 19 epidemiological studies.
20 Health organizations like Health and Human 21 Services --
22 CHAIRMAN CARR:
You're talking to the 23 right guys.
24 DOCTOR ELLETT:
-- do.
This can be done.
25 It's just a question of looking at how the data could NEAL R.
GROSS l
1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
4 91 1
be released.for studies.
It's important.
2 CHAIRMAN CARR:
Okay.
Early on I remember 3
when the original Nautilus crew was-put under an 4
intensive program, all of them, for caternets.
We 5
were all-base-lined and logged and looked at for. years.
6 and they finally quit that because they weren t-o 7
finding anything.
But if we've got a good database to 8
start with, it looks like we ought to start following 9
it.
I don't know that anybody's doing that.-
10 MR. MEINHOLD:
Part of the difficulty,
'o f 11-
- course, is that the exposure ' in the industry-really-12 started in the middle
'70s, your large numbers' of
-13 people.
As Doctor Sinclair's slide
- shows, solid 14
' tumors take a long time to get started.
15 CHAIRMAN CARR:
All the more reason we 16 ought to start tracking them.
.17 MR. MEINHOLD:
- Yes, you've got to start 18 tracking them.
That's why. the study is difficult.
19 Even though' t here's a lot of exposure, there hasn't 20' been a lot of time.
21 CHAIRMAN CARR:
But a lot of us are beyond 22 that age-related routine where it starts.
23 Let's see, I think that finished up. my--
24 so you're not sure you can give us any specifics on 25 the research needs then that we need to dig into?
I w-NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
.t 92
' l; 1
DOCTOR UPTON:
Well, certainly the follow-1 2
'up of-irradiated human populations where the data can 3
be harvested, that's very important.
I think that 4
we're just on the. dawn of an enormous explosion. of 5
knowledge relating to cancer, getting to identify the' 1
9-6 genes that are
- involved, develop tools to detect 7
genetic change with high sensitivity.
8 So, I
think t '.i n t the epidemiologicar 9
studies need.to be paralleled by the studies in the 10 laboratory with cultured cells, with animal systems.
11 There's a spectrum of research, all o f which will help 12 to narrow our uncertaintles.
That's why I would think 13 that it would make sense for a national study to take
~
14 a look at where the science is, where the problems 1
15
- are, what is being done to address those problems.
16 Are they being pursued vigorously?
Do we have the 17 scientific talent we need?
If the answers are we need 18 more of this or that, then let's develop a national 19 plan to address those needs.
20 CHAIRMAN CARR:
Thank you.
21 Any other questions?
22
- Well, let me thank you all for this 1
23 informative briefing about the development of 24 radiation protection standards.
The mission of the 25 U.S.
Nuclear Regulatory Commission is to protect the I
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
93 r--
L-1-
public and the environment from potential hazards 2
associated with the uses of nuclear. materials.
3 Radiation protection standards -are an 4
Jntegral and essential.part of our regulatory program 5
to ensure.the protection of the p ub l i c..-
Your briefing
'y G
today has helped us to understand the process through 1
.7 which expert bodies like the National Research 8'
- Council, NCRP and the.ICRP derive estimates of the 9
risks associated with ionizing radiation'and formulate 10 recommended standards for radiation protection, 11 Your briefing has also highlighted the 12 many precautions taken throughout the process
.to t
13 ensure proper protection of the public.in light of the 14 uncertainties associated with the low doses and dose 15 rates typically associated with-licensed and exempted 16 nuclear activities.
17 Given the large. uncertainties that still 18 exist at these low doses, we need to continue to focus 19 our collective efforts on reducing these uncertainties 20 and on sustaining the scientific capability to improve 21 our understanding of the health. and environmental 22 significance of ionizing radiation.
23 I thank you for your presentations today 24 and for the careful and thorough work of your 25 organizations over the years in support of radiation i
u_.
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433
S 94 1.
l?
protection programs:of this Agency and the many users
-2 of. radioactive materials.
3 Unless there are additional comments, we 4
stand adjourned.
S-(Whereupon, at 12:04 p.m.,
the-above-t G
entitled matter was concluded ~.)
7 8
9 10 11,
i 12 4
r--
13 t
14
' 15-16 17-18
-i 19 20 21 22-23 i
24 25 E
NEAL R.
GROSS 1323 Rhode Island Avenue, N.W.
Washington, D.C.
20005 (202) 234-4433 T
CERTIFICATE OF TRANSCRIBER This is to certify that the attached events of a meeting
~
of-the United States Nuclear Regulatory Commission entitled:
TITLE OF MEETING: BRIETING ON DEVELOPMENT OF RADIATION PROTECTION STANDARDS PLACE OF MEETING: ROCKVILLE, MARYLAND DATE OF MEETING:
AUGUST 1, 1990 were transcribed'by me. I further certify that said transcription is accurate and complete, to the.best of my ability, and that the transcript is a true and accurate record of the foregoing events.
AAix.
- C+vt
' %J d
Reporter's name:
Peter Lynch
.,C e
4 e
NEAL R. GROSS COUef R$POtttt$ AND TRANSCRittR$
1323 RNOOG l$ LAND AVINUf, N.W.
(202) 234-4433 WASHINGTON, D.C.
2000S (202) 232-6600 Q
7/31/90 - reviseo SCHEDULING NOTES
Title:
Briefing' on Development of Radiation Protection Standards
-Scheduledi 10:00 a.m., Wednesday, August 1,-1990 (OPEN)
J
' Duration:
Aoprox 1-1/2 hrs
Participants:
SUMMARY
OF BEIR V RESULTS 30 mins
- Dr. Arthur C. Upton Chairman Department of Environmental Medicine New York University Medical Center
- Dr. William H. Ellett Senior Program Officer Board of Radiation Effects Research National Research Council DEVELOPMENT OF ICRP AND NCRP RECOMMENDATIONS 30 mins REGARDING RADIATION PROTECTION
- Dr. Warren Sinclair President National Council on. adiation M
Protection and Management
- Mr. Charlie Meinhold Division Head Radiological Science Division Brookhaven National Laboratory NRC Staff Dr. Bill M. Morris 10 mins
- Development of regulatory standards for radiation protection
e i
L THE BEIR V COMMITTEE A. Upton (Chair)
E. Hall D. Harti (Vice Chair)
D. Herbert B. Becker D. Hoel K. Clifton G. Howe j
C. Denniston S. Jablon j
E.Epp A. Kennedy 1
J. Fabrikant A. Knudson D. Grahn D. Thomas D. Preston, Scientific Advisor i
_ __ b a
~
~ _
~
m TYPES OF RADIATION INDUCED CANCER.
MODELED Leukemia Breast 1
Respiratory System j
Digestive System
=
All others (as a group) l
- t
/
<y.
--.3..
~
..._,.hy.,
.-f.,
..d,
~
+
-.x
+
a..,. - - - -.
s.
x..
s-
-. max
.--.a
.s n,
I '. I d[! '
d
- . d f'
~
w a\\ p_'-
1 1
'4 f
no I
.4o
~~s, n
,M,,
12
( 102% hear >
.is.
(
4 10 -
4
%s e
N c.
5 Q
22 -
%s i
%g%s('s eo ae l
2 u,
i
%g%
em
.o a.
en g.
sLW w
N g,
. X
- .~ m..
..m as m
Leukemia (+ 4.4%/ par) h I
i i
i i
i 0.5 1950 1965-1969-1963-1967-' 1971 1975-1979-1954-1968 1962 1966' 1970 1974
' 1978 1982 INTERVAL OF FOU.OW-UP
'j i
l
-1 f
i e
a e
f
- 1Hi$n$".#
s
--O--
~
I t
I
, ?
m:
e i
e A
s
~
%g O
g
=
t
% g 4
y P
h
(
g Yi t
I i
I I
I I
g" O
g a
g M
N w
o
.. i 4
4
- - - - - - - ~ ~ " ' ~ ~
^
~^
^^
ti i
$ dl.k. Q
~
^
~
paa E is g
tmh
- l wi>
l ao; l -- --- (m l
"? --- --..
)
10
'o a
1 h5
. Number of essee in the HI
' indiented does Irnervel g.
d i
i i
i i
i i
o 0,
1 2
3 4
5 6
7 8
9 MARROW DOSE EQUIVALENT (Sv) b 9-e 1
9 V
k) Y 7/ c4'.
t i
l h
e i
i e
!~
Solid Cancers 1
Did l...-
l
~
g fl 4
e 10
=
48
=
j It
=
4 im 54' 43 m
l.
15
=
tie 4G
=
i 9
g It s
S l
S 4
IIlllll1111111llllllllllll 1
- . - - -.. -. -. -. ~... -..... - -.
CANCER MORTALITY - ACUTE DOSE I
0.1 Sv to 100,000 persons of a given sex Male Female
~
Leukemia Solid Leukemia Solid 110 660 80 730 TOTAL 770 810 AVERAGE
- 8. x 10(-4) per rem 4
9'
a MORTALITY BY TYPE OF SOLID CANCER FEMALES Breast Respiratory Digestive Other 70 150 290 220 O
e
=
i e
l I
i i
TOTAL MORTALITY BY AGE AT EXPOSURE l
t FEMALES 1
5 1532 55 505 l
f I
15 1566 65 386 25 1.178 75 227 i
35 557 85 90 1-k i
i 9
l i
i l
I i'
I I
i L
TOTAL CANCER MORTALITY MALES 1
l i
l 1
rad per year age 18-65 i
i i
i l
2880 i
(2150-5460)
?
14% of normal exoectation Average years of life lost 1
1 15 years 1
per excess cancer I
j
~
t i
f 1
l
l 1
COMPARISON OF BEIR CTTES' RISK ESTIMATES l
t Linear. Relative Risk Models Lifetime Plateau (Cancer fata.lities per million person rem)
BEIR-1 1971 690 Early Deaths 1970 S. Pop
+
i BEIR-111 1980 500 Early Deaths 1970 S. Pop
(
BEIR-V 1990 790 Excess Deaths 1980 S. Pop l
l (1000)
(Early Deaths) l l
?
l l
f i
i i
f
= = -.
= - -. _ _.
RANGE OF CA RISK ESTIMATES BY BEIR COMMITTEES i
1 (Cancer fatalities per 1,000,000 person rem) l i
l i
BEIR-l 130-690 L (Additive, 30 yr.)-(RR, Life) l
}
BEIR-ill 10-500 QL (Additive)-L (RR, Life) i L
BEIR-V 400-1700
' 90% Credibility interval (540-1240) 90% stat. confidence interval i
=
s c
[
l h
.,-n.
w,,.
--n,
ESTIMATES OF GENETIC RISK I
i l
l cases per 1,000,000 liveborn (1
rem per generation) i First Generation Equilibrium Autosomal 6-35 100 i
(
Congenital Abnormalities
<10 10-100 Unbalance <1 I
i Translocations
<5 5+
L i
i Others
<3 6+
i t
+
=m
1 ESTIMATES OF CANCER RISK AND DETRIMENT-THE BASIS OF ICRP AND NCRP RECOMMENDATIONS Warren K. Sinclair l
President i
National Council on Radiation Protection and Measurements Presentation to the Commissioners of the Nuclear Regulatory Commission August 1,1990 j
I
^ i
1 l
i i
l Concerns in Low Dose l
Radiation Protection l
l Stochastic Effects e
No threshold e
Magnitude of effect same at all doses e
Frequency proportional to dose at low doses l
Hereditary Effects j
Induction of Cancer l
i Special Concem l
l e
Risk of Mental Retardation in Fetus
[
l Deterministic (nonstochastic) effects are not a concem because limits are below thresholds.
l t
Exposed Human Populations for Risk Estimation i
i A Bombs Medical Diagnosis Japanese Survivors Multiple Fluoroscopies Marshall Islanders Pre-natal Irradiation l
Thorotrast injections j
Medical Therapy Occupational l
i Pelvic Radiotherapy-Uranium Miners l
Spinal Radiotherapy (A.S.)
Radium 22s Ingestion l
Neck, Chest Radiotherapy (thyroid)
Scalp Irradiation (tinea capitis)
Breast Radiotherapy Radium 224 Treatment l
i 1
i
Risk of Cancer After 1 Rad, Whole Body irradiation 6
10 x 6 -
-'e
- s L
1 AR Other Cancers
)
y
- i
.9 4-w.
g g
a C
C 2-Leukemia "Ri$,l*
i g
- iMjh, O
s i
I i
1 O
10 20 30 40 50 j
Time After Irradiation / Years b
v s-
-- i-u
+
w w
. ~,
?
From ICRP Publication 26 l
1977 i
For the purposes of radiation protection j
involving individuals, the Commission j
concludes that the mortality risk factor for radiation induced cancers is about l
10-2 Sv-1 l
(This value is rounded from the value of 1.25 x 10-2Sv-1obtained from the average of 1.5 x 10-2Sv-1for females a.:d 1.0 x 10 -2 Sv-1for males)
Epidemiological Information Since 1977 e Update in Japanese A Bomb Survivors e Update in Clinical Studies such as Ankylosing Spondylitics in U.K.
e International Cervix Series e Other Clinical Updates, such as on Breast and Thyroid
i i
New Information on Cancer Risk in the Japanese A-Bomb Survivors j
(Since 1977-80) i
- 1. Three new cycles of Japanese data, 1975-78,1979-82,1983-85 l
2.
Increase in excess solid tumors from l
~100 to ~ 300 l
t 1
3.
More information on age dependence 4.
More information on time course 5.
Revisions in the survivor dosimetry
[
(DS86 vs. T65D) l i
1
(
Lifetime Risks in the Japanese (Preston & Pierce, RR,1988)
Population of All Ages l
Linear Extrapolation solid tumors 11WSv leukemia 1WSv Total 12WSv Nonlinear Extrapolation Total m SWSv l
(DREFN 2.5)
Adult Population Nonlinear Extrapolation Total
%3WSv (DREFN 2.5)
i 1
i UNSCEAR 1988 i
Probability of Lifetime Excess l
Cancer Mortality l
High Dose, High Dose Rate Population of All Ages (Japan)
Additive 4%/Sv f
.Multiplicative 11%/Sv i
Working Population Additive 4%/Sv i
Multiplicative 8%/Sv l
l Low Dose, Low Dose Rate i
l Divide by DREF.of 2 to 10 i
i
)
f i
c i
1 BEIR V 1989 j
l Probability of Lifetime Excess l
Cancer Mortality
]
0.1 Sv U.S. Population (all ages) 0.8%
1 mSv/y Lifetime Exposure (all ages) 0.56 %
l 10 mSv/y Exposure (age 18-65) 3.0%
i High Dose, High Dose Rate l
l Population of all Ages (U.S.)
9%/Sv l
t Working Population l
t Low Dose, Low Dose Rate 1
Divide by DREF of ~2 or more l
1.
1
+
l
\\
l
\\
8
\\ e,
'.5
\\ 1
\\
- 4, o
t
=
\\S
\\
9-
\\g
+
p o
N *+ \\
g e5 g
n
\\
W
\\\\
\\.
es\\\\
~"
\\
i L,
(3) /4?lepolN aeouso peonput 1
t t
Dose Rate Effectiveness Factors I
i UNSCEAR 1977 2.5 BEIR 1980 2.25 NCRP 64 1980 2-10 l
l NIH 1985 2.3 I
UNSCEAR 1986 Up to 5 UNSCEAR 1988 2 to 10 l
BEIR 1989 2 or more l
Human experience, breast, thyroid-
~1- ~ 3 Hiroshima-Nagasaki 1987-89 l
i Leukemia
~2 i
~1 i
Solid Tumors i
i 4
I l
I i
i
- +
l
i
.i ICRP Risk Estimates for Radiation Protection High Dose, High Dose Rate i
Population of All Ages
- Average UNSCEAR (mult) 11 %
10WSv Beir V 9%
Working Population j
UNSCEAR (multi?
8WSv i
Low Dose, Low Dose Rate Divide by DREF of 2 i
Population of all ages SWSv l
Working Population 4WSv
- and others i
1 1
~
t l
l Lifetime Mortality in 10,000 1
[
per Gy of Low Dose Rate l
Low-LET Radiation ICRP 26 ICRP 1990 i
Bone Marrow 20 70 i
Bone Surfaces 5
5 Bladder 25 Breast 25 20 Colon 80 I
Liver 15 Lung 20 80 i
30 Oesophagus 10 l
Ovary i
SWn 2
Stomach 105 Thyroid 5
8 l
Remainder 50 50 i
TOTAL 125 500 l
l f
t i
Tissue Weighting Factor (wT )
0.01 bone surfaces, skin i
0.05 bladder, breast, liver, oesophagus, thyroid, l
remainder 0.12 bone marrow, colon, lung, stomach 0.20 gonads l
f i
1.00 whole body i
i I
ICRP Detriment (Low Dose) oer Sv Population of All Ages 1%
Serious Hereditary Disease Cancer Mortality.
5%
1.5%
Cancer Morbidity
~
7.5%
Working Population Serious Hereditary Disease 0.6%
4 /o Cancer Mortality 1.2%
Cancer Morbidity 5.8%
3 1
I i
NCRP AND ICRP APPROACH TO i
SETTING DOSE LIMITS AUGUST 1,1990 i
i Charles B. Meinhold Radioingical Sciences Division i
Brookhaven National Laboratory Upton, New York l
l l
i l
n
/
A
!CRP PUB 26 (1977).
Coherent system of dose limitation recommendations Total cancer and genetic risk to the first two generations (stochastic ef-fects) must be compatible with the fatal accident rates in safe industries.
Deterministic or nonstochastic effects must be avoided.
2 s
9
2
'l i
- 4
! {
!i j
!!i!,
ii!1 4
e^_
.c s
y 7n~
r e
y
/
. fa v4 u
s S
0 m1 n
i 0<
s r
1 s
6
- =.
~;
/
+
+
~
1977 -1990
~
-industry becoming sater.
New Japanese survivor data.
New risk projection models.
Fetal risk concern.
7 ij i
I nd
7 x
.. 4 l
TRENDS WITH TIME IN FATAL ACCIDENT RATE l
1957-1989 q
Mean Rate Annual over period Change of l
(10*y")
Rate -
]
i l
All groups 182
-2.4 0.1 (SE) a l
Trade 75
-2.3 0.1 (SE)-
j i
Manufacture 98
-1.6 0.2 (SE)
Service 114
-3.0 0.2 (SE)
Government 128
-1.2 0.2 (SE) f i
l
.I
1.
TRENDS WITH TIME IN 1
. FATAL ACCIDENT RATE 1957-1980
]
Mean Rate Annual 1
over period Change of (10*y")
Rate i
1 Transport &-public utilities 362
-1.2 0.3 (SE)
Construction 677
-1.4 0.2 (SE) l
. Mines and quarries 940
-2.5 0.5 (SE)
Agriculture (1973-80) 561
-0.3 1.0 (SE) 9
)
f
-l
t
- i
[
. i j;i.f
.lt i
j, l
~..
R O
F ST I
M N
L O
I 1
9 I
OT T
N TA 7
R O
I 8
O UD 9
SRA P.
1 NUR 0
E OS E
1 R.
TO G N
I P
AP N U
J R
D XZ I
C N
N EI N
E O
~
M I
M O
C E
R 4
l
~'
OCCUPATIONAL-EFFECTIVE DOSE EQUIVALENT LIMIT Risk of 10-2 gy-1 (10 rem-1) d Average annual dose equivalent = 2.3 mSv (230 mrem)
Therefore,50 mSv (5 rem) 11
m OCCUPATIONAL DOSE LIMIT But:
Risk estimates are likely to increase.
Safe industries becoming safer.
12-m
~
..)'
3 1
OCCUPATIONAL ~ DOSE LIMIT i
.i First q
Discontinue (Age-18) 5 rem j
Second Stresses upper boundary nature of y
dose limit "It is only when the cost of further i
dose reduction is truly unreasonable that the11imit should be approached."
13 l
l i-
l 1
OCCUPATIONAL DOSE LIMIT t
Third Operations should be. aimed at j
keeping individual lifetime exposures in tens of mSv below his or-her age in j
years.
For example: Worker at age 50 should j
have a lifetime; dose equivalent of less than 500 mSvf50 rem).
l 14 i
f
=
PROTECTION OF EMBRYO / FETUS
- 1. Less concern.aboutl teratogenic effects.
- 2. New concern about the developing central nervous system.
8-15 weeks==.4 Sv-' (4 x 10-3 rem ')
-15
. llr
i
.t
- i ;
- i -
- !jj if I 1
?
t n
i
)
m e
r
~
SC
(
S
)
v U
m S
T e
m E
r F
5 5
/O s
(
6 1
v Y
. S s
i R
B m
o.
nh M
5 gt E
an io
=
d
.m l
r a
ey
. t t
o f n T
Aa c
.=
+.
< i ii e
~
s i
I I
L EXPOSURE IN EXCESS OF THE LIMITS:
! OCCUPATIONAL
- 1. Unlikely to affect lifetime risk.
j
- 2. Suggests inadequate system of protection.
i
- 3. Return to work based on improved j
control of.workplace.
1l 1
1 17 j..
)
i i
i
~.
=
LM c
SPECIAL DOSE EQUIVALENT LIMITS
- 1. 5 rem (50 mSv) based on safe industries.
- 2. Normally easily obtainable.
- 3. Exceptions, i.e., space flight.
- 4. New limits based on informed consent of workers and demonstrated need.
- 5. Focus on lifetime risk.
18 l
ame
p a,\\A-4 4
M J
A a
v---am4--*,4 e+
a.
AdmMn se 4
b
-4mA0, a
6-A-6
--=.v.4
--h la J
4 h
h t,
..I s
t
'4
. 9.
(:
i-t-
mums
- lC c.
g 1
\\
j om c
9 l..
I i
l-
= t I
,7 I
r 8
' (; <
- i\\ ;
, 1 4
Y
+
i' f
- l!: j !il l ! 1j ll r
! ;i
=
~
t r
u e
e b
h g
t a
E y
i r
e e r
R t
s nd ev e U
i u
s w a
S d
e h
d ot O
n rl i
N i
t r
.t o P
X e
so yt I
uw S E a
l f
ply d
T im pp L
s.
n I
ap MA h
r a
eo a
I t
LN i
f f t
s E O w
ai at s n ST n
u di I
m r
OA o
or yi tl s
f o DP i
e i
l r
s n a
U a
dt t
sl a C
p rn ro od u C
m a a d
dt m
O o
n s eiv l
C a n ei i
t o h hd 7;
Sc Twin 7
9 1
1 2
l; I
p 1
m 1
i t
CRITERIA l
1977 Limit judged against the average
- l 1990 Limit also judged against the maximum i
i
-(Limit of Tolerability) 1 Unacceptable 10-3 4
Tolerable 10 1
Acceptable 10-s I
21 i
a i
t
o-
~
=
i.
g MULTI-ATTRIBUTE (50 mSv) 18-65 y
Additive Multiplicative l!
Lifetime risk (%)
5.66 8.56 j
Loss of lifetime 20-13 j
Loss of life-1.12 1.11 expectancy-j l
Most probable age 68 77 of attributable death i
E i
i i
,[
~ ;
.=
=
+,
' ^
m.-
~
~
MULTI-ATTRIBUTE 1
ANNUAL DOSE FROM 18-65 3
10 mSv-20 mSv 50 mSv l
Lifetime risk (%).
1.8 3.6 8.6 i
Loss of lifetime 13 13 13
- Loss of life
.23
.46 1.11 expectancy-l t
Most probable 78 78 78
]
age of attributable death j
2a I
s l
~
u L
e.-
b b
RECOMMENDED DOSE LIMITS; Occupational L
. Effective Dose 100 mSv in 5 years j
50 mSv in any 1 year i
i l
Annual dose equivalent m i
i the lens of.the eye 200 mSv j
the skin (100 cm 1 500 mSv l
the hands and feet 500 mSv
- l 1
l 6
24 1
3
~. _
I
n,j" l
l1
,Iv e'
m 2
N o
E t
M.
tn v O
eS W
n am l
e v
F 2
i m
u O
o qt s
e E
a n
R ee e
U s
m m
o S
a do O
S d
t P
ib 52 X
ma t
E n
i s
L 'r )m a
L.
e A.
n t
g~
nhe N
e t r aom O
r n
T gm P
I 0
t A
o ee0 rh2 P.
N Pt(
U C
C O
Llllllll l
l)Il,lll ll i
1l1\\
e
+.s RECOMMENDEDLDOSE LIMITS Public Effective Dose 1 mSv per year, averaged over any 5 consecutive years Annual dose equivalent in 20 mSv the lens of the eye the skin (100 cm j 50 mSv the hands 50 mSv 26
.,g.
~
's :
"o.
a-NETWORK :OF HUMAN EXPOSURE SOURCE ENVIRONMENT INDIVIDUAL EXPOSURE.
Radon Home Accelerators Experimental halls Waste disposal Soil, water-l 27 i
~
4 k#@
l b
%'N/r/j\\e[Y#
/////f 64 IMAGE EVALUATION
\\/
gf/
TEST TARGET (MT-3)
// g, %
j *f[4'"'%A?
0 dh e
gy 4.
'&<*>9
'4 1.0 if 2 M yljL23 l,i
[5 Ele t.8
=
1.25 1.4 1.6 4
150mm 4
6" g %y
/f4%
4}#y>p,,,_,43f*;%
///
i 93 o
w-
~'d CLASSIFICATION Practices (planned)
[
x-ray facilities, accelerators i
Potential waste disposal facilities, interlocked
~
exposure facilities i
Intervention.-(pre-existing) radon in existing homes-post-accident, etc.
j 4
f 28 i
4 c:.
u.
e f
- ;" k ne-4 PRACTICES AND POTENTIAL; EXPOSURE-REQUIRFMENTS Justification of the practice.
j 1
i Optimization of the. protection..
j l
c Dose limit.
j
-]
i f
i c;..-
.,,...,,-_...__._.._____.____________________________________,___..___m_.
[
O-t E
L 1
l l
INTERVENTION 1
1.
- 1. Justification -:do more good than i
harm.
l
- 2. Optimization - form, scale, and L
duration.
l
- 3. Limitation.- situation specific,.
intervention levels.
30 4
m.
e-
,+-n y-
,,me.
4 e
..,.y.,,.~.
s 4
k i
BRIEFING ON DEVELOPMENT OF REGULATORY 4
i STANDARDS FOR RADIATION PROTECTION AususT 1, 1990 i
BILL MORRIS
Contact:
Bill Morris l
Phone: 492-3750 l
i i
l-
.t i
..~
=
it a
L DEVELOPMENT OF REGULATORY STANDARDS FOR RADIATION PROTECTION
- O
~ MONITOR AND SUPPORT DEVELOPMENT OF SCIENTIFIC-AND TECHNICAL INFORMATION O
MONITOR AND SUPPORT DEVELOPMENT OF
- GUIDANCE, RECOM4ENDATIONS AND STANDARDS INTERNATIONAL COMISSION ON f
RADIOLOGICAL PROTECTION l
NATIONAL COUNCIL ON RADIATION 3
PROTECTION INTERNATIONAL ATOMIC ENERGY AGENCY ENVIRONMENTAL PROTECTION AGENCY COMITTEE ON INTERAGENCY RADIATION RESEARCH AND POLICY COORDINATION l
FEDERAL GUIDANCE 1
_i 4
DEVELOPMENT OF REGULATORY STANDARDS FOR t
RADIATION PROTECTION ICONTINUED)
O EVALUATION OF OPERATING EXPERIENCE O
IDENTIFICATION OF POTENTIAL MODIFICATIONS
~ l TO REGULATIONS l
l O
MODIFY NRC REGULATIONS AND REGULATORY j
GUIDANCE.
ANALYZE BENEFITS AND IMPACTS INCLUDE MARGINS TO ADDRESS UNCERTAINTIES IN' IMPLEMENTING RECOPMENDATIONS I
ASSURE THAT REQUIREMENTS ARE REASONABLE, INSPECTABLE, AND PRACTICAL TO IMPLEMENT l
l 2
- , _ u,_.,,
~ _.._
k y,.-
g !
g
=
s
(
A 1
HEALTH-EFFECTS OF j
EXPOSURE 10 l
IOW LEVEIS OF.
~
~
+
IOXIZING
~
RADIA110X BEIR V r,
- 1 Committee on the Biological Etiects of Ionizing Radiations
- Board on Radiation Effects Research
- Commission on Life Sciences National Research Council i
t
?
NATIONAL ACADEMY PRESS Washington, D.C.1990 e
I lJ i*
l
+.
~
1 i.
. - - =
n o
(,,
t t
1?
\\c
?!.
i COMMITTEE ON THE BIOLOGICAL L
p EFFECTS OF IONIZING RADIATIONS
~
l
- ARTHUR C. UPTON, Chairman, Department of Environmental _'
Medicine, New York University Medical Center, New York, New York DANIEL L. HARTL, Mce Chairman, Department of Genetics, Wasnington University School of Medicine, St Louis, Missouri -
BRUCE B. BOECKER, Inhalation Tbxicology Research Institute, Albuquerque, New Mexico KELLY H/CLIFTON, University of Wisconsin Clinical Cancer Center,
.I Madison, Wisconsin CARTER DENNISTON, Department of Genetics, University of j
Wisconsin, Madison, Wisconsin EDWARD R. EPP, DMsion of Radiation Biophysics, Massachusetts General Hospital, Boston, Massachusetts JACOB 1. FABRIKANT, Donner bboratory, Univerisity of California, Berkeley, California DOUGl.AS GRAHN, Argonne National bboratory, Argonne, Illinois ERIC J. HALL, Radiological Research Laboratory, Columbia University, New York, New York DONALD E. HERBERT, Department of Radiology, University of South
- Alabama, Mobile, Alabama DAVID G. HOEL, National Institute of Erwironmental Health Science, Research Triangle Park, North Carolina GEOFFREY.R. HOWE, National Cancer Institute of Canada, Uniwrsity.
of Tbronto, Tbronto, Ontario, Canada SEYMOUR JABLON, National Cancer Institute, Bethesda. aryland ANN R. KENNEDY, Department of Radiation Oncology, P ; spital of the University of Pennsylvania, Philadelphia, Pennsylvania ALFRED G. KNUDSON, JR., Fox Chase Cancer Center, Philadelphia, Pennsylvania DUNCAN C.-THOMAS, Department of Preventive Medicine, University -
of Southern California Los Angeles, California i
DALE PRESTON, Scientifi: Advisor to the Committee, National Cancer Institute, Bethesda, Maryland e
9 m
,g 4
e 6
Y wT+,-
y 1
t e
' Nat'lonal Research Council Stat -
' WILLIAM H ELLETE Study Ditector, Board on Radiation Effects
-j Research, Commission on Life Sciences '
RAYMOND D. COOPER, Senior Program Omcer, Board on Radiation Effects Research, Commission on Life Sciences RICHARD E. MORRIS, Editor, National Academy Press F
= -'
~
Sponsor's Project Omcer WILLIAM A. MILI.S, Oak Ridge Associated Universities l
iv S
ne
~
n I
w e
}
'e s
.=
.g.
THE BkIR V REPORT The BEIR V report updates earlier estimater of the risks of somatic and i
genetic effects of low-level irradiation, taking into account new information gained during the-decade since completion of the BEIR III study.-
For estimating the risks of carcinogenic effects, the BEIR V ' Committee placed primary reliance on the a-bomb survivor data, since they provide the only coherent' body of information on the effects.of a wide range of reasonably well quantified doses of'whole-body radiation in persons of both sexes and all ages.
. For comparative purposes, data f rom other irradisted populations also were analyzed.
The a-bomb survivor experience was assessed with the use of a machine-readable database, obtained from RERF, comprising a total of 3399 records on survivors, grouped by age at exposure, time after exposure, DS86 dose, city, t.
l and sex.
The kerma categories in the records were replaced by corresponding,
. organ doses, and an RBE of 20-was assigned to the neutrons (other RBE values also were evaluated for their influence on the resulting risk estimates). The
[
records were truncated at 4 Gy, data for higher doses being considered of I
- doubtful reliability.
The rates of ncrtality f rom various forms of cancer were. analyzed in relation to age. at.rradiation, time-after irradiation, sex, dos 9,and other variables, with a view toward their compatibility with relai.1ve (multiplicative) li and absolute (additive) risk models. Parameter estimates for thse models were then derived from the data, and the results were evaluated for goodness of fit..
- To assure sufficient numbers of cases for adequate modeling, cancer deaths were ultimately combined into the following five diagnostic categories: " leukemia" l
1 L
1 l
i
l-
.j n'
(excluding CLL), *cencor. cf the n breast *, ' cancer of. the digestive system",
7
-" cancer of-the respiratory tract", and *other cancers".
Various exposure-time-response ~ models were fit to the data, using.the
-AMFIT program, which fits a general-form of " Poisson regression" model.
The l'
observed number of events'in each cell of the cross-tabulatann were thus treated as a Poisson variate with parameters given by the predicted number of events under the model 1.e., the product of the person-years in that cell times the fitted rate.
Although the BEInt III report included risk estimates for cancer derived by the additive projection model as well=as by the multiplicative projection model, the BEIR. V committee found neither model to fit the data unless modified appropriately for sex, age et irradiation, and time after exposure.
Given such modifications, the multiplicative risk model was found to provide a more parsimonious description of the data and to be less subject to error resulting f rom misclassification of causes of death as discussed below.
i Hence the BEIR V Committee's preferred risk estimates (Tables 1,2) were based on the'latter type of model, f
For a given radiation dose equivalent d in sievert (Sv) the individual's age-specific cancer risk y(d) was expressed as:
y(d) = y,[1 + f(d)g($)].
(1) where y denotes the age-specific background risk of death due to a specific cancer for an individual at a given age which also depends upon the' individual's sex and birth cohort (year of birth), f(d)- represents a function of the dose,d which is always a linear or linear-quadratic function; i.e.,-f(d)-
= a d o r f (d) = ' a d + a d '.
In general, the excess risk function, g($) was n
n observed to depend upon a number of parameters; for example, sex, attained age, age-at exposure, and time af ter exposure. The age-specific risk could also be 2
l modeled-as an additive risk:
y (d) = yo + f(d)g (S ).
(2)
Although both models gave similar results, as expected -- since tho' function g(S) was allowed to-depend on age, time,. ate. -- this would not have been the case if g(S) were restricted to having a constant value other than for sex and age at exposure.
'The models were fitted using maximum likelihood -- i.e., the values.of the 1
unknown parameters which maximised the probability. of the observed number of y
cases (the " likelihood function") were taken am the best estiscates, and, where.
I 3
applicable, confidence limits and significance tests were derived from standard large-sample statistical theory.
1
-It-was expected.that the form-.cf the background term might. vary considerably between popu1'ations at risk and would not be of particular interest in terms of radiation risk.-
Hence the committee chose not to model it, but rather to estimate the basaline rate nonparametrically by allowing for a large number cf multiplicative. rate parameters, as is of ten done when fitting hazard, models.to ungrouped data.
Each model was then described by the Committee in terms of " point" estimates of its various - parameters, their respective standard errors and significance tests, and an overall " deviance" for the model as a whole. Because of the extreme sparseness of the data, comparison of deviance to its degrees of freedom was not used as a tu of fit of the model; however, since differences in deviance between nested alternative models (pairs of models for.
which all terms in one model were included in the other) have an asymptotic chi squared distribution, with degrees of freedom equal to the difference in the l-degrees of freedom between the models being compared, this test was used to I
i 3
1 I
)
assess the improvement in fit ac a result of adding terms t'o the' dose-response function and used repeatedly by the committee to minimise ' potential over-specification of the models.:
Approximate confidence limits on-parameter estimates were constructed in
'l l'
I-1 l-the usual-way, by adding and subtracting the standard error times 1.65 (for.90%
1 confidence) or 1.96.(for 951 confidence). However, in cases where the Committee-had reason to believe that the use of a normal distribution to estimate confidence limits was not valid, it reported " likelihood based" limits found by searching iteratively for the parameter values that led to a corresponding l-increase in the deviance.
L L
For leukemia, the Committee's preferred model was a relative risk model l-(equation 1) with terms for dose, dose', age at exposure, time after exposure l
\\;
H (minimum latency of 2 years is assumed), and interaction effects.
Between L
individuals exposed before age 20 and those exposed later in life, there I;
appeared to be no ef fect of age at exposure but simply a different time pattern l
within each groups hence a simple step function was found to fit both groups:
1.
rather well (splines can be used to smooth the transitions when desired--e.g.,
in the calculation of probability of causation).
The parameters for the leukemia model were as follows:
_ f(d) = a d + a,ds n
( ' explS I(T < 15) + SsI(15 < T < 25)) if E f 20 (3) g(S)
= 4 exp(6,1(T e 25) + SsI(25 e T e 30)] if E's 20, where T is years af ter exposure, E is age at exposure, and ' the indicator function I(T < 15) is defined as l'if T I 15 and 0 if T > 15.
The estimated parameter values and their standard errors, in parenthesis, are i-a = 0. 243(0. 291), a, = 0. 271(0. 314),
l' 4
L l
L'
m:
"t Si - 4.885(1.349), As = 2.380(1.311),pi = 2.367(1.121), -
p = 1.638(1,321).
The standard errors for the dose-effect coefficients, estimated by means
' of the likelihood method mentioned above, are both imprecise and highly skewed.
For all cancers other than leukemia, a'10-year minimum latency was assumed
~
= this was done simply by excluding all observations (cases and person-years) less -
than _10 years af ter exposure. For purposes of overall nonleukemia cancer risk estimation, the ecmmittee simply chose to sum the' risks of the components of the nonleukemia cancer group (i.e. respiratory cancer, digestive cancer, etc.),
each of which was estimated by the models described below.
For. cancer of the respiratory tract (ICD 160-163): The Committee's preferred model was a relative risk model (equation 1) with terms as followsi.
f(d) = a,d i
g(S) = expl6 sin (T/20) + SsI(s)),
(4) where T - years after exposure and I(S) =1 if female, 0 if male, with o, =
- 0. 636(0. 291),0, = 1. 437(0. 910), $s = 0. 71)(0. 610).
Under the Committee's model, the relative risk for this site decreases with time after exposures i.e., the coefficient for time after exposure, -1.437, maans that the relative risk decreases by a factor of about 5 over the period of: 10-to-30 years post-exposure.
The Committee noted that few data are available, as yet, on respiratory cancer among those exposed as children, and that the relaties risk is 2 times higher for females (owing to their much lower baseline rates) than for males, although the observed absolute excess risks are similar.
When testing departures from a constant relative risk model, the addition of a parameter for time af ter exposure resulted in the greatest improvement in 5
f i
(
l 1
^
- l
~
'deser
'ng the-data, a finding consistent with the decreasing relative risk' observed in the - Ankylosing Spondylitis _ study (Da87), which - influenced the' Committee's choice of parameters. The inclusion of a parameter for sex did not -
greatly ' improve the model's fit to the data significantly but caused some
. improvement. There was no improvement when a-term for age-et-exposure was added to the regression model, its value being sufficiently close to sero.as,to haves y
no influence on the' estimated risk.
For cancer of the-f emale-breast (ICU 174), the.model was based on a
' parallel analysis of several cohorts.
The important modifying factors were found to be age at exposure and time after exposure.
The dependence of_'rlsk-on age at exposure was complex, doubtless being heavily influenced by the woman's hormonal and reproductive status at that time.
Lacking data on these biological variables, the Committee found that the best fit was obtained with g
i L
the use of an indicator variable for age-at-exposure less than 16, together with additional indicator or trend variables depending on the data set.
Both incidence and mortality models were developed.
Although these differed, the highest risks were seen in women under 15-20 years of age at exposure. Risks were low in women exposed at ages greater than 40, suggesting that risks i
decrease with age at exposure. Finally, risks were found to decrease with time l,
af ter exposure in all age groups.
The preferred model for breast cancer age-specific mortality (female only) was a relative risk model (equation 1) with terma as follows:
f(d) = a,d
' exp($, + ps1n(T/20+0s1n'(T/20)] it E < 15 (3) g($)
= 1 1
exp[$s1n(T/20) + ps1n*(T/20) + 0 (E - 15)] L f EL 15, 6
1
.i:
t
~-
m.
P k
sp t
'e,
?
where K is age at exposure and T is years af ter exposure' with oi =
- 1. 220)0. 610), b, ' = 1. 385(0. 554),0 s = -0.104(0. 004,0 3 = -2. 212(1.'376), pe =
-0.0628(0.0321).
For cancer of the digestive system (ICD 150-159), the most alanificant aspect of the LSS data was found to be the greatly increased risk (factor of
- 7) for those exposed under the age of 30, although the Cosunittee had no-explanation f or it.
There was no evidence. of a significant change in the 3;
relative risk with time after exposure.
The Consnittee's preferred model is as follows:
f(d) = o,d (6) g(6) = exp($,1(S) + cE) where Z(S). equals 1 for females and 0 for males and l
! O if E 3 25 1
k Os(E - 25) if 25 < E j 35 oE =
100s if E > 35
'\\
l l;
- with E age at exposure.
The estimated parameter values are ai
=
=
. 0. 809(9. 327), $ s = 0. 553(0. 462),6s = ' -0.198(0. 0628).
For cancers other than those listed above (ICD 140-209 less those listed above), the excess was found to contribute significantly to the total radiation-
-induced cancer burden. Finer subdivision of the group did not, however, provide
- sufficient cases for modeling-individual substituent sites.
When attempted, the models were unstable, resulting in risk estimates for which there was little -
L
. confidence. -The general group of "other cancers" was reasonably fit by a simple model with only a negative linear effect by age-at-exposure at ages greater than 10.
There was no evidence of either an effect by sex or by time after 7
1
y
- ,2 1
'l s
1 j
exposure.
f 1
The' preferred model is as-follows ~
j m
- f(d) = a, d '
\\
- g(6) = 1 if E e 10 and exp [0,(E - 10)] if E 3.10, (7) where E = age at exposure and ai = 1. 220(0. 519), S i = -0. 0464 (0. 0234).
- As concerns the influence of dose rate on the carcinogenic effectiveness of radiation, the BEIR V-Committee expressed the view-that low-LET radiation can.be expected-to decrease in effectiveness when highly protracted, possibly by a f actor of 2 or 'more for certain ne'eplasms, if the carcinogenic response of human tissues in the low-to-intermed ate dose range is consistent with that i
which has been observed in experimental inodel systems (NCRP, 1980 UN, 1986).
The Committee ref rained, however, f rom specifying a precise value for the DREF, :
except in the case of leukemia, where its preferred linear-quadratic modc1 (equation 3). contained an implicit DREF of approximately 2.
For heritable radiation-induced detriment (Table 3), the BEIR V Committee's risk estimates differ 'from those of the BEIR III Committee in including no y
allowance for effects on the incidence of-multifactorial' diseases, which were considered by the-BEIR V Committee to be too uncertain to quantify.
In other respects, the estimates of the two Committees were similar.
For radiation injury to the developina embryo the BEIR V Committee's risk estimates were larger than those of the BEIR III Committee, reflecting new information (Figs. 1-3) on the incidence and severity of metal retardation in prenatally irradiated a-bonb survivors, i
I i
p 8
l
'e
- ; '. j..
I l
Table'1.- Estimated' excess lifetime mortality from cancers of various organ systems-j af ter acute; exposure: to 0.1 RGy acute 'whole-body low-LET radiation, in relation to J
age at exposure ~and' sex (from BEIR, 1989)'
Males (deaths per 10')
Age at JExposure Total Leukemia
- Nonleukemla' Respiratory' Dimestive
'Other
.5
.1276
-111 1165
'17 361 787
~15 1144 109 1035 54 369 612 J
25-921 36 885 124 389 372
(35-566 62 504
-243 28 233 45' 600 108 492 353 22 117 55 616 166 450 393 15 42 65 481 191-290 272 11' 7-75 258 165 93 90 5
i
.85-110 96 14 17 LAvercge' 760 110 650 190 170 300 Females (deaths per 10')
dge at Exposure Total Leukemia
- Nonleukemia' Breast Respiratory Dimestly{
Other 15-1532 75 1457 129 48 655 625 51 5 1566 72 1494 295 70 653 476 25 1178 29 1149 52 125 679 293 35 557 46 511 43 208 73 187 45-541
.73 468 20 277 71 100 55 505 117 388 6
273 64.
45 65 386 146 240 172 52 16 75-227 127 100 72 26 3
85-90 73 17 15 4
-Averege' -
810 80 730 70 150 290 220
Based on a ' single exposure to radiation, and on a lifetable weighted average over each of th3 age 1 groups listed, in a stationary population having U.S. mortality rates.
I-l_
- Estimates for leukemia are based on the use of a linear-quadratic model and therefore include en implicit.DREF of approximately 2.0.
Estimates for solid tumors are based on the use of a linear model and therefore include a DREF of 1.0.
l
' Bas d on the sum of cancers of respiratory tract, digestive tract, breast, and other orgens.
L Values rounded to nearest 10.
9 l'
l
["
~
^
^
3: :n
.l.. <
g<
ll R '. Tablo* 2b Projected lifetime cancer mortality and associated loss of: life expectancy f rom L
. continuous whole-body' irradiation in a population of both.sexest(BEIR V, 1989).
z ;
7'
_ Exposure 1from Exposure Throughout Age 1-8 to Aga 65 Life (ImSv/yr)
(10 mSv/Yr)
Excess. Cancer Deaths No. per 10* _
56 320'
-X of normalE 3
16 fn '.
= Loss of Life Expectancy (Yr)
Average per person exposed 0.2 0.5 Average per excess death 17 16
-i r
3
.(Calculations based on. cancer and survival rates'for the U.S. population and on use of the risk models presented in Table 10, which include an implicit DREF of about 2.0 for leukemia
'and DREF of I'for solid tumors).
I u
e f\\
t r
?-
J l
i l.
l 10 i
1 5
C'k.
q i
+
. Table.3._
Estimated genetic effects of 10 mSv; (1 rem) per generation._
-R j
-Type of.
Additional Cases per-10' Natural Liveborn /rea/ Generation 4
Genetic Prevalence.
_First Generation Eaullibrium Disorder; (per 10',[
BEIR III BEIR IV BEIR III BEIR V Autosomal-
' d ominant
.and.x-linked 10,000 65 6 240-200.
100
'I Recessive 2,500:
<1
<1 very slow very slow increase increase i
Congenital 20,000-
<10 slight anomalies 30,000 increase
. Chromosomal abnormalities 4,400 (6
<1 Subtotal 70 40 200
-100 Hultifactorial 650,000 20-909 Total
~700,000 70 40-
~60-1,100
-100 i
t l-h
.~
I l
L j
11 l
l; J
1
+.
L 3
+,
4
..c l
e t
4 i
i 70 00
.a x.....
's
~
n' 50
'I W5'
/,
Nc.
m g 40
.,, n :,,..,,
3 o a.
xvs s
o vsg
',"./
$w so
./
wa
.7.J
/
w4 s
u t:
f r
?, O 20 s
aD
-./
'g
',*,, s :.
/
/
a s ser.,
10
,,.a"
,o O I
4 1- - -- J -
i' l.
3 0 010 020030 0 50 1 00 1 50 FET AL DOSE (G))
-i s
-1 Figure 1.
The prevalence of severe mental retardation among a-bomb survivors exposed in t utero, by. dose and gestational age, in Hiroshima and Nagasaki. The vertical lines-indicate 90% confidence intervals.
I')~
12 i,
Mf 't.
_ er 4.>.
T 1-5 m,,o o
- 130 120 m
en 7
~
no : I I,,,
J T-
- x. *t T,i>
w I'l
.t.', ]
1
-i-!
8 I
t u
l 4
y 4
h I
90 E-
- }.. m -.
~ ['
1 I
2-oi U
m
+ e.. c. i
.o W.-e c S M cel
(
yn,
ti
. h o n s ov W
w e
- 0i 0 9 0,
+ consn%
ALL AGES 07 8 15 1645 20+
4
- t
)
AGE IN WEEKS AFTER CONCEPTICN
?
s I
~
1 l
Figure 2.
Mean IQ scores and 951 confidence limits, by gestational age in weeks and fetal
' dose. in' a-bomb survivors exposed in utero in Hiroshima' and Nagasaki.
The numbers in parentheses are severely retarded cases, IQ f,64.
l i-I~
l r
t' l
13 1
. f.
ni I
i
. p [r_
' [' '
+
4
- i t,/-
. 0
..:< ; 1 l'
{
lL
+
p.
1
-21 :
e t
l' It l~
-i
{
f oO-$d g.-.
U a
.f.I
'h i s,,
lf[
- i",'
- g,, ho Ie J,
,l.
'3
, e'. i !,
i.
y
.At
}
, e j,,,
e
~
s
,j m-I-
a i
g A.
4 o
a l ;I to 3
o2 4
l n
3 -
. ' a %.
i y
,.a
.c e.
. y a s p 3.,, g
+
e aeasan,
. e a ai ew,
I
.., w. - 4, e n,,.
~
07 0 13 t ri ac.
,v,.
GESTATIOfJA!. AGE N NJ.51 k
1 i
i Figure 3. ~ Average school subject : score in the first grade with 95% confidence limits, by gestational age and fetal dose, in a-bomb survivors exposed in utero in Hiroshima and Nagasaki..
4 14
s-
- c... w -
) Jsp l*..'.
y i
,,1 c,
L ABSTRACT p
l-i I
BEIR V Estimates of Excess Cancer Mortality h
l W.-H Ellett, D. G. Hoel, and R. D. Cooper-4 L
The BEIR V committee i6itially modeled'ench of ne major types of cancer l
(l' ukeata, breast cancer, lung cancer, c'ncers of the digestive system, and a o
-c residual group *all other cancers") in terms of organ dose equivalent and a
('
full set of modifying factors (e.g. sex, age at exposure, age at risk, and l
L time after-exposure). Modifying factors which proved not to be important for a particular cancer site were eliminated in subsequent models.
In this way, a group of " preferred" relative risk models which include temporal factors were developed which can be used to calculate the risk of cancer mortality due to an exposure occurring at any age and time interval. This paper reports'en the time and age dependence of the committee's risk models.and presents risk estimates with their 904 confidence intervals for exposures occurring at various ages.
i e
[e r
0 1
~
n, +,.
a.
.. m 9WY)g $f h = = Cg y h
O'YfVO BEIR V Estimates of Excess Cancer Mortality W. H E11ett,* D. G. Hoe 1~,** and R. D. Cooper * -
Board on_ Radiation Effects Research, National Research Council, 2101 Constitution Avenue, N.W., Washington, D.C.
20418 USA
- Siometry and Risk Assessment Division, National Institute of Environmental Health Sciences, P.O. Box 12233, Research Triangle Park, N.C.
22709 USA ~
In 1986, a consortium of U.S. Federal Agencies asked the National
_ Research Council to organize a new Committee on the Biological Effects of
< - - c Ionizing Radiations (BEIR) to develop risk estimates de-novo for cancer due to-low LET radiation.(1)
The timing of this request was largely due to the assignment of new estimated organ doses to the individual A-bomb survivors in the hdiation Effects Research Foundation (RERF) Life Span Study.
In.
addition, 11 more years of survivor mortality experience had occurred'since collection-of mortality data for the 1980 BEIR-III report.(2)
In this time interval, the number of deaths due to cancer had nearly doubled.
The cohort of about 76,000 survivors having newly estimated doses was defined more rigidly than previously in that persons without significant shielding s
i.
information were excluded.
i While this truncation of the cohort largely affected-distal survivors who had small doses and contributed little,.if
.anything, to the dose response, it also excluded survivors who were self-e
- reported to be in the open but had not received flash burns.
In addition, L
persons who died at ages greater than 75 years were excluded from the committee's ' analyses as there is reason to believe the information on cause of
-death is considerably less reliable for this age group.
&& f&L u.
6 =- t Y
D L
l.
4
' The BEIR V Committer did not confine its data sources to: Japan but also considered data bases-for a numb'r of other epidemiological follow up studies e
carried out in Great Britain,- Canada, Israel, and the United States.
The availability of original data permitted the committee's risk estimates to be based on their own biometric analyses rather than che conclusions reached sw i
by other investigators. Moreover, a variety of source materials allowed intercomparisons of results from studies of different_ ethnic groups.
Nevertheless, with cne exception of breast cancer, the committee's risk
,.. c.
estimates of cancer mortality are largely based on the A-bomb survivor data (see footnote).
The Committee's Approach to Risk Modeling Preliminary examination of the A-bomb survivor data indicated that, with the exception of leukemia, the dose response for mortality due to all cancers combined was linear at organ doses below 4 Sv but flattened out at larger doses.
RERF investigators have attributed this decrease in effect per unit dose at high doses to large random errors in the doses assigned to heavily exposed persons,(3) but cell killing may also be a contributing factor.
Because interests have centered on effects at low doses, the committee decided to limit the data set to organ doses below 4 Sv.
This resulted in the exclusion of 2 cases of death from leukemia and 22 cases of death due to solid Results of tests made at various maximum dose levels are described cancers.
in Table 1 where it is seen that the coefficient for dose squara te nanntive-and not statistically significant for all solid cancers combined.
In contrast, the leukemia data showed a linear quadratia response when bone 2
=
,efl ' -, -;.'
.l i
~
i narrow doses greater'than 5 Sv were-excluded. At higher doses, the response for'laukemia decreases rather rapidly as has been observed in several animal' studies.
The committee fitted a number of dose-response models to the mortality data for all aolid cancers cymbined to test for the effect-of such temporal.
modifying factors as age at exposure, age at risk, etc.
Although.most of these models gave about the same estimates of lifetime risk,- none was really
- 2 '
satisfactory.
When all solid cancers were analyzed as a single group, the' goodness of fit was not impressive and many of the models appeared to emphasize the risk to those exposed as children.
When more detailed studies of cancer by site indicated there were often large differences among cancer types in their temporal responses, the attempt to model solid tumors as a group was abandoned. Although this move limited'the precision of the' final risk-estimates, it was a wise choice.
The average _ response over time often lc has little utility in formulating radiation control decisions or in deciding l
j on connensation for.possible radiation injury.
Nevertheless, modeling the dose response for individual cancers from a-
-limited data set also has limitations.
The committee fitted models to ten sites, or groups of sites, with the number of cancer deaths ranging from 34 to 2034.
Clearly, the larger groups produced more stable risk estimates.
In the end, a balance was struck between specificity and stability by modelinc, cancer 1
mortality. in just five groups:
leukemia, breast cancer, cancers of the respiratory system, cancers of the digestive system, and a residual group of "all other cancers "
While this is not an optimal grouping for calculating 3
^
4 e
i probability of' causation, it was considered to be the best that can be done at this time. No doubt, further follow up of the A-bomb survivors will allow greater specificity, s
Time and Age Dependence for Cancers at Specified Sites i
All of the models preferred by the committee are on a relative risk
. scale.
In general.:the radiogenic excess among survivors in the Life Span Study has increased ever time much like the age-specific baseline. rates among the. unexposed. ' Relative risk models could be fitted with weaker modifying factors for age and time than absolute excess risk models and, judged on the basis of deviance, usually yielded a better fit to the data.
Moreover, it is necessary to model the baseline cancer rate rather precisely when temporal factors are. included in absolute risk modele. The committee's experience 1
indicated this can be an appreciable source of uncertainty in the final risk estimates.
On a relative risk scale, the committee's general model for age specific cancer mortality is:
'If(d) -)f (1+ f(d)g(#))
o where 2f (d)- is the age-specific cancer risk for an individual of a given age, 2f is the age-specific baseline rate for a specific cancer, f(d) represents a o
linear or. linear quadratic' function of th'e dose equivalent, and g(p) depends l'
on a number of modifying factors such as sex, attained age, age at exposure, and *1:e cin:: cxpo.ur. ;iodels were ficceu using maximum likelihood.
For
- all cancers except leukemia, the best fit to data was provided by f(d) -4f d.
i Addition of a quadratic term, $(2d ' increased the initial slope as O(2 was 2
invariably negative.
In no case was (2 statistically significant.
4
{
Details of the committee's model are presented in the BEIR V report.(I)
Because the variation of g(#) with cancer types can be important in the design of efficient sampling strategies for dose ascertainment and risk projection, special emphasis is given to it here.
For cancers of the digestive system, I
the relative risk is higher for females than males.
g(#)- exp[# 1(S)+6 l 1
E where #1 is 0.55, E is age at exposure, and 1(S) equals 1 for females and zero for males. There is also a marked decrease in risk for those individuals t
exposed at more than 30 years of age, i.e. KE - 0 if E $ 25; 42(E.25) if 25 <
E s 35; and 10 42 if E > 35; where 42 is nearly 0.2.
Similarly, there is a decrease in the relative risk for increasing age at exposure in the group "all other cancers."
g(#) - 1 if E $ 10 and exp [# (E 10)) if E > 10 where #1 is 0.046 1
The committee's preferred model for cancers of the respiratory system difi'ers significantly from the usual model where the relative risk is modulated by age of exposure.
Such models invariably project a very large lifetime risk for those exposed as children.
Since A bomb survivors who were s
exposed as children are just now entering the age range at which lung cancer occurs, it is obvious that such projections are based on the experience of older persons nat those exposed as children.
In contrast to the usual model, the one preferred by the comalttee has no parameter for age at exposure.
In fact, when a term for age of exposure is added to this model, its maximum likelihood estimated value is so close to zero as to have no influence on the risk.
For cancer of respiratory system, the committee's model is:
g(#) - expl#1 n(T/20)+42 (S)}
1 I
5 t'
I.
i 1
J whe e T is years after exposure ahd~.I(S) equals 1 for females, and zero for males. With this model, the relative risk decreases rapidly with time after exposure, $1 -.1.44.
Of all factors tested, the factor for time after exposure produced the greatest improvement in fit, as compared to a constant 3
relative risk model.
This improvement, however, was not significant at a $4 confidence level.
The decision to include this factor was largely based on the decline in the relative risk of lung cancer observed in the Ankylosing
.. z:
Spondylitis study (5) Future follow up will, no doubt, refine the estimate of
$1 which has a large standard deviation, 10.91. While the factor for sex pl
.71, was also pat highly significant, it did improve the fit of cie model to the data.
The committee was fortunate in having four data sets for breast cancer mortality: the' A bomb survivors, irradiated postpartum mastitis patients, and Canadian and U.S. tuberculoses patients who received radiation I
to the breast for medical purposes. Although the exposure pattern for these cohorts ranged from high doses delivered nearly instantaneously to a rem or less every few weeks, there was no significant effact of dose rate.
In contrast, experimental studies of breast cancer in rodents show a large dose rate effect.
Obviously, care must be exercised in extrapolating animal data to radiocarcinogenesis in humans.
i 6
mr-h
[*
h The time pattern of radiogenic breast cancer is complex, in all likelihood, i
because of the age variation in hormonal factors in the population at risk.
The committee's model reflects this complexity in that age at exposure, i
E,and time since exposure, T,are strong modifiers of the relative risk.
g(4)-$(exp[pt+41n(T/20)4p3n(T/20))ifE4515 2
1 2 2
1 2 (exp [41n(T/20)+p3 n (T/20)+p4(E 15)] if E > 15 where pg is 1.4, #2 is 0.10, #3 is 2,2, and p4 is 0.062.
The, committee j
L modeled both breast cancer mortality and incidence but found no evidence for
~
an increased effect among those less than 10 years of age at exposure.
The major breakpoi.c in age dependence <ppeared to be at about 15 years of a3e, but this could not be tested in detail as the data available to the committee i
were in 5-year intervals.
In Figure 1, the projected breast cancer incidence pattern of U.S. women is shown as a function of attained age following an acute exposure, t
Leukemia sortality also shows a marked reduction of risk with both increasing age at exposure and time after exposure.
When appropriate temporal t
parameters are included, a relative risk codel fits the leukemia data as well l
as an absolute risk model.
contrary to earlier speculations, leukemia l
mortality in the Life Span Study population of A bomb survivors has not t-returned to baseline rates.
Therefore, ~a lifetime risk plateau is assumed in l
I the model.
In keeping with some recent RERF studies of somatic mutations in i
j lymphocytes, the committee found that dependence of relativa viek aa ge of exposure..E. changed at age 20, about the same age at which the thymus stops produ:*.ng T cells.
There does not appear to be any effect of age at exposure within the two age groups, E s 20 and E > 20 years, but there is a different time pattern for the expression of leukemia within,each age group.
7 1
l 1
t i
I f (d) - d 40(232 exp[pl+#2 (15 < T s 25)+S 1(T > 25)) if E 5 20 1
3 i
[ exp[pi+$4 ( T $ 25)+p5 (2! < T s 30)+A 1(T > 30)) if E > 20 I
I 3
where c(2 - 1.120(1.082); $1 is 3.442(1.016); $2 is 2.478(1.101);
- 3 is 4.857 (1.343); #4 is 2,492(0.870) and $$ is 3.218(1.115).
These equations are formally equilavent to the leukemia risk model given in the BEIR V report (1) but, at the suggestion of committee member Dr. Duncan
~"
Thomas, the excess risk in the period when the risk is highest, 2 15 years af ter exposure, rather than 30 years, is used as a baseline.
Expressing the r
BEIR V leukemia model in this new format has several advantages.
The estimates of the parameter values are more stable and inclusion of the coefficient of the dose term, A, in the exaonential provides a better 1
description of the skewed distribution of i':s possible values while also minimizing the correlation between the linear and quadratic terms. Although the step functions in the leukemia model may not be appealing on biological grounds, this model fits the A bomb survivor data considerably better than the leukemia risk model in the BEIR III report.(2) In this regard, it should be noced that the committee used a number of diagnostic tests to examine the degree of correspondence between a given model and the data on which it is based, see Annex 4F of the BEIR V report.(1)
The Committee's Estimates of Lifetime Risk The committee's risk estimates take into account only the excess number of deaths due to cancer, i.e. the difference between the number of deaths in an exposed and an unexposed lifetable population.
In contrast, the 1988 i
UNSCEAR report (6) and the BEIR III report (2) calculate the lifetime excess risk somewhat differently in that the difference in cause specific death rates 8
l-between an exposed and unexposed population is applied to persons alive at a L
given age in an exposed lifetable population.
In this way, the detriment due to_ fatal radiogenic cancers that occur earlier than nonradiogenic cancers is i
also taken into account.
Both the excess death and the early death method are correct; the best measure of the detriment is probably somewhere between the two estimates they provide. Vaeth and Pierce (6) have pointed out that to a good first approximation the numerical difference in lifetime risk for these two methods is: early deaths equal excess deaths divided by 1.P, where po is o
. x:
\\
probability of death due to cancer in an unexposed population having the same age distribution. Tor the 1980 U.S. stationary population used as a baseline in the BEIR V analyses, po, averaged over sex, is 0.202.
The BEIR V committee considered three exposure regimes in their lifetable calculations of lifetime risk: acute exposure, lifetime exposure from age O to 100 and continuous exposure from 18 to 65 years cf age. Table 2.
r These calculations assume the same dose equivalent to all organs.
The 904 confidence intervals for the point estimates of excess mortality in Table 2 were obtained by Monte Carlo sampling of the probability distributions of the parameters in the committee's equations for relative risk and over 10,000 successive lifetable calculations, one for each set of parameters.
The confidence intervals for leukemia in Tab'le 2 (and Table 3, below) are based on the renormalized model discussed above.
The statistical uncertainty of the lifetime risks includes only that part of the uncertaintv due to samnling variation. Other sources of uncertainty considered by the committee were:
model misspecification, dosimetry errors, populations differences, and the
.possible effect of sex where this factor was not part of the model.
It was judged that these factors would increase the 904 confidence intervals by a 9
h i
factor of about 1.4 Dose Rate Effects i
The risk estimates for solid cancers presented in Table 2 do not include a dose rate effectiveness factor (DRET).
In some cases, depending on the LET of the radiation, use of such a factor may be necessary.
The committee did not believe a DREF was justifiable for acute doses of either I
high or low LET radiation. However, for exposure to. low LET radiations at low I
doso rates, the committee thought that the risks listed in Table 2 would be
..e lower. Unfortunately, they were only able to quantify the DREF for low LET radiation as being in the range of 2 to 10.
It is their belief that at this time there is no good scientific basis for selecting a particular value for solid cancers in humans.
It should be noted that in the BEIR III report,(2) all of the linear quadratic based risk estimates for solid cancers contained an implicit dose rate reduction factor of 2.5.
This single factor accounts for most of the differences between the BEIR III committee's estimates of I
'i cancer risk based on relative risk and those of the BEIR V committee.
Because the BEIR V committee's linear quadratic equation for leukemia risk implicitly includes a dose rate reduction factor of 2, the committee thought that no additional dose rate reductica factor was applicable to the risks estimates for leukemia.
Lifetime Risk by Age of Exposure and Type of Cancer Even thour,h the committee's models show less variation of the risk with age at exposure then constant relative risk models, considerable differences between age groups remain evident. This is illustrated in Table 3 for the acute exposures. As indicated in this table there is still considerable 10
_,__-----u
r
~
statistical uncertainty in the risk estimates for those exposed at less than 20 years of age.
t The division of the risk of cancer mortality between types of cancer for exposure from age 18 to age 65 is shown in Table 4 Because children are not included in this group, the relative prevalence of the various types of cancer differs from that given in the BEIR V report (1) for exposures occurring at all ages. The estimated risks for acuce occupational exposures between the ages of 18 and 65 are large; a 10 mSv dose equivalent per year (one fifth of the
.. =
legal limit) throughout a 47 year working life increases the baseline cancer risk by about 14% for U.S. males and about 17% for U.S. females.
Vhen the difference between excess death and early death are taken into account, the BEIR V committee's risk estimates for acute exposures are not much different from those given in the 1988 UNSCEAR report (7) which are based on a constant relative risk model.
Early deaths for the UNSCEAR model yields 4
1,070 deaths per 10 person Sv for the 1982 Japanese population, while for the i
1980 U.S. stationary population, BEIR V yields 1,000 early deaths per 104 person Sv.
In contrast, the current ICRP estimate is ten times smaller, 100 deaths per 10' person Sv.(8)
It is apparent that the ongoing reevaluation of radiation riske by the ICRP is well warranted.
Acknowledgement Members of the BEIR V committee are Arthur C. Upton, Chairman, Daniel L.
f Hartl Vice Chairman, Bruce B. Boecker. Kelly H. Clifton, Carter Denniston.
Edward R. Epp, Jacob I. Fabrikant, Eric J. Hall, Donald E. Herbert, David G.
Hoel, Geoffrey R. Howe, Seymour Jablon, Ann R. Kennedy, Alfred C. Knudson, Jr., and Duncan C. Thomas.
Dale Preston served as Scientific Advisor to the committee, 11
1 REFERENCES i
1.
Committee on the Biological Effects of Ionizing Radiations.
Health i
Effects of Exposure to Low Lavels of Ionizing Radiation BEIR V.
National Academy Press ISBN O 309 3995 9, 2.
Committee on the Biological Effects of Ionizing Radiations.
The Effects on Populations of Exposure to Low Lavels of Ioniting Radiation: 1980.
National Academy Press ISBN O 309 030951.
{
3.
Pierce, D. A., Stra's, D. O and Vaeth, M.
Allowing for Random Errors in Radiatior, Exposure Estimates for the Atomic Bomb Survivor Data. RERF TR 2 89.
Radiation Effects Research Foundation.
Hiroshima (1989).
4 Shimizu, Y., Kato, H. and Schull, V. J.
Studies of the Mortality of A Bomb Survivors. Radiat. Res. 121, 120 141 (1990).
...c.
5.
Darby, S. C., Doll, R., Cill, S. K. and Smith, P. C.
Long term
)
Mortality af ter a Single Treatment Course with X rays in Patients Treated for Ankylosing Spondylitis.
Br. J. Cancer 55 179 190 (1987).
I 6.
Vaeth, M. and Pierce, D. A.
Calculating Excess Lifetime Risk in Relative Risk Models.
RERF CR 3 89 Radiation Effects Research j
Foundation Hiroshima (1989).
1 7.
United Nations Scientific Committee on the Effects of Atomic Radiation: Sources Effects and Risks.
1988 Report to the General Assembly. United Nations ISBN 92 1 142143 8.
- 8. -Recommendations of the International Commission-on Radiation I
Protection. ICRP Publication 26.
Pergamon Press (1977).
l I
l I
l
! 1
\\
. ~.
1 4
I i
t I
r t
c h
r i
t Footnote on Page 2) Diskettes of the ~ data from Life Span Study 9(4) as provided by RERF to the BEIR V committee are available from the Editorial Office, Radiation Effect Research Foundation, Hiroshima, 732, Japan (FAX 082-263 7279) 13
E TABLE 1 EffectofExcludingHighDoseg(ioup on the Fitted Dose Response Relationship.
Source: BEIR VL l
Exclusion Linear Dose Coefficient Test for Adding (SV) per Sv a Quadratic Tera
- Laukemia
- t-None 0.58
.08
>5 0.76 1.76
>4 0.48 2.14 '
>3 0.25 1.45
.c
>2 0.05 1,49 Solid Cancers None 0.78 2.04 0.82 1.88
>4 0.97 0,41
>3 0.98
.o.31
>2 1.14 0.66 aSigned square roots of Score Statistics for a test of the null hypothesis (no effect).
These statistics are asyntotically distributed as standard normal deviates.
3 m
14
+
m
. m
'.(..
P r
TABLE 2 Excess cancer Mo{)tality Lifetime Risk per 100,000 Exposed (*)
Source:
BEIR V(
Hales Females All Organ Leukemia Solid Cancers 14ukemia Solid Cancers Dose Equivalent 0.1 Sv 110 660 80 730 All ages (60 350)b (420 1,040)
(50-230)
(550 1,020)
Acute 1.0 mSv/yr 70 450 60 540 All ages (20 230)
(320 830)
(20 180)
(430 800) continuous 10.0 mSv/yr 400 2,480 310 2,760 Age 18 65 (150 1,200) (1,670 4,560)
(110 980)
(2,120 4,190) aThese risk estimates do not include a dose rate reduction factor, see text.
b904 confidence interval 1
l h
1 L
J e -
[7 l
i y
T TABLE 3 Excess Risk Estimates and 906 Confidence Intervals with'the Preferred Models.
(0.1Svacutegy)posureto 100,000 persons of each age and Source: BEIR V(
' sex).
Males Age All Solid Breast at Exposure 14ukemia Cancers cancer
.e 5
111 (33 655)*
1165 (673 1956) 15 109 (35 640) 1035 (642-1775) 25 36 (16 215 885 (534 1442) 35 62 (33 221) 504 (272-947) 45 108 (60 315) 492 (257 863) 55 166 (90 399) 450 (217 815) 65 191 (97 436) 290 (137-572) 75 165 (84-380) 93 (38-233) 85 96 (49 221) 14 ( 5 44)
Femalesb 5
75 (24-455) 1457 (1001 2114) 129 ( 28 440) i 15 72 (24 434) 1494 (1051 2095) 295 (112 652) 25 29 (14-179) 1149 ( 809-1527) 52 (10-144) 35 46 (26 182) 45 73 (42 214).
511 ( 315 819) 43 ( 9 132) 468 ( 277-776) 20 (
3-108) 55 117 (66-292) 388 ( 221 616) 6.(
8 71) 65 146 (76 333) 240 ( 134-426) 0 ( 38) 75 127 (64-291) 100 ( 53 197)
-1 (
3 13) 85 73 (37-168) 17 (
7-40) 1 (
2 2)
?
- (5%, 954) 200 replications.
.bUnpublished committee calculations l
L 1A l
L l'
t TAB 12 4 ExcesscancerMogtalitybySitefor 100,000 persons exposed between age 18 and 65.*'
10 mSv per year to all organs laukemia Respiratory Digestive All Other
, Breast Males 394 1120 468 844
..a 1
Females 306 935 945 713 132 aThese risk estimates do not include a dose rate reduction factor, bUnpublished committee calculations.
=
17
'1
- e.
.. ~
j a
n
.i. :
j l
i
.. :a.
.)
f 1
I 9
t FIGURE 1.
The projected incidence by age at exposure and attained
. age of radiogenic breast cancer f:dioving an acute dose of low IIT radiation.
for women exposed after age 45.Incidenceoecreasewithattaingdage 3
Source: BEIR V ( )
~
II.b..
t N R P B 1 t44...
- +ps 4 :- -
w%
40 ag,,, 9,,,,,
5
- j. \\
E
,/
\\
l
- - - as g-s __.
., N _.
,g
as-so /
1
/
m
/
~.
/
\\
- p
~
/
o a
\\
g
/
s\\
/
s w
O
/
's\\
Y
/
7 s,
~,,
lf 10 ~l
./*
R,,-----
l 0
31 45 60 75 I
A1TAINED AGE (y)
---r s a l
NATIO*:ALAhDEMYPR'
\\
Cha\\ r/Au.
I*
/
p;e F1 e
Booki No;' l'1 k -
N Yr M n.
-h MA
~
1 m
- 8 l
a.qrg.icas To x
l
'm 2,,
...- -~
jMWdn'd'hndn%WtVnWd%%%%fWdnd%%A4fffwggigg TPAH5MITTAL TC:
Document Control Desk. 016 Phillips ADVANCED COPY TO:
The Public Document Roem h
DATE:
N/J/9#
g FROM:
SECY Correspondence & Records Branch E
h L
Attached are copies of a Comission meeting transcript and related meeting 6
l document (s). They are being forwarded for entry on the Daily Accession List and l
placement in the Public Document Room.
No other distribution is requested or g
reovired.
5 Meeting
Title:
b IL A Na'Mm'd 6
h & f: l95 L
Meeting Date:
F/// 90 Open I Closed l
l item Description *:
Copies Advanced DCS
'8 to POR g
8 i
- 1. TRANSCRIPT 1
1 I
U//LJ inW W i
l l
4 IL m e mM i
I 2.
L-Ibh
/
/
bE?Z/
k.
/
i
- 3. N - &IR. E
/
5 4.
s.
b 6.
- POR is advanced one copy of each document, two of each SECY paper.
l C&R Branch files the original transcript, with attachments, withcut SECY
- papers, p /gh i
hE O uns _
90MHN0MN0%'amWomWmYNNmWWKKKKKKKKHHHKKKKKKKKKKKKKKKK