ML19338G385
| ML19338G385 | |
| Person / Time | |
|---|---|
| Issue date: | 10/31/1980 |
| From: | Aldrich D, Blond R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES), SANDIA NATIONAL LABORATORIES |
| To: | |
| References | |
| CON-FIN-A-1042 NUREG-CR-1433, SAND80-0981, SAND80-981, NUDOCS 8010290176 | |
| Download: ML19338G385 (69) | |
Text
.
t 5
NUREG/CR-1433 SAND 80-0981 Unlimited Release 1
i EXAMINATION OF Tile USE OF POTASSIUM IODIDE (KI) AS AN EMERGENCY PROTECTIVE MEASURE FOR NUCLEAR REACTOR ACCIDENTS I
David C. Aldrich Sandia National Laboratories Albuquerque, New Mexico 87185 Roger M.
Blond U.S.
Nuclear Regulatory Commission Washington, DC 20555 a
3 Date Published:
March 19R0 l
1 Sandia National Laboratories Albuquerque, New Mexico 87185 operated by Sandia Corporation for the U.S.
Department of Energy i
Prepared for Office of Nuclear Regulatory Research Probabilistic Analysis Staff U.S.
Nuclear Regulatory Commission Washington, D.C.
20555 Under Memorandum of Understanding DOE 40-550-75 NRC FIN No. A1042 l
34t#@@l %
3-4
L I
l 1
1 ABSTRACT j
l i
i Following the recent accident at Three Mile Island, there I
has been a resurgence of interest in the use of thyroid block-I ing as an emergency protective measure for reactor accidents.
An analysis has been performed to provide guidance to policy-
[
makers concerning the effectiveness of potassium iodide (KI) i j
as a blocking agent in realistic accident situations, the dis-i tance to which (or area within which) it should be distributed, i
j and its relative effectiveness compared to other available
(
protective measures.
l l
I The analysis was performed using the Reactor Safety Study (WASil-140 0 ) consequence model.
Four categories of accidents l
l 1
were addressed:
gap activity release accident (GAP), GAP l
without containment isolation, ccre melt with a melt-through release (Melt-Through), and core melt with an atmospheric I
release (Atmospheric).
Thyroid dose calculations show that l
the GAP category does not pose a significant health hazard i
to the public at any distance from the reactor.
For the GAP 5
l without containment isolation and Melt-Through categories, doses in excess of recommended protective action guidance i
levels (PAGs) (5-25 rem) are confined to areas within approx-
{
imately 10 and 15 miles of the reactor, respec t ive l'y.
For l
the Atmospheric category, however, thyroid doses are likely I
i to exceed PAGs out to 100's of miles.
i I
A cost-benefit analysis for the use of KI was also per-formed.
Cost-benefit ratios (S/ thyroid nodule prevented) are given assuming that no other protective measures are taken.
Uncertainties due to health effects parameters, accident prob-
[
abilities and costs are assessed.
The effects on predicted i
i ratios of other potential protective measures, such as evacu-j ation and sheltering, are addressed.
The impact on children (critical population) is also evaluated.
The estimated cost-l i
benefit ratios are high, and it appears that the distribution of KI is only marginally cost-effective, at best.
4 Finally, using statistics provided in NCRP Report No. 55, a simple risk-benefit analysis showed the risk of adverse re-l action posed by KI at the recommended action levels and dosages i
to be small compared to its potential benefits.
Iloweve r, several recent reports suggest that adverse reaction rates for some j
segments of the population may be higher than those estimated i
by the NCRP.
l l
l 5-6
I t
1 CONTENTS i
Page 4
i f
j Figures......................................
8 i
Tables.......................................
9 Acknowledgements.............................
11-12 l
Prologue 13 1.
Introduction.................................
15 2.
KI as a Protective Measure...................
18 3.
Accident Releases Considered.................
22 4.
Thyroid Dose and Health Effects calculations..............................
25 l
- Thyroid Dose 29
- Thyroid Dose Calculations..................
29 5.
Other Protective Measures....................
39 1
6.
Cost-Benefit Analysis 42
- Costs......................................
44 J
- Potential Impact of the Accidents 46
- Potential Reduction in Thyroid Nodules.....
46 t
- Accident Probabilities.....................
51
- Cost-Benefit Ratio.........................
53
- Sensitivities 54 1
7.
Risk-Benefit Analysis........................
59 8.
Summary, Conclusions and Recommendations.....
61 1
References 67 4
7
Figures No.
Page 1
Percent of Thyroid Blocking Afforded by 100 mg of Staole Iodine as a Function of Time (in hours) of Administration Before or After a 1 pCi slug Intake of I-131.
19 2
Conditional Probability of Exceeding Thyroid Doses of 0.01 and 0.2 rem versus Distance for an Exposed Adult Located Outdoors.
Probabili-ties are Conditional on a Gap Activity Release accident.
36 3
Conditional Probability of Exceeding Thyroid Doses of 1, 5, 10 and 25 rem for an Exposed Adult Located Outdoors.
Probabilities are Conditional on a GAP w/o Isolation Accident.
37 4
Conditional Probability of Exceeding Thyroid Doses of 1, 5, 10 and 25 rem for an Exposed Adult Located Outdoors.
Probabilities are Conditional on a Core Melt Melt-Through Accident.
38 5
Conditional Probability of Exceeding Thyroid Doses of 1, 10 and 25 rem for an Exposed Adult Located Outdoors.
Probabilities are Conditional on a Core Melt Atmospheric Accident.
40 l
e S
8
Tables No.
Page 1
Summary of Release Categories Representing Hypothetical Nuclear Reactor Accidents (from Ref. 1).
23 2
RSS Calculation of Expected Cases per Million Person-Rem of Benign and Cancerous Thyroid Nodules (from Ref. 1).
28 3
Mean Thyroid Dose (rem) versus Distance for Exposed Adult Located Outdoors.
The mean thyroid dose for a child would be approxi-mately a factor of 2 higher.
31 4
Conditional Probability of Thyroid Damage versus Distance for Exposed Adult Located Outdoors.
Probabilities are conditional on the accident occurring.
Probabilities would be approximately a factor of 2 higher for a child.
32 5
Fractional Components of Mean Thyroid Dose for Exposed Individual Located Outdoors.
34 6a GAP w/o Isolation.
Conditional Mean Number of Thyroid Nodules Within Selected Distance Intervals.
Aungformpopulationdensityof 100 persons / mile is assumed.
Risk coefficient = 334 thyroid nodules per 106 person-rem to thyroid.
47 6b Core Melt Melt-Through.
Conditional Mean Number of Thyrold Nodules Within Selected Distance Intervals.
A unifgrm population density of 100 persons / mile is assumed.
Rigkcoefficient= 334 thyroid nodules per 10 person-rem to thyroid.
48 6c Core Melt Atmospneric.
Conditional Mean Number of Thyroid Nodules (Albated Thyroids)
Within Selected Distance Intervals.
A uni-form population density of 100 persons / mile 2 is assumed.
R thyroid nodulesper10gskcoefficient=334 person-rem to thyroid.
49 7
Potential Reduction in Mean Number of Thyroid Nodules (Ablated Thyroids) by Use of KI.
99%
~
effective KI is assumed.
Numbers are deter-mined from Table 6.
50 9
- I Tables (cont'd)
No.
Page 8
Potential Reduction per Year of Reactor Operation in Mean dumoer of Thryoid Nodules by Use of KI.
99% effective KI is assumed.
RSS probabilities are assumed, 52 9
Estimated Cost-Benefit Ratios for Use of KI
($ per nodule prevented).
99% effective KI is assumed.
RSS probabilities are assumed.
55 10 Cost-denefit Analysis for Use of KI by Children.
Assumptions:
r 668thyroidnoduelsper10gskcoefficient=
person-rem to thyroid, no 0.1 dose effectiveness factor for I-131, Core Melt Atmospheric accident category only, RSS accident probacilities.
56 11 Cost-Benefit Analysis for Use of KI by Children.
Assumptions:
APS upper-bound risk coefficient fo roid nodules per 10g children of 6500 thy-person-rem to thyroid, no 0.1 dose effectiveness factor for I-131, Core Melt Atmospheric accident category only, RSS accident probabilities.
58 12 Summary Table for KI Cost-Benefit Analysis (from Table 9).
63 W
10
ACKNOWLEDGEMENTS The authors are indebted to their colleagues, D.
M.
- Ericson, N.
C.
Finley, J.
D.
Johnson, J.
M. Taylor (Sandia),
B.
K.
Grimes, J. A. Martin (NRC) and B.
Shleien (FDA), for their many helpful discussions and suggestions.
PROLOGUE During the first few critical days of the accident at Three Mile Island, many spontaneous decisions were made concerning offsite emergency protective measures.
The sense of the moment dictated action.
Plans were conceived and implemented with little or no time available to determine the potential benefits and costs associated with alternatives.
Specific plans were developed to evacuate the population within 20 miles of the re-actor; the Governor ordered a five mile precautionary evacuation of pregnant women and small children; and Potassium-Iodide medi-cation (KI) was manufactured and shipped to the area for possible distribution.
To provide an adequate planning basis for potential future accidents, it is necessary to determine how frequently they would occur; to estimate their anticipated impacts on the surrounding population; and to evaluate the potential benefits of alternative protective measures.
Several studies have focused on these impor-tant questions.1,2,3 It is also important to estimate the costs associated with various protective measure strategies.
With this information (i.e., probability of accident occurrence; impact on public; benefit of various protective measures; and associated costs), a rational basis would be available to make planning decisions.
It is the intent of this report to focus on one emergency protective measure (Potassium Iodide) and present information 13
needed to make a decision concerning a program for its use.
There are many uncertainties associated with the information, methods, and techniques which are used in this analysis.
As our knowledge and experience expands, the results,and conclu-sions of this type of study should be reevaluated'and, if necessary, changes should be made to the emergency planning strategy.
14
. - - ~_
i Il i
l 1.
Introduction i
l
' Potential accidents at nuclear reactors, however unlikely,
)
l cocid result in substantial offsite radiation exposures, and l
pose a serious threat to.the health and safety of the surround-l ing public.
If an accident were sufficiently severe, the re-l l
sulting radiological consequences could include immediate deaths I
and injuries, delayed cancer deaths, thyroid nodules, and long-term contamination of land and property.1 Any immediate ef fects, i
even for the worst accidents, would probaoly be confined to areas relatively close to the reactor (a few tens of miles)l,2 and could be significantly reduced by implementing immediate protective mea-sures.
Ilowever, cancer deaths and thyroid nodules could occur over much larger distances (100's of miles) and would therefore be less affected by immediate protective measures taken near the site.
The risk to the thyroid of exposed individuals posed by potential accidents is especially great for several reasons:
- Radioactive isotopes of iodine are produced in abundance l
by the fission process.
l
- Iodine and iodine compounds are normally quite volatile.
Therefore, a sizeable fraction of core radioiodine inven-i tories could be available for release to the atmosphere.
- Inhaled or ingested radioiodines are quickly absorbed into the bloodstream and concentrate preferentially in the thyroid.
- Iodines are eliminated from the thyroid with a
relatively long biological half-life.
i i
15
the radiation dose to the thyroid is likely to far As a result, exceed the dose to the rest of the body, and thyroid damage is individuals than any other accident-induced likely to affect more health effect.1,3 Taken in large enough quantities, potassium acts to block the absorption of radioiodines by the iodide (KI) thyroid, reducing the thyroid dose.
For this reason, KI has been discussed for many years as a potential protective measure for use in the event of a serious reactor accident.4*
The availability of KI would provide a supplemental strategy to be considered along with other possible protective measures, llowever, KI should not be considered a panacea for reactor acci-its ef fective use could significantly reduce the dents.
Although it would number of thyroid nodules resulting from an accident, have no impact on long-term land contamination or immediate health and only a moderate impact on delayed cancer deaths.
- effects, the only protective action that will reduce Use of KI is also not thyroid dose, nor is it without its difficulties and problems:
- The drug is not completely risk free; adverse reactions are possible.
- Making KI available would involve a cost to society; dollars that perhaps could be used to reduce risk more effectively elsewhere.
O Potassium iodate, a drug similar to KI, hasbeendistrgbuted for-use within a few miles of reactors in Great Britain.
A 3 recommends planning analysis by Beyea and von Ilippel in the U.S.,
on the recent for the use of KI over much larger distances l
f order of 100 or more miles from all reactors.
I 16
- There are serious storage and distribution logistical f
problems associated with ensuring that the public would i
i receive the drug in sufficient time to be effective.
l 1
i j
- It must be assured that any KI distribution strategy 1
i implemented would not reduce the effectiveness of other protective actions taken, e.g.,
if people are required to receive KI at a distribution center, they may be
" caught" by the cloud while outdoors, and receive a i
j higner dose than if they had stayed at home.
I l
A timely decision on the potassium iodide issue is required of responsible policymakers.
This report summarizes a study j
performed to provide them with technical guidance on that issue.
i It is intended (1) to provide insight concerning the effective-l j
ness of KI in potential accident situations, (2) to help determine the merits of KI as an emergency protective option, (3) to estab-lish the population and the distance to which (or area within i
which) it should be distributed, and (4) to determine under what i
conditions it should be implemented.
Simple cost-benefit and i
i risk-benefit analyses have Deen performed as part of this study.
The effects of other protective measures, such as evacuation and sheltering, are assessed as well.
Specific alternative strategies for stockpiling and distributing KI have not been addressed, although that would be essential to reduce costs and assure effectiveness before making KI available.
The analysis reported here was performed using the Reactor Safety Study (RSS) consequence model,1 CRAC, for a range of poten-tial reactor accidents. Four categories of accident releases are l
17
examined; from fuel pin gap activity release accidenta to com-plete core meltdowns with containment failure directly to the atmosphere.
It is important to note that there is a great deal of uncertainty in our knowledge of these releases and their probabilities, as well as dose-health effect relationships for the thyroid.
In some cases, these uncertainties hinder our ability to provide definitive gui6ance.
Ilo w e v e r, they are addressed to the extent possible in our analysis.
2.
KI as a Protective Measure Inhaled or ingested iodine is rapidly and almost completely absorbed into the bloodstream.
Almost one third of the iodine concentrates in the thyroid where it has a biological half-life of approximately 120 days.
The absorption of radiciodines by the thyroid is greatly reduced if body fluids are saturated with stable iodine prior to exposure.4 The blocking effectiveness of staole iodine is shown in Figure 1 as a function of the time of administration.
After a short-term exposure, the majority of radioiodine uptake by the thyroid occurs within 10-12 hours, and the initial administration of a blocking agent is therefore of little value oeyond that time.
Essentially complete curtail-ment (90% or greater) of radioiodine uptake by the thyroid requires that stable iodine be administered shortly before or immediately after the initiation of exposure.
A block of 50 percent or more is attainable only during the first few hours after exposure.
18
100 %
90-l d
t
\\
r s
\\
\\
70-
\\'
\\.' 9 6er so-gg '. g{ \\
40-
\\
J0-2g.
10-a
-40
-Je
-20
-70 g
70 29 gg 4p Figure 1.
Percent of Thyroid Blocking Afforded by 100 mg of Stable Iodine as a Function of Time (in hours) of Administration Before or After a 1 pci Slug Intake of I-131.
Ref:
Radioactive Iodine in the Problem of Radiation Safety ( USS R) (19 72), USAEC Translation Series, AEC-tr-7536.
Available from NTIS, US Depart-ment of Commerce, Springfield, VA 22151.
S 19
B 1
Several chemical compounds of stable iodine are suitable as clocking agents, including potassium iodide (KI) and potas-sium iodate.*
The Food and Drug Administration (FDA) has 4
recommended and approved oral administration of potassium iodide (KI) in dosages of 130 mg (tablet or liquid form) as a blocking 1
agent.4,6 Continued administration of this daily dose appears to maintain an essentially complete block.
A minimum of three l
to seven days administration would protably be required, and use of the drug is not expected to exceed 10 days.6 There is presently no definitive guidance concerning when, or under what conditions, KI should be used as a blocking agent.
The NCRP recommends that it be considered for use if the pro-i jected thyroid dose ** to an individual in the general public r
exceeds 10 rem.4 Protective Action Guides (PAGs) promulgated by the EPA for projected thyroid dose range from 5 to 25 rem.7 Protective action is recommended at the lower level for j
sensitive populations (pregnant women, children), or if there are no local constraints to providing protection at that level.
l Protective actions would be warranted in all cases if the pro-i
{
jected dose exceeds the higher value.
However, only evacuation l
- Radiological emergency plans in Great Britain include thyroid-blocking using 100 mg tablets of potassium iodate, since in the British experience, the shelf-life of the iodate is appreciably longer than that of iodide tablets.
The iodate form could be employed in the U.S. only by compliance with FDA requirements that include gathering tne pertinent clinical data for the o
j iodate.
- The projected thyroid dose is the estimated dose that would be received within a few days following the release if no protec-tive actions are taken.
20
, _ _ _ _ _ ~ -. _ _ _
and controlled area access were discussed in the EPA document,7 and the use of KI was not specifically cited as an appropriate protective measure.
There is considerable experience with the use of KI as a therapeutic drug.4 It has been used for a number of years in high doses, and on a long-term basis, for the treatment of vari-aus pulmonary disorders.
The reported incidence of adverse reactions to the drug is low, and the risk posed by the short-term use of the relatively low doses that would be involved with 4
response to an accident is judged to be minimal.
The NCRP estimates the adverse reaction rate to be between 1 x 10-7 and 1 x 10-6 per dose, and concludes that the administration of KI would not result in significant immediate side effects, even if given to large segments of the population.*
Because the prompt administration of KI in the event of an accident is critical to its ef fectiveness as a protective mea-sure, some method of rapid distribution to the public is required.
There is little current definitive planning for such methods.
Stockpiling supplies of KI in " distribution centers" such as schools, police stations, or firehouses has been recommended.4 An alternative would oe to provide each household with a suffi-cient supply fer all members of the household.
The feasibility and effectiveness of these and other alternative strategies, as well as their likely implementation costs, should be investigated.
~
- Note that warning would be given cautioning against the use of KI by individuals who are sensitive to iodine.
21
3.
Accident Releases Considered Release magnitudes for potential accidents of offsite significance range from relatively small releases of gap activity to the large releases predicted for full core-melt dccidents in which the containment fails directly to the atmos-phere.*
The RSSI grouped this spectrum of reactor accidents into nine release categories for pressurized water reactors (PWR) with large dry containments and five for boiling water reactors (BWR) with Mark I containment.
These categories are presented in Table 1 along with their estimated probabilities of occurrence, release magnitudes, and other parameters that characterize the release.
It should be noted that, because of the lack of complete understanding of the physical processes associated with core-melting and the resulting release of radioactive material to the environment, there is a large degree of uncertainty and overlap in these groupings.
There is also a significant uncertainty associated with their estimated probabilities,8 a point which will be discussed later in this report.
- A large light water power reactor typically contains about 10 billion curies of radioactive material.
The spectrum of pote in this study would release from 10 g-tial accidents addressed (1000 curies) to about one half (5 billion curies) of this radio-active material directly to the atmosphere, 21'
t Table 1.
Summary of Release Categories Representing Hypothetical Nuclear Reactor Accidents (from Ref. 1)*
i
.,.u....,.,........,i.....,,,
i..
. - n., i.
.... u
..,, ~
..i..-
.i.....
f
- c...,
..i.->"'
u a
i.
n a.- u.'a u
..w' 4
o
..i, a
o 3. ir e
e u
.e-un
.. i..- i
-a
.. ir.
u T
.. ic*
u u
u
.. ir i u
u u
un
- i. >
i. ire a..
u o
- 1. ire u.
u.
...ri
..ici mi
.. irs u
u u
u o
a. iri
.u.
..re s. ir.
i..-.
..ir.
,. ics
.. ir.
u..
in i..
u n. ici
.. ir.
.. ir.
.. r-i
..r-
..... i..,-.
.. ir i 4. irs
.. irs i. r-s
..rs i.,-.
r.
.. irs a..
in u
a
.. ici u
u
.,. < a 4. i.- >
i.
r.
i. ir.
...c i. e-.
i. ir.
i. w.
i. ir' i. ir '
..w' i. w.
i. ien
.. a-.
u o
i. ir.
u o
o n
no u
..wi
.o o
s. ic '
s
.. w.
su a
u x.
u
.. ic ar u.
..u*
- 1. wi io u
u n
2.
u
..i-i u.
.w u,
.. c.
.. w.
u u
u n
v. ic.
..w.
.. ic i
... n->
.. iv.
..u.
i, ac e
- i. i.-.
u u
n.
.. w.
.. iv.
.. iv ".. iv.
..wu
..ma (Q
i.
. i.
.n..i -,.i i.u.. i.... u.i.u
,....,n.u...n.....
u.,i u n, i....u.u, - n,... n i.,.......... u...
U<
,..o i.u
..... c.....
..,-u..-...c.....,.c..
.i.. o..
..-....,.au....i...o..,,.o.,.#,..
.-. -i.
-a..
nn,...
- >,-. u,
.a a. :-
- i..u
-n i.,.,... w '
. n..,-,,.- i n. w.a., #,.
i..n...
n.
.. i... i. u i.
i.. i....,.-. -.. u u-,. i-.,. -. -u. i -o u m.a.
u
.n
..a
..u 5'8]
l r---
- The time of release is the time interval between the initiation of the accident and the S
release of radioactive material from the containment structure to the atmosphere.
The duration of release is the period of time during which radioactive material is emitted M
to the atmosphere.
The warning time for evacuation is the projected time interval between D
awareness of impending core melt and the release of radioactive material from the contain-h l
ment building.
For those accidents in which core-melting does not occur, there is no projected warning time.
Finally, the height of release and the energy content of the released plume influence the height to which the plume rises and, thus, the exposure to 4
persons near the site.
w W
For the purpose of this study, the PWR accident release spectrum has been grouped into 4 categories:*
RSS Release Categories 1.
Gap Activity Release Accident (GAP)
PWR9 2.
Gap Activity Release Accident without Containment Isolation (GAP w/o Isolation)
PWR8 3.
Core Melt with Melt-Through Release (Core Melt Melt-Through)
PWR6-7 4.
Core Melt with Atmospheric Release PWRl-5 (Core Melt Atmospheric)
PWR9 represents a gap activity release accident in which only the activity initially contained within the gap between the fuel pellet and cladding would be released into the containment.
All engineered safeguards are assumed to function properly.
PWR8 is the same as PWR9, except that the containment fails to isolate properly on demand.
Again, all other engineered safe-guards, including containment sprays, are assumed to function properly.
PWR categories 1 through 7 are accidents in which core melt is assumed to occur.
PWR 6 and 7 are dominated by accident sequences involving containment failure by containment base mat melt-through.
PWRl-5, on the other hand, consist of accidents in which containment failure is assumed to occur directly to the atmosphere as a result of either inadequate isolation of contain-ment openings or penetrations,. a reactor vessel steam explosi an,
- These 4 categories are comprised of the RSS release categories from which they are defined, each weighted by its respective probability as calculated in the RSS.
l 14 1
hydrogen burning, or overpressure.
To reduce the required time and cost of computation, BWR accidents have not been considered specifically in this analysis.
However, the information and conclusions presented for large dry containment PWRs should be roughly applicable to other PWR designs and for BWRs as well, given a similar type of accident and mode of containment failure.*
4.
Thyroid Dose and Health Ef fects Calculations Dose to the thyroid is estimated as the sum of 1) external dose from the passing cloud (cloud exposure), 2) external dose from contaminated ground (ground exposure), 3) internal dose during the first 30 days from all inhaled radionuclides except I-131, and 4) internal dose during the first 30 days from inhaled I-131.
Thyroid dose from ingestion via the grass-cow-milk-man pathway and chronic exposure has not been included in this analysis because those pathways would not require an immediate emergency response in the event of an accident.
- BWR5 represents the BWR gap activity release accident.
BWRl-4 are accidents that involve core-melt.
For the specific BWR design investigated in the RSS, the probability of containment failure by containment vessel melt-through is essentially zero, i.e., the containment is assumed to always fail directly to the atmosphere.
BWR4 is dominated by accident sequences involving containment isolatica failure in either the drywell or wetwell, whereas BWRl-3 are dominated by accidents in which the contain-ment fails from either a steam explosion in the reactor vessel or containment, or from overpressure resulting in release through the reactor building or directly to the atmosphere.
Other con-tainment designs (e.g., PWR ice condenser, BWR Mark II or BWR Mark III) would have somewhat different probabilities for the various containment failure modes.
i l
25
The dose received by a child's thyroid is likely to be different than that received by an adult for several reasons, 3
including differences in thyroid mass, breathing rate, frac-tional iodine uptake, and metabolic rate.
The RSS assumed age dose f actors
- of 1.0 for children of ages 0-1 years, 1.9 for ages 1-10 years, and 1.6 for ages 10-20 years.
Somewhat higher factors (up to 5) have been assumed in other studies.3,9 There is considerable uncertainty concerning the effects of radiation exposure on the thyroid.I'#'9 Thyroid nodules are the effect of primary concern and would typically be observed from 10 to 40 years after exposure.1 A nodule is an abnormal growth that could be either benign or malignant (cancerous).
Nodules that are thought to be possibly malignant would most likely be surgically removed.
Most thyroid cancers are well differentiated, slow growing, and relatively amenaole to therapy.
Their associated mortality rate is therefore much lower than that for most other forms of cancer.
The RSSI conservatively assumed a 10 percent mortality rate for malignant thyroid nodules.
Based on the results of animal experiments and clinical 1
data for humans, the RSS assumed that internal irradiation of the thyroid by I-131 would be only 1/10th as effective as exter-nal x-rays in producing both benign and malignant nodules.**
- Ratio of child to adult inhalation dose.
- 0n a purely radiological basis, it is thought that the more uniform distribution of dose within the thyroid from external irradiation might increase the efficiency of inducing clinical hypothyroidism.
26
I This f actor of 0.1 for I-131 dose was disputed by the American Phys!. cal Society (APS) study group on reactor safety,9 which assumed a range of factors from 0.3 to 1.0.
Because this issue remains unresolved, calculations have been performed in this analysis both with and without a 0.1 factor for I-131 dose effectiveness.
Sufficiently high radiation doses
- would result in ablation of the thyroid with no subsequent risk of either benign or malignant nodules.1 However, because of the high doses required, thyroid ablation is unlikely to occur except for persons very near the reactor following the most severe accidents.
Ablation would probably require surgical removal of the thyroid, and the affected individual would need to take substitute hormone pills on a daily basis.
Thyroid damage, including both nodules and ablation, has been addressed in this analysis.
The RSS calculation of the expected number of thyroid nodules per million person-rem ** is reproduced in Table 2.
The assumed total incidence rate is 334 thyroid nodules per 106 person-rem, of which 60 percent are benign and 40 percent are malignant.
Although nct specifically computed, a dose-effects coefficient for a child's thyroid can be derived from the RSS
- The RS5~ assumed that doses in excess of 5000 rem (50,000 rem from I-131) would result in thyroid ablation.
A value of 3000 rem has been assumed in this analysis.
- Number of cases per million population per rem 27
i i
Ii$
Table 2.
R3S Calculation of Expected Cases per Million Person-Rem of Benign and Cancerous j
Thyroid Nodules (from Ref.1).
Life Latent Years Age Benign Nodules Cancers Age Group Fraction of Expectancy Period at Dose Risk Expected Risk Expected (years)
Population (years)
(years)
Risk Factor'3 Coefficenth Coefficienth Casesc Casesc 0 - 0.99 0.014 71.3 10 30 1.0 8
3.4 4.3 1.8 1 - 10 0.146 69.4 10 30 1.9 8
66.6 4.3 35.8 11 - 20 0.196 60.6 10 30 1.6 8
75.3 4.3 40.5 t
21 - 30 0.164 51.3 10 30 1
4 19.7 4.3 21.1 i
31 - 40 0.118 42.0 10 30 1
4 14.2 4.3 15.2 41 - 50 0.109 32.6 10 22.6 1
4 9.9 4.3 10.6 t
51 - 60 0.104 24.5 10 14.5 1
4 6.0 4.3 6.5
[
61 - 70 0.080 17.1 10 7.1 1
4 2.3 4.3 2.4 71 - 80 0.044 11.1 10 1.1 1
4 0.1 4.3 0.2 80+
0.020 6.5 10 0
1 4
0 4.3 0
i r
TorAL 200 134 aRatio of child to adult inhalation dose. See Tables VI-8-5 and 9-8 in reference 1.
- b. Ntrber of cases per million population per rem per year.
cExpected cases per million person-rem.
I 1
3 3
data to be approximately a factor of 2 higher.*
Beyea assumes the RSS values as lower bounds, and upper bounds of 650 thyroid 6
nodules per 10 person-rem for adults, and 6500 thyroid nodules 1
l per 10 person-rem for children.
6 i
Unless otherwise stated, the calculations performed in this study assume the RSS risk coefficient of 334 thyroid nodules per 6
10 person-rem.
This corresponds to an assumed risk, or prob-ability, of a thyroid nodule for an individual of 3.34 x 10-4/ rem,
]
i.e.,- 100 rem to an individual implies a probability of contract-j ing thyroid nodules of 3.34 x 10-2 For this assumed coefficient, a dose to an individual of 3000 rem gives a thyroid nodule prob-ability of approximately 1.0. Therefore, the following is assumed:
j Thyroid Dose
'l
< 3000 rem p(thyroid nodule)
= (3.34 x 10-4/ rem)(dose in rem) i
> 3000 rem p(thyroid nodule)
=0 j
p(ablated thyroid) = 1.0 4
1 l
The effect of uncertainty in the thyroid dose-effect relation-ship is assessed by repeating some calculations using the upper bound values proposed by Beyea3 and the APS.9 Thyroid Dose Calculations A series of calculations was performed using CRAC,1,10 to i
determine 1) the magnitude of the threat to the thyroid of 1
i
- For age group 1-10:
(years at risk) (age dose factor) (risk coeffigient) 30 x 1.9 x (8 + 4.3) = 707 thyroid nodules
=
l per 10 person-rem (see Table 2).
l 5
29
exposed individuals, 2) the distance to which that threat is likely to be of concern, and 3) tne relative contributions of different exposure pathways and radioisotopes to the thyroid dose, for each of the four accident categories definea in the previous section.
All calculations were performed for a 3200 MWt PAR using one year of meteorological data taken from a single reactor site.*
From the year's data, 91 different weather sequences were selected by stratified sampling 1 and used to generate probability distributions of thyroid dose versus distance.
Breathing rate and snielding parameters appropriate for a person located outdoorsl,2,12 are assumed:
breathing rate = 2.66 x 10-4 m /s, shielding factors = 1.0 3
(cloud exposure) and 0.7 (ground exposure).
For each accident aategory, Table 3 presents the mean thyroia cose that would be received by an exposed adult located outdoors at selected distances from the reactor.
The corres-ponding dose to a child's thyroid would oe approximately a factor of 2 higher.
Table 4 presents the associated probability of thyroid damage for the same individuals.
The values shown equal the doses in Taole 3 multiplied by the RSS risk coefficient of 3.34 x 10~4 per person-rem to the thyroid.
l l
I
- Site-to-site variations in meteorological histories have been shown to have little e publichealtheffects.((ectonthepredictionoflong-term Therefore, the use of meteorological data from a single site is considered sufficient for this study.
30
9 i
Table 3.
Meana W yroid Doseb (rem) versus Distance for Exposed Adult Incated Outdoorse Re mean thyroid dose for a child would be approximately a factor of 2 higher.d Accident Category Distance (miles)
GAP GAP w/o Isolation Core Melt Melt-% rough Core Melt Atmospheric 1
5.7 x 10-2 55 25 1.3 x 104 5
4.0 x 10-3 3.9 1.7 5.8 x 103 10 1.1 x 10-3 1.1 5.2 x 10-1 3.2 x 103 25 1.7 x 10-4 1.7 x 10-1 7.6 x 10-2 1.1 x 103 50.
4.2 x 10-5 4.2 x 10-2 2.0 x 10-2 3.8 x 102 100 1.1 x 10-5 1.1 x 1 T2 5.9 x 10-3 1.0 x 102 150 3.8 x 10-6 3.8 x 10-3 2.0 x 10-3 36 200 1.9 x 10-6 1.9 x 10-3 1.0 x 10-3 16 a91 weather sequences were used to calculate a probability distribution of dose at each distance. %e mean doses presented are the mean of those distributions bCalculated doses include: dose from inhaled radionuclides from cloud passage, plus external dose due to the passing cloud plus 1-da sure to ground contamination.
cBregthing rate = 2.66 x 'Tg ep/s.Shielding factors = 1.0 (cloud exposure) and 0.7 (ground exposure).
m dRSS assumed age dose <
or of 1.9 for children aged 1-10 (see Section 3).
I d
U d
D tante 4.
(.bnditional Probability of W yroid Damage versus Distance for Exposed Adult Iocated Outdoors. Probabilities aJ3 conditional on the accident occurring.
Probabilities would be approximately a factor of 2 higher for a child.c Accident Category Distance (miles)
GAP GAP w/o Isolation Core Melt Melt-% rough Core Melt Atmospheric 1
1.9 x 10-5 1.8 x 10-2 8.4 x 10-3 0.6 '
d 5
1.3 x 10-6 1.3 x 10-3 5.7 x 10-4 0.7d 10 3.7 x 10-7 3.7 x 10-4 1.7 x 10-4 0.7d I
25 5.7 x 10-8 5.7 x 10-5 2.5 x 10-5 0.4d 50 1.4 x 10-8 1.4 x 10-5 6.7 x 10-6 1.3 x 10-1 100 3.7 x 10-9 3.7 x 10-6 2.0 x 10-6 3.3 x 10-2 150 1.3 x 10-9 1.3 x 10-6 6.7 x 10-7 1.2 x 10-2
[
L 200 6.3 x 10-10 6.3 x 10-7 3.3 x 10-7 5.3 x 10-3 aNo 0.1 effectiveness factor for I-131 dose is assumed. Values presented equal doses in Table 3 multiplied by assumed risk coefficient of 334 thyroid nodules per 106 person-rem to the thyroid.
b 2yroid damage includes thyroid nodules (both benign and cancerous) and ablated thyroids.
CSee Section 3.
i dProbabilities are less than 1.0 because for some accidents and weather conditions, the energy of release is sufficiently high to result in significiant plume rise.
In these cases, the plume would travel over the heads of individuals near the reactor, and resulting thyroid doses would be low.
t s
1 1
r i
e I
I 4
The probability of thyroid damage to an individual follow-t j
ing a gap activity release accident (GAP) is extremely low, I
ranging from less than 2 x 10-5 (1 in 50,000) 1 mile downwind i
of the site to less than 4 x 10-9 (1 in 250,000,000) at 100 miles.
Probabilities are somewhat higher for the GAP w/o Isolation and Core Melt Melt-Through accidents.
Thyroid damage I
probabilities for the Core Melt Atmospheric accidents are much higher, and such accidents could pose significant health hazards 1
j to persons at distances of more than 100 miles from the site.*
These results agree with those of previous studies.2,3 Fractional components of the mean thyroid dose are pro-
)
vided in Table 5 for selected distance intervals: 0-25 miles, i
l 25-100 miles, and distances greater than 100 miles.
Within these intervals, the relative contributions to thyroid dose I
will not differ significantly.
The dose is divided into com-ponents for the inhalation of radiciodines, inhalation of non-radioiodines, cloud exposure and ground exposure.
Radioiodine inhalation is further divided into components for I-131 and f
othe r - iodines.
It is evident from Table 5 that the thyroid dose is dominatad by the inhalation of radiciodines for each
)
of the four accident categories.
Inhalation of I-131 alone
]
- Caution must be used in interpreting the large distances indi-l cated.
The RSS consequence model assumes an invariant wind direction following the release of radioactive material, llow-l l
ever, because of the time required by the cloud to travel large l
i dis tances, it is likely that the wind direction will, in fact, i
j shift and that the predicted dose levels would not be observed at the reported radial distance.
Rather, the distance applies more closely to the distance along the trajectory of the released cloud.
33 5
T Table 5.
Fractional Components of Sm 'Ihyroid Dose for Exposed Individt:al Incated Outdoors Distance Interval Inhaled Radiciodinesa Inhaled Cloud Ground a
b (miles)
I-131 Other Iodines Non-radioiodines Exposure Exoosurec A.
GAP 0-25 0.67 0.25 0.02 0.03 0.03 25-100 0.70 0.22 0.02 0.04 0.02
>100 0.77 0.16 0.03 0.02 0.02 B.
GAP w/o Isolation 0-25 0.68 0.25 0.02 0.02 0.03 25-100 0.71 0.23 0.02 0.02 0.02
>100 0.78 0.16 0.02 0.02 0.02 C.
Core Melt Melt 'Ihrough 0-25 0.65 0.16 0.10 0.06 0.03 25-100 0.63 0.15 0.10 0.09 0.03
>100 0.63 0.09 0.09 0.16 0.03 D.
Core Melt Atmospheric 0-25 0.67 0.21 0.07 0.01 0.04 25-100 0.72 0.20 0.05 0.01 0.02
?"100 0.77 0.16 0.05 0
0.02 3
aBreathing rate = 2.66 x 10-4 m /s.
bShielding factor for exposure to cloud = 1.0.
cl-day exposure to ground contanination.
Shielding factor = 0.7.
A
4 accounts for 60-80 percent of the total dose, and other iodines contribute another 10-25 percent.
Inhalation of non-radio-iodines, cloud exposure and ground exposure are all small contributors to total thyroid dose.
The probabilities of exceeding thyroid doses of 0.01 and 0.1 rem versus distance from the reactor are shown in Figure 2,
conditional on the occurrence of a gap activity release accident (GAP).
The probabilities are calculated for an exposed adult located outdoors.
The selected dose levels, 0.1 and 0.01 rem, are far lower than any recommended action levels, and are still confined to areas very close to the reactor.
Therefore, it is evident that the GAP accident does not pose a significant hazard to the public.
Figures 3 and 4 show the probability of exceeding thyroid doses of 1, 5, 10 and 25 rem versus distance for the GAP w/o Isolation and Core Melt Melt-Through accidents.
The 5, 10 l
and 25 rem dose levels were chosen because they represent the range of action levels that have been recommended for the initiation of emergency protective measures.
The 1 rea level was added as a lower bound for doses of interest.
It is evident from these results that, for all practical purposes, projected thyroid doses of concern are confined to areas within a few 10's of miles of.the reactor for these types of accidents, and in most cases to areas considerably closer.
For the GAP w/o i
Isolation accidents, doses in excess of 5 rem are confined to about 10 miles; those in excess of 25 rem to about 5 miles.
The 35
5
- 1. 0 i
- E LJx 5
d 0.1 u.
a o
2 g<
2 e~-
~
~
m u.
< o m
m
@d o o
_; o IC Eo
- e E $ 0. 01 z -
oo v z 5
W o
Es GAP
'l
l 0.001 O.I 1
10 100 DISTANCE (miles )
Figure 2.
Conditional Probability of Exceeding Thyroid Doses of 0.01 and 0.1 rem versus Distance for an Exposed Adult Located Outdoors.
Probabilities are Condi-tional on a Gap Activity Release Accident (GAP).
36 m..
- 1. 0 i
i i,ii_
s
~
l rem w*
5 rem A
10 8 < 0.1 r rem 7
2 25 rem 3
~_,.
g m 4 m
u_
oo E
O m
Aoo z
$ 0. 01 E !E o x u-o
_z y
M GAP w/o ISOLATION m
I
l 0.001 0,1 1
10 100 DISTANCE ( miles )
Figure 3.
Conditional Probability of Exceeding Thyroid Doses of 1, 5,
10 and 25 rem for an Exposed Adult Located Outdoors.
Probabilities are Conditional on a GAP w/o Isolation Accident.
i 37
- 1. 0 N
u az 25 10 5
>- e-
~
s
$d I
o u.
[O
<0 a
m zo 9ose i
z o 0. 01 a:
o$
~
e Z
e y
5 CORE MELT MELT - THROUGH o
L 0.001 i
b i'
O. I 1
10 100 DISTANCE ( miles )
Figure 4.
Conditional Probability of Exceeding Thyroid Doses of 1, 5, 10 and 25 rem for an Exposed Adult Located Outdoors.
Probabilities are Conditional on a Core Melt Melt-Through Accident.
38
i i
same dose levels are confined to approximately 15 and 7 miles, respectively, for the Core Melt Melt-Through category.
The conditional probabilities of exceeding thyroid doses cf 1, 10 and 25 rem for the Core Melt Atmospheric category are shown in Figure 5.
The thyroid dose levels of concern are likely to be exceeded at very large distances from the reactor (and correspon-dingly over very large areas) if this type of accident were to occur.
5.
Other Protective Measures It was shown in the previous section that, for eacn of the four accident categories addressed, the thyroid dose is dominated by the inhalation of radiciodines.
Therefore, in order to be effective in reducing the thyroid dose and resulting health im-pacts, a protective measure must reduce the inhalation dose.
KI does this by blocking the absorption of inhaled radioiodines by the thyroid.
Iloweve r, other protective measures, including both evacuation and sheltering, can also act to reduce inhalation dose.
Evacuation, which is the expeditious movement of the population, is considered to be the primary protective measure in most radiological emergency planning within the United States. 13,14,15,16 Evacuation could potentially be 100 percent effective in reducing all dose if accomplished before arrival of the radioactive cloud.
On the other hand, it could be in-39
I I
- 1. 0 _
i i, i...._
l rem
_~
~
10 rem
- EW 25 rem 4
m
]
N ez u_ <
0.1
>2 e
_~
u_
mo om EM o
qa a e 9g e ce o >-
zI o - 0. 01 :-
o az E
tri o
x CORE MELT ATMOSPHERIC 0.001 10 100 1000 DIST ANCE ( miles )
Figure 5.
Conditional Probability of Exceeding Thyroid Doses of 1, 10 and 25 rem for an Exposed Adult Located i
Outdoors.
Probabilities are Conditional on a Core Melt Atmospheric Accident.
40 1
4 effective in reducing inhalation doses if not initiated until after the cloud has passed.*
t Sheltering might also provide some reduction in thyroid dose and could potentially be implemented at much larger dis-tances than evacuation.
Sheltering is the deliberate action by the public to take advantage of the protection against radiation exposure af forded by remaining indoors, away from doors and win-dows, during and after the passage of the cloud of radioactive material.
The shielding inherent in normally inhabited structures i
offers some degree of protection against external penetrating radiation from airborne and surface-deposited radionuclides.
Furthermore, the exclusion of a significant amount of airborne t
radioactive material from the interior of a structure, either by natural effects or by certain ventilation strategies, can re-t duce the amount of inhaled radionuclides as well.17 A recent 18 study suggests that a factor of 2 reduction in inhalation dose can be assumed for sheltered individuals.
That factor has been j
assumed in the following cost-benefit analysis.
Finally, other potential measures such as breathing through 1
either respirators or common household items, e.g.,
handkerchiefs and towels,19,20 may provide additional protection against dose
- Even in situations where the radioactive cloud has passed, eva-cuation could be valuable to reduce exposure to ground contamina-i tion.
However, since thyroid dose is dominated by radioiodine inhalation, it would not be reduced significantly in this case.
It is also possible that evacuating persons could receive increased inhalation doses if, for example, they remained in the cloud for a longer period of time or moved toward, rather than away from, the reactor while in the plume.
41
from inhalation of radionuclides.
However, further research is required to determine their effectiveness in realistic accident situations, and they have not been addressed in this analysis.
6.
Cost-Benefit Analysis 5
The decision to use potassium iodide (KI) as a protective measure should be based, at least in part, on its cost-effective-ness relative to other available protective or safety measures.
To analyze the costs and potential benefits of KI, the following information is needed:
Costs; Potential impact of accidents; Potential reduction in accident impacts; and Accident probabilities.
The cost of implementing a KI program would include: the purchase price of the KI in tablet or liquid form (both original and per-iodic replacement costs); costs for stockpiling, distributing and monitoring the status of the drug; and administrative expenses associated with the program.
The potential impact of the accident is measured here by the mean number of thyroid nodules that would occur within selected distance intervals.
The reduction in acci-dent impact is measured as the difference between the number of thyroid nodules predicted if no protective actions are taken (normal activity) and the number predicted if various protective actions are implemented.
Accident probabilities are expected occurrence rates per year of reactor operation.
By combining the costs with the accident probabilities and the estimated reduction 41
in effects, a cost-benefit ratio is generated.
The cost-benefit ratio for KI is interpreted as the expected number of dollars required to prevent a single thyroid nodule.
The cost-benefit ratio has been evaluated for the GAP w/o Isolation, Core Molt Melt-Through, and Core Melt Atmospheric accident categories over selected distance intervals out to 200 miles from the reactor.
Because few, if any, thyroid nodules are likely for the gap activity release accident (GAP), that category has not been addressed.
Calculations were performed for a 3200 MWt PWR using CRAC in the same manner as described in Section 4.
Several additional assumptions were made to facilitate the analysis and to allow the presentation of results in a concise and easily interpretable manner.
All calculations assume that KI is 99 percent effective in reducing the dose to the thyroid from inhaled radioiodines. This is obviously a limiting case since it assumes that all affected individuals take the drug before or immediately after the cloud passes.
A uniform population density of 100 persons per square mile was also assumed.*
Results for real, or site-specific, population distributions can be estituated by scaling the 100 persons / mile 2 results within each distance interval.
Finally, calculations were performed both with and without the 0.1 dose effectiveness factor for I-131 discussed in Section 4.
- Because costs are also assumed to be proportional to popu-lation density, this assumption does not impact the cost-benefit ratios calculated.
43
Costs The stockpiling, distribution, monitoring, and administrative 1
costs of a KI program would depend on the specific strategy of j
implementation and are difficult to estimate.
Therefore only i
the. original purchase and replacement costs of the drug are addressed in this analysis.
The following assumptions are made:
1)
Cost of KI per individual (14 tablets in a bottle)
=
S0.50.*
r 2)
KI is replaced every five years (i.e.,
5 year shelf 4
life).**
i 3)
KI is available for all persons within a given distance interval.
4)
No redundancy of KI locations (i.e., no extra tablets l
are available).***
l The cost per year to provide KI for all persons within an interval 4
is therefore equal to the nutaber of persons in the interval x S0.50/ person x'l/5 years.
l l
I
- This value is consistent with the price range (SO.41 to 0.75, depending on quantity) quoted by a U.S.
drug firm that manu-factures KI.
- KI tablets and solution currently approved by the U.S. Food and Drug Administration (FDA) for marketing bear 2-year expira-i tions.
However, improved product stability should be possible, j
Therefore, a 5-year shelf-life is assumed here.
- Considering the importance of prompt distribution and administra-j.
tion of KI, some redundancy of storage locations would be desir-i able'. -However, the extra cost from this redundancy has - not been included here.
44 eww e
ci-en
-'"-'e T-w wewww rw-vt v,+www yee ww y
y
<W7--+-w v'r-,'--
w y
y='-"-
y w
For the uniform population density of 100 persons / mile 2 assumed in this analysis, the number of persons located within selected distance intervals are as follows:
Distance Interval No. Persons in Cumulative No.
(miles)
Interval Persons 0-5 7,900 7,900 5-10 23,600 31,400 10-25 165,000 196,000 25-50 589,000 785,000 50-100 2,360,000 3,140,000 100-150 3,930,000 7,070,000 150-200 5,500,000 12,000,000 Using this information, the estimated annual cost for a KI program within each interval is given below.
Distance Interval (miles)
Cost (S/ year) 0-5 790 5-10 2,400 10-25 16,000 25-50 59,000 50-100 240,000 100-150 390,000 150-200 550,000 At the assumed cost of $0.10 per person per year, the annual cost to implement a KI program for the entire U.S. would be about
$20 million.*
- 0ther distribution strategies, such as regional storage, could substantially reduce this cost.
Iloweve r, because of longer implementation times, the effectivenss of these strategies may also be reduced.
45
i 5
[
Potential Impact of the Accidents The mean number of thyroid nodules
- that would occur within i
selected distance intervals for the three accident categories addressed are given in Tables 6a, 6b and 6c.
Results are pre-sented separately for four protective measure combinations:
- 1) normal activity, i.e.,
no protective actions taken,**
- 2) normal activity plus 99 percent effective KI, 3) sheltering ***,
and 4) sheltering plus 99 percent effective KI.
Although results i
are not specifically presented for evacuation, they would range l
from zero within all distance intervals to approximately those values shown for normal activity (see Section 5).
Pctential Reduction in Thyroid Nodules The potential reductions in the mean number of. thyroid nodules that would result by the use of KI are presented in j
Table 7.
The values provided were determined from those given 2
in Tables 6a, 6b and 6c.
As an example, for the GAP w/o Isola-tion accident, the mean number of nodules in the 0-5 mile interval is 1.77 for normal activity and 0.09 for normal activity plus i
- For the Core Melt Atmospheric accident category, thyroid
{
doses can be sufficiently high to result in ablated thyroids as well as nodules.
Mean numbers of ablated thyroids in each distance interval are given in parentheses in Table 6c.
i 1
l
- Shielding factors = 0.75 (cloud exposure) and 0.33 (ground exposure).
1-day exposure to ground contamination (see reference 1).
- Shielding factors and ground exposure time are the same as for normal activity.
50 percent reduction in inhalation dose.
O l
= _ _..
-.~
Table 6a. GAP w/o Isolation. ConditionalMeanNumberofThyroidNodulesWithinplected Distance Intervals.
A uniform populat%n de is asstuned.
Riskcoefficient=334thyroidnodulesper10gsityof100 persons / mile person-rem to thyroid.
Without 0.1 dose effectiveness factor for I-131 c
Distance Interval Normal Activity Sheltering a
b (miles)
Normal Activity 99% KI Sheltering 99% KI 0-5 1.77 0.09 0.90 0.06 5-10 0.35 0.02 0.18 0.01 10-25 0.43 0.03 0.22 0.02 25-50 0.32 0.02 0.16 0.01 50-100 0.36 0.02 0.18 0.01 100-150 0.17 0.01 0.09 0.01 150-200 0.11 0.01 0.06 0
With 0.1 dose effectiveness factor for I-131 0-5 0.66 0.07 0.35 0.05 5-10 0.13 0.02 0.07 0.01 10-25 0.16 0.02 0.08 0.02 25-50 0.11 0.02 0.06 0.01 50-100 0.12 0.02 0.06 0.01 100-150 0.05 0.01 0.03 0.01 150-200 0.03 0.01 0.02 0
%hielding factors = 0.75 (cloud exposure) and 0.33 (ground exposure). 1-day exposure to ground contamination.
bShielding factors and ground exposure same as for normal activity. Inhalation reduction factor = 0.5.
't
8 Table 66.
Core Melt Melt-through. Conditional Mean Number of Thyroid hbdules Wjthin Selected Distance Intervals. A uniform population den is acsumed.
Riskcoefficient=334thyroidnodulesoer10gityof100 persons / mile person-rem to thyroid.
Without 0.1 dose effectiveness factor for I-131 Distance Interval Normal Activity Sheltering (miles)
Normal Activitya
.D 99% KI Sheltering 99% KI 0-5 2.34 0.36 1.22 0.23 5-10 0.53 0.09 0.28 0.06 10-25 0.66 0.12 0.36 0.09 25-50 0.52 0.10 0.28 0.07 50-100 0.56 0.11 0.30 0.08 4
'100-150 0.30 0.07 0.17 0.05 150-200 0.21 0.05 0.12 0.04 With 0.1 dose effectiveness factor for I-131 i
0-5 0.91 0.34 0.50 0.22 5-10 0.21 0.09 0.12 0.06 10-25 0.27 0.12 0.16 0.09 25-50 0.21 0.10 0.13 0.07 50-100 0.23 0.11 0.14 0.08 100-150 0.12 0.07 0.08 0.05 150-200 0.08 0.05 0.06 0.04 1
aShielding factors = 0.75 (cloud exposuro) and 0.33 (ground exposure).
1-day exposure to ground contamination.
bShielding factors and ground exposure same as for normal activity.
Inhalation reduction factor = 0.5.
l l
4
?
Table 6c. Core Melt Atmospheric. Conditional Mean Number of Thyroid Nodules ( Albated 'Ihyroids)2 Within Selected Distance Intervals. A uniform population s)ensity of 100 persons / mile is assumed. Risk coefficient = 334 thyroid nodules per 10u person-rem to thyroid.
4 1
Without 0.1 dose effectiveness factor for I-131 Distance Interval Normal Activity Sheltering (miles)
@ rmal Activity 99% KI Shelteringb a
99% KI 0-5 81 (137) 49 (0) 76 (92) 31 (0) 5-10 192 (292) 81 (0) 210 (146) 48 (0) e 10-25 1110 (610) 181 (0) 918 (102) 109 (0) 25-50 2110 (210) 193 (0) 1190 (30) 115 (0) 50-100 2970 (20) 234 (0) 1520 (0) 140 (0) 100-150 1580 (0) 119 (0) 802 (0) 70 (0) i 150-200 992 (0) 76 (0) 503 (0) 45 (0)
With 0.1 dose effectiveness factor for I-131 0-5 73 (73) 46 (0) 76 (25) 29 (0) 5-10 231 (63) 75 (0) 158 (8) 46 (0) 10-25 735 (31) 168 %
403 (3) 102 (0) 25-50 836 (22) 177 (0) 448 (0) 107 (0) 50-100 995 (0) 214 (0) 520 (0) 129 (0)
]
100-150 473 (0) 108 (0) 247 (0) 64 (0) 150-200 280 (0) 68 (0) 147 (0) 41 (0) aShielding factors = 0.75 (cloud exposure) and 0.33 (ground expuure). 1-day ennsure to ground contamination.
i bShielding factors and ground exposure sane as for normal activity.
Inhal:ttion reduction factor = 0.5.
Table 7.
Potential Reduction in Mean Number of Thyroid !bdules ( Ablated 'Ihyroids) by Use of KI.
99% effective KI is assumed. Numbers are determined from Table 6.
Without 0.1 dose effectiveness factor With 0.1 dose ef fectiveness factor for I-131 for I-131 Distance Interval (miles)
Normal Activity Sheltering Normal Activity Sheltering GAP w/o Isolation 0-5 1.68 0.84 0.59 0.30 5-10 0.33 0.17 0.11 0.06 10-25 0.40 0.20 0.14 0.06 25-50 0.30 0.15 0.09 0.05 50-100 0.34 0.17 0.10 0.05 100-150 0.16 0.08 0.04 0.02 150-200 0.10 0.06 0.02 0.02 Core Melt Melt-Through 0-5 1.98 0.99 0.57 0.28 5-10 0.44 0.22 0.12 0.06 10-25 0.54 0.27 0.15 0.07 25-50 0.42 0.21 0.11 0.06 50-100 0.45 0.22 0.12 0.06 100-150 0.23 0.12 0.05 0.03 150-200 0.16 0.08 0.03 0.02 Core Melt Atmospheric 0-5 32 (137) 45 (92) 27 (73) 47 (25) 5-10 111 (292) 162 (146) 156 (63) 112 (8) 10-25 929 (610) 809 (102) 567 (31) 301 (3) 25-50 1920 (210) 1080 (30) 659 (22) 341 (0) 50-100 2740 (20) 1380 (0) 781 (0) 391 (0) 100-150 1460 (0) 732 (0) 365 (0) 183 (0) 150-200 916 (0) 458 (0) 212 (0) 106 (0) lillli
99 percent effective KI (Table 6a).
The difference between these two numbers (1.68) is the reduction afforded by using KI.
Accident Probabilities The probability of occurrence estimated by the RSS1 for the accident categories addressed in this analysis can be obtained from the data in Table 1.
Estimated Probability RSS Categories (per reactor-year)
GAP PWR9 4 x 10-4 GAP w/o Isolation PWR8 4 x 10-5 Core Melt Melt-Through PWR6-7 4.6 x 10-5 Core Melt Atmospheric PWR1-5 1.4 x 10-5 The RSS probabilities were used with the results in Table 7 to determine the potential reduction in the mean number of thyroid nodules per year of reactor operation by implementing a KI strategy.
Those values, which are shown in Table 8, include contributions from all 3 of the accident categories considered.*
Note that the contribution from the Core Melt Atmospheric category dominates (95-100%).
- The expected reduction per reactor year = Zi (potential reduction)i (accident probability)i, where i is the acci-dent category.
51
. ~.
U Table 8.
Potential Reductiona per Year of Rccctor Operation in Mean Number of Thyroid Nodulesb by Use of KI.
99% effective KI is asstrned. RSS probabilities are assumed.
a Without 0.1 dose effectiveness factor With 0.1 dose effectiveness factor for I-131 for I-131 Distance Interval (miles)
. Normal Activity Sheltering Normal Activity Sheltering 0-5 2.5 x 10-3 2.0 x 10-3 1.4 x 10-3 1,n y ig-5-10 5.7 x 10-3 4.3 x 10-3 3.1 x 10-3 1.7 x 10 10-25 2.2 x 10-2 1.3 x 10-2 8.4 x 10-3 4.3 x 10-25-50 3.0 x 10-2 1.6 x 10-2 9.5 x 10-3 4.8 x 10-(
50-100 3.9 x 10-2 1.9 x 10-2 1.1 x 10-2 5.5 x 10-100-150 2.0 x 10-2 1.0 x 10-2 5.1 x 10-3 2.6 x 10-150-200 1.3 x 10-2 6.4 x 10-3 3.0 x 10-3 1.5 x 10-I aReductions calculated from values in Table 7.
Expected reduction =
{ (potential reduction)i (accident probability)i, where i is the accident category.
per reactor-year i
bIncludes ablated thyroids.
o
i
)
i The uncertainties in the probabilities used above are large.
i Error bounds of factors of 1/5 and 5 on the values above were estimated in the RSS.
In 1978, the risk assessment review group (Lewis Committee),0 chartered by NRC to review the Reactor Safety f
i-Study, concluded "We are unable to determine whether the absolute probabilities of accident sequences in WAsil-1400 are high or low, but we believe that the error bounds on those estimates are, in general, greatly understated. "
Operating experience data for light water reactors (LWR) can also be used to estimate an upper bound for the probability of core melt.21 Through the end of 1979, there had been approximately 450 years of LWR experience in the U.S., without a core melt event.*22 Assuming a X distribution for such potential events, it can be shown that the probability of core melt is less than 1.5 x 10-3 with 50 percent confidence, and less than 6.7 x 10-3 with 95 percent confidence.**21 These l
upper bound probabilities are approximately factors of 25 and 1
i 100 times the-RSS values above (4.6 x 10-5 + 1.4 x 10-5 = 6.0 I
f x 10-5),
4 Cost-Benefit Ratio i
l Combining the estimated costs and the results in Table 8, i
estimated cost-benefit ratios for the use of KI are presented 4
t
- Although the accident at Three Mile Island involved serious core damage, it was not a core melt event.
- WorldwggeLWRexperience through 1979 was closer to 1000 reactor-1 years.
Using this value rathe in probability estimates of 7 x 10~y than.450 years results vith 50 percent confidence, and 3 x 10-3 with 95 percent confidence.
lk 53
in Table 9 in terms of $ per nodule prevented, i.e.,
the expected number of dollars required to prevent a single thyroid nodule.
The estimated ratios range from 3.2 x 105 S/ nodule prevented (for the 0-5 mile interval, normal activity, and no 0.1 dose effectiveness factor for I-131) to 3.7 x 108 S/ nodule prevented (f ; the 150-200 mile interval, sheltering and 0.1 I-131 dose effectiveness factor).
Sensitivities Table 10 summarizes a cost-benefit analysis performed speci-fically for the use of KI by children.
The risk coefficient 6
assumed, 668 per 10 person-rem,* is a factor of 2 higher than that assumed in Table 9.
Other assumptions include: no 0.1 dose effectiveness factor for I-131, RSS accident probabilities, normal activity, and a uniform population density of 100 persons /
mile 2 Only the Core Melt Atmospheric accident category was addressed.
However, as shown earlier, this has a negligible effect on the predicted results.
The cost-benefit ratios in Tables 9 and 10 are not significantly different for the intervals close to the reactor.
This is because the doses within those intervals are suf ficiently high to result in thyroid nodules for essentially all exposed individuals, regardless of the coefficient assumed.
At larger distances, the cost-benefit ratio in Table 10 is a factor of 2 lower, as expected.
- This isalsoveryclosetogheriskcoefficientassumedbyBeyea for adults (see Section 4).
54
=.
l I
Table 9.
Estimated Cost-Benefit Ratios for Use of KI (S per nodule pre-venteda) 99% effective KI is asstuned. RSS probabilities are assumed.
i Without 0.1 dose effectiveness factor With 0.1 dose effectiveness factor for I-131 for I-131 Distance Interval (miles)
Normal Activity Sheltering Normal Activity Sheltering 5b 4.0 x 105 5.6 x 105 7.9 x 105 0-5 3.2 x 10 5-10 4.2 x 10 c 5.6 x 105 7.7 x 105 1.4 x 106 5
6 1.9 x 106 3.7 x 106 5d 1.2 x 10 10-25 7.3 x 10 6
7 25-50 2.0 x 10 e 3.7 x 106 6.2 x 106 1.2 x 10 7
2.2 x 107 4.4 x 107
~
50-100 6.2 x 106f 1.3 x 10 7
7.6 x 107 1.5 x 108 7f 3.9 x 10 100-150 2.0 x 10 150-200 4.2 x 107f 8.6 x 107 1.8 x 108 3.7 x 108 aIncludes both nodules and ablated thyroids. Approximately 4% of the thyroid nodules will be fatal.
bApproximately 80% of the reduced thyroid damage cases are ablated thyroids,19% are nodules and 1% are thyroid cancer fatalities (from Table 7).
cApproximately 70% are ablated thyroids, 29% are nodules and 1% are thryoid cancer fatalities.
dApproximately 40% a.7 ablated thyroids, 58% are nodules and 2% are thyroid cancer fatalities.
roximately 10% are ablated thyroids, 86% are nodules and 4% are thyroid cancer fatalities.
e Approximately 96% are nodules and 4% are thyroid cancer fatalities.
~
Table 10. Cost-Benefit Agalysis for Use of KI by 0111dren. Assumptions: risk coefficient = 668 thyroid nodules per 10 person-rem to thyroid,a no 0.1 dose effectiveness factor for I-131, Core Melt Atmospheric accident category only, RSS accident probabilities.
D
'1hyroid Nodules (mean)c Distance Normal 3
Interval Normal Activity cost-Benefit Ratio j
(miles)
Activity 99% KI Potential Reductionc Reduction (nodules /yr)c (S/ nodule prevented) 0-5 270 91 179 2.5 x 10-3 3.2 x 105 5-10 625 157 468 6.5 x 10-3 3.7 x 105 10-25 2510' 361 2150 3.0 x 10-2 5.3 x 105 l
25-50 4190 386 3800 5.3 x 10-2 1.1 x 106 50-110 5930 467 5460 7.6 x 10-2 3.2 x 106 100-150 3170 238 2930 4.1 x 10-2 9.5 x 106 150-200 1980 151 1830 2.6 x 10-2 2.1 x 107 aIncludes age dose factors and risk coefficients from RSS (see Section 3).
bIncludes both nodules and ablated thyroids.
2 cAsstunes a uniform population density of 100 persons / mile,
1 P
e g
.. - ~
Finally, Table 11 summarizes an identical analysis performed for children using the APS upper bound risk coefficient of 6500 6
thyroid nodules per 10 person-rem to the thyroid.
In this case, 5
the estimated cost-benefit ratios range from 4.9 x 10 S/ nodule 6
prevented within 0-5 miles to 2.2 x 10 S/ nodule prevented within 150-200 miles.
Note that the ratio for the 0-5 mile interval i
is actually higher than in Tables 9 and 10.*
i The cost-Denefit ratios given in eacn of the preceding tables were calculated for selected distance intervals from a single reactor.
However, if tnere were two reactors at a patti-
]
cular site, the probability of an accident at that site would be twice as high and tne cost-oenefit ratio for each distance interval would be a factor of 2 lower.
Similarly, in many areas of the U.S.,
several reactors at dif ferent sites may contri'o 2te to an individual's risk of thyroid damage.
The extent to which tnis would reduce the cost-benefit ratio for KI depends on a number of factors, including the specific location with respect to neighboring plants, winc direction frequencies, reactor power levels, etc.
For example, there are approximately 13 reactors **
currently operating witnin 200 miles of New York City.
Using i
4 i
- For this assumed risk coefficient, the thyroid dose is still i
nign enough to cause significant numoers of thyroid nodales, i
even with 99% effective KI.
1 i
l
- Reactors (power level > 200 MWe) within 25-50 mile interval:
Indian Point 2 and 3; 50-100 miles:
Oyster Creek, Haddam Neck, I
Millstone 1 and 2; 100-150 miles:
Salem, Vermont Yankee, Peach Bottom 2 and 3; 150-200 miles:
Tnree dile Island 1 and 2, I
Pilgrim.
hoe D
- D ww w
57 I
---e.
.,me a
w w
,e w.
w
,, - - = -.,- -,
2-s-
---w rea--+-
tw
E Table 11.
Cost-Benefit Analysis for Use of KI by 0111dren. Assumgtionn: APSa upper-bound risk coefficient for children of 6500 thyroid nodules per 10 person-rem to thyroid,b no 0.1 dose effectiveness factor for I-131, Core Melt Atmospheric accident category only, RSS accident probabilities.
c
'Ihyroid ty) ules (mean Distance Normal Interval "ormal Activity Cost-Benefit Ratio d
(miles)
Activity 99% KI Potential Reduction Reduction (nodules /yr)d
($/ nodule prevented) 0-5 374 262 112 1.6 x 10-3 4.9 x 105 5-10 1020 586 434 6.1 x.10-3 3.9 x 10 5 10-25 5590 2430 3160 4.4 x 10-2 3.6 x 105 25-50 12,600 3500 9100 1.3 x 10-1 4.5 x 105 50-100 31,600 4530 27,100 3.8 x 10-1 6.3 x 105 100-150 28,400 2320 26,100 3.7 x 10-1 1.1 x 106 150-200 19,300 1470 17,800 2.5 x 10-1 2.2 x 10 6 American Physical Society [9].
bIncludes age dose factor of 5.0.
cIncludes both nodules and ablated thyroids.
d 2
Assumes a uniform population density of 100 persons /miie,
N
l the data provided in Table 9 above, and ignoring wind direction frequencies and differences in reactor power level and design, the cost-benefit ratio specific to New York City can be estimated to be approximately a factor of 4 lower than if only the nearest reactor (Indian Point 1 or 2) was considered alone.*
Similarly, for the city of Chicago (which has more than 10 operating plants within 200 miles), the cost-benefit ratio is approximately five times lower than the ratio if only a single reactor was considered.
7.
Risk-Benefit Analysis As discussed in Section 2, the risk posed by the use of KI as an emergency protective measure for reactor accidents was judged by the NCRP to be minimal.
Nevertheless, a brief analysis is presented here to determine under what conditions, if any, the risk posed by the drug might outweigh its potential benefits.
a Assuming a risk of adverse reaction of 10-6 per 130 mg tablet of KI (see Section 2) and that a total of 10 tablets would be administered to each person following an accident, the risk posed to that person by the drug equals 10-5 To estimate the thyroid dose for which the potential benefit (reduced risk
- From Table 9, for normal activity and no 0.1 I-131 dose effectivenessfactor,NYgcost-benefitratioforasingleIndian Point reactor = 2.0 x 10
$/ thyroid nodule.
Including all 13 reactors:
l 1
2' 4
4 3
=
,+
,+
cost-benefit ratio 2.0x10o 6.2x10o 2.0x10 +
4.2x107 7
and cost-benefit ratio = 5.2 x 105
$/ thyroid nodule.
59
~
=....
of nodule occurrence \\ and risk of KI are equivalent, the follow-ing additional assumptions are made:
risk coefficient for individual 2 3.34 x 10-4/ rem, no 0.1 dose effectiveness factor for I-131, and 99 percent effective
- use of KI reduces total thyroid dose by 90 percent.**
Then 10-5 = 0.9 x (3.34 x 10-4/ rem) l x (equivalent dose), and the equivalent dose = 3 x 10-2 rem.
What if other assumptions are made?
Iligher risk coef ficients, such as those for children (see Section 3), would result in lower predicted equivalent doses.
The administration of KI to everyone i
within 360' of a site, rather than only to exposed persons, would increase the equivalent dose.
For example, if the radioactive plume was 15 wide, the equivalent dose would be a factor of 24 (i.e.,
360/15) higher ***
(= 0.8 rem).
Assuming only 50 percent effective KI (rather than 99%), as well as 360' admi.nistration, f
the equivalent dose would become 2 rem.
Finally, if a 0.1 dose effectiveness factor for I-131 is also assumed, the equivalent dose is increased to approximately 5 rem.****
- 99 percent reduction in dose from inhaled radioiodines.
- Actual ~ percentage reduction depends on the composition of the release.
For the accident categories addressed in this study, roughly 90 percent of the thyroid dose is due to inhaled radiciodines (see Table 5).
i
- 24 times as many individuals would now take the drug.
The i
adverse. reaction risk would therefore be 24 times higher.
l
- I-131 contributes approximately 75 percent of the dose from l
inhaled iodines (see Table 5).
With a 0.1 dose effectiveness i
factor, the effective dose from inhaled iodines is reduced i
by a factor of (0.75)(0 1) + (0.25) = 0.33.
The potential l
bengfit of 50 percent effective KI = 0.9 (0.33)(0.5)(3.34 x 10- )(equivalent dose).
Setting this equal to 24 (10-5), the l
equivalent dose = 5 rem.
t 60
~r Q h 4:y em s m
/
,Y,.al
- \\'\\\\)'%[htf<g'gf
% ' f>_ ii'
/g, e v},j#g,e
\\/
yp f
Q
- ve,
/g%*+,
V
- s imieE Ev l9
/,jj'/j fs
/\\'*$t9)?
///g'(A;f 9
Ny///
6*(p
'O jp g77 V+,,9
/4,N XV
IMAGE EVALUATION TEST TARGET (MT-3) i i
1.0 s a E22
[ yl El l
1.1
[m llLE
\\\\M=
l.25 IA 1.6 1
1 i
MICROCOPY RESOLUTION TEST CH ART h'
sp %
s2
%g Ab+a//p
/
y
+ g;,))gr/w
- > 4 A/4
< g y-x-
-r__
o o
i J
The range of equivalent doses calculated above for various l
assumptions are all below the level' recommended by the NCRP for use of KI (10 rem,'see Section 2).
Therefore, at the recom-i the risk posed by the drug does appear to be small mended level, compared to its potential benefits.*
However, several recent reports suggest that the risk associated with the drug may be significantly higher than.10-6 per dose for certain segments of the population.23,24 T.f this is confirmed, the risk-benefit conclusion for KI would have to be reassessed.
8.
Summary, Conclusions and Recommendations This study was undertaken to provide guidance to policy-makers concerning the use of potassium iodide (KI) as an emergency protective measure for reactor acci. dents.
Although,the effective use of KI could significantly reduce the number of thyroid nodules resulting from a serious accident, it would have no, or only-minor, impact on other accident consequences; including immediate deaths or injuries, delayed cancer deaths, and long-term land contamination.
Therefore, the availability of KI would provide only a supplemental strategy to be considered along with other possible protective measures.
f The study was performed using the Reactor Safety Study (WASH-1400) consequence model, CRAC.
Four categories of accidents were addressed:
gap activity release accidents (GAP), GAP without If the adverse reaction risk was 10-7 rather than 10-6 per dose (see:Section 2), the risk posed by XI would be minimal compared to its potential benefits.
~
61
=
.i I
J containment isolation (GAP w/o Isolation), core melt with a l
melt-through release (Core Melt Melt-Thrc. ugh) and core melt i
J with an atmospheric release (Core Melt Atmospheric).
A series of thyroid dose calculations showed that the GAP category does not pose a significant health hazard to the public at any dis-I tance from the reactor.
For the GAP w/o Isolation and Core Melt l
Melt-Through categories, doses in excess of recommended protec-tive action luidance levels (PAGS)(5-25 rem) are confined to areas within approximately 10 and 15 miles of the reactor, respectively.
I For the Core Melt Atmospheric category, however, thyroid doses are likely to exceed PAGs out to 100's of miles.
A cost-benefit analysis for the use of KI was also performed, the results of which are summarized in Table 12.
Cost-benefit i
ratios ($ per thyroid nodule prevented) are presented for selected distance intervals, assuming that no other protective measures are taken.
j The ef fect of evacuation and sheltering on the predicted ratios is shown in Table 9 and is discussed in Section 5.
Evacua-tion has the potential to be 100% effective in reducing all dose if accomplished before arrival of the radioactive cloud.
Shelter-4 ing was assumed in this analysis to provide a factor of 2 reduction in thyroid dose.
Therefore, in both cases, the thyroid dose reduc-tion afforded by the supplemental use of KI would be reduced, and the KI cost-benefit ratios presented in Table 12 would be corres-
}
pondingly increased.
The uncertainties in the estimated cost-benefit ratios are r
very large.
Key assumptions made in deriving the ratios are i
noted in Table 12.
The KI was conservatively assumed to be 99%
62
Table 12.-
Summary Table for KI Cost-Benefit Analysisa,b (from Table 9)
Normal Activity Distance Interval Cost-Benefit Ratio (miles)
(S/ thyroid nodule prevented 0-5 3 x 105 5-10 4 x 105
~
10-25 7 x 105 25-50 2 x 106 50-100 6 x 106 100-150 2 x 107 150-200 4 x 107 aKey Assumptions 1.
99% effective KI (i.e., all persons take drug before cloud passes).
2.
No other protective measures are taken.
3.
WASH-1400 accident probabilities.
4.
Estimated cost of KI program = $0.10 per person per year.
Assumed cost includes only the purchase price of KI, i.e., no costs for distribution, monitoring and administrative expenses.
5.
Only 1 reactor (3200 MWt PWR) within 200 miles.
6.
WASH-1400 dose-effects coefficients (no 0.1 effectiveness factor for I-131 dose).
buncertainties are large and scale approximately linearly with assumed KI effectiveness, accident probabilities, cost, multiple reactors, and dose-effects coefficients.
O
effective (i.e., all persons take the drug before the cloud passes).
Realistic effectiveness values could be significantly smaller.
'iAS!!-1400 accident probabilities were assumed.
Prob-I ability uncertainties have been estimated to be at least an order of magnitude (see Section 6).
Estimated costs for a KI program were conservatively based on only the purchase price of the drug and did not include costs for distribution, monitoring, and administrative expenses.
The ratios presented in Table 12 are j
appropriate if there is only a single reactor within 200 miles.
Many actual sites would be influenced by several reactors, and cost-benefit ratios could be reduced by factors of 2 to 5 (see Section 6).
Uncertainties in dose and health effects parameters are also large and could result in either higher or lower cost-benefit ratios.
To some extent, the large uncertainties in the above assump-a tions hinder our ability to provide definitive guidance.
Never-i theless, for the assumptions made, the calculated cost-benefit ratios are high; and even including uncertainties, KI appears to be only marginally cost-ef fective, at best.
- Finally, using statistics provided by the NCRP4, a simple l
risk-benefit analysis showed the risk of adverse reaction posed by KI'at the recommended action levels and dosages to be small compared to its potential benefits.
Iloweve r, several recent
- Although tL2 total cost associated with a case cf thyroid nodules was not specifically addressed, an approximate upperbound of
$17,000 can be inferred from the information presented in refer-ence 25 assuming 1) average hospital care costs of $2,000,
'2) that hospital costs are 60% of 'all direct costs, and-3) that-indirect costs (economic losses due to mortality and morbidity) are 4 times higher than direct costs.
64 j
l --
~
reports suggest that there..is a significantly higher risk j
associated with use of the drug among certain segments of the l
- population.23,24 If this is confirmed, the risk-benefit conclu-sion.for KI would have to be reassessed.
l Based on the above analysis, the following additional i
recommendations and comments are made:
The risk of thyroid nodules was shown to be dominated by the large releases associated with core melt j
accidents in which the containment fails directly to the. atmosphere.
Therefore, if design modifications, such as filtered centainment venting systems, are f
implemented to reduce the likelihood of those releases, the potential benefit of KI could be substantially reduced.
{
Before any KI program is implemented, specific alterna-tive strategies for stockpiling and distributing the i-j
-drug should be examined to reduce costs and assure effectiveness.
The use of common household items (e.g.,
handkerchiefs and towels) as respiratory filters may provide signi-1 i
ficant additional protection against dose due to inhaled i
. radionuclides and shoald be considered further in the i
development of protective strategies.
If a KI program is implemented, responsible government agencies should give priority to establishing guidance j
(PAGs) concerning when, or under what conditions, the drug should be used.
65 i
e
--,,,.c
Finally, whether or not a public KI program is imple-mented, it might be wise to have sufficient quantities of the drug available at or near reactor sites for use by 1) site personnel, 2) offsite emergency response personnel, and 3) controlled populations in offsite institutions (e.g.,
hospitals, prisons) where immediate evacuation would be difficult or infeasible.
8 66
p References 1.
Reactor Safety Study Appendix VI:
Calculation of Reactor Accident Consequences, WASH-1400 (NUREG 75/014),
U.S.
Nuclear Regulatory Commission, October 1975.
2.
- Aldrich, D.
C.,
P.
E.
McGrath and N.
C.
Rasmussen, Examina-tion of Offsite Radiological Emergency Protective Measures for Nuclear Reactor Accidents Involving Core Melt, SAND 78-0454, Sandia Laboratories, Albuquerque, NM (1978).
3.
Jan Beyea, Some Long-Term Consequences of Hypothetical Major Releases of Radioactivity to the Atmosphere from Three Mile Island, Draft Report to the President's Council en Environmental Quality, Center for T.nergy and Environmental Studies, Princeton University, September 1979.
4.
Protection of the Thyroid Gland in the Event of Releases of Radioiodine, NCRP Report No. 55, National Council on Radia-I tion Protection and Measurements, August 1977.
l, 5.
Personal communication with Dr.
G.
N.
Kelly, National Radio-logical Protection Board, Harwell Didcot, United Kingdom.
6.
" Accidental Radioactive Contamination of Human and Animal Feeds and Potassium Iodide as a Thyroid-Blocking Agent in a Radiation Emergency," Department of Health, Education and Welfare, Food and Drug Administration, Federal Register, Friday, December 15, 1978, part VII, p.
58790.
7.
Manual of Protective Action Guides and Protective Actions for Nuclear Incidents, EPA-520/1-75-001, September 1975, U.S.
Environmental Protection Agency.
8.
H.
W.
Lewis, et al., " Risk Assessment Review Group Report to the U.S. Nuclear Regulatory Commission," NUREG-CR-0400, September 1978.
9.
" Report to the American Physical Society by the Study Group on Light-Water Reactor Safety," Review of Modern Physics, 47, 1975.
10.
- Wall, I.
B.,
S.
S.
- Yaniv, R.
M.
- Blond, P.
E.
- McGrath, H.
W.
Church, and J.
R.
Wayland, Overview of the Reactor Safety Study Consequence Model, U.S.
Nuclear Regulatory Commission, HUREG-0340 (1977).
11.
P.
E. McGrath, D.
M.
Ericson, and I.
B. Wall, "The Reactor Safety Study (WASH-1400) and Its Implications for Radio-logical Emergency Response Planning," International Symposium on the Handling of Radiation Accidents, 28 February 1977, Vienna, Auctria, IAEA-SM-215/23.
67
12.
D.
C. Aldrich, D.
M.
Ericson, Jr., and J.
D.
Johnson, Public Protection Strategies for Potential Nuclear Reactor Accidents:
Sheltering Concepts with Existing Public and Private Structure, SAND 77-1725, Sandia Laboratories, Albuquerque, NM (1977).
13.
Planning Basis for the Develop;uent of State and Local Govern-ment Emergency Response Plans in Support of Light Water Nuclear Power Plants, U.S.
Nuclear Regulatory Commission and 'nviron-mental Protection Agency, NUREG-0396, EPA 520/1-78.
o, 1978.
14.
- Aldrich, D.
C.,
R.
M.
Blond, and R.
B.
Jones, A Model of Public Evacuation for Atmospheric Radiological Releases, SAND 78-0092, Sandia Laboratories, Albuquerque, NM, June 1978.
15.
- Aldrich, D.
C.,
L.
T.
Ritchie, and J.
L.
Sprung, Effect of Revised Evacuation Model on Reactor Safety Study Accident Con-sequences, SAND 79-0095, Sandia Laboratories, Albuquerque, NM, February 1979.
16.
- Aldrich, D.
C.,
D.
M.
- Ericson, Jr.,
R.
B.
- Jones, P.
E.
McGrath and N.
C.
Rasmussen, " Examination of Offsite Emergency Protec-tive Measures for Core Melt Accidents," ANS Topical Meeting on Probabilistic Analysis of Nuclear Reactor Safety, May 8-10, 1978, Newport Beach, CA.
17.
- Aldrich, D.
C.,
and D.
M.
Ericson, Jr., Public Protection Strategies in the Event of a Nuclear Reactor Accident:
Multi-compartment Ventilation Model for Shelters, S AND7 7 -15 5 5, Sandia Laboratories, A1 )uquerque, NM, January 1978.
18.
A.
F.
- Cohen, B.
L.
- Coheq, D.
C.
Aldrich (ed.), Infiltration of Particulate Matter into Buildings, SAND 79-2079, Sandia Laboratories, Albuquerque, NM, l'ovember 1979.
19.
al.
G.
- Guyton, H.
M.
De cke r and G -
T.
Auton, " Emergency Res-piratory Protection against Radivlogical and Biological
- Aerosols, A.M.A.
Arch. Ind. Health 20, 91-95 (1959).
20.
Respiratory Protective Devices Manual, Am. Industrial Hygiene Association, American Conference of Government Industrial Hygienists, 1963.
21.
F.
L.
Leverenz and R.
C.
Erdmann, " Comparison of the EPRI and Lewis Committee Review of the Reactor Safety Study,"
prepared for Electric Power Research Institute by Science Applications, In :., EPRI Report NP-ll30, July 1979.
22.
"World List of Nuclear Power Plants," Nuclear News, Vol. 22, No. 10, August 1979.
23.
Curd, John G.,
et al.,
" Potassium Iodide Sensitivity in Four Patients with Hypocomplementemic Vasculitis," Annals of Internal Medicine, December, 1979, Vol. 91, No.
6, pp 853-857.
68
. =
t 24.
Rosenstein, Beryl J.,
et al.,
Iodide-Induced Ilypothyroidism Without a Goiter in an Infant with Cystic Fibrosis," Journal l
. of Pedriatrics,. August, 1978, Vol. 93, No.
2, pp 261-262.
i 25.
Scotto, Joseph and Leonard Chiazze, " Cancer Prevalence and lIlospital Payments," J.
Natl. Cancer Inst., Vol. 59, No.
2,
. August, 1977, pp 345-3498.
i 4
d I
4 i
L e
b i
9 4
1 69-70 y
7,-
3 y
y..y.
.-y..,
,....r,
,m,.._,,y___
d i
3
' DISTRIBUTION:
i
' Roger Blond (50)
Probabilistic Analysis Staff Office of Nuclear. Regulatory Research
_ USNuclear Regulatory Commission
- Washington, DC 20555
.A.
Bayer i
INR-Kernforschungszentrum Karlsruhe 75 Karlsruhe 1 Postfach 3640 l-West Germany i
Douglas Cooper Department of Environmental Health Physics liarvard School of Public Ilealth 663 Iluntington Avenue Boston, MA 02215 i
Keith Woodward Picard &'Lowe Associates 1200 18th Street NW Washington,. DC 20036 j
Joseph Logsdon US Environmental Protection Agency AW 461 j
401 M Street NW Washington, DC 20024 l
S.
Chakraborty Abteilung fur die Sicherheit der Kernanlagen Eidgenossisches Amt fur Energiewirtschaft CH-5303 Wurenlingen i
Switzerland f
N.
Rasmussen l
Nuclear Engineering Department Massachusetts Institute of Technology Cambridge, MA 02139 i
Dr. Ulf Tveten Institutt for Atomer2rgi 4
PO Box 40 2007 Kjeller Norway Dr..Seppo Vuori Valtion Teknillinen Tutkimuskeskus Statens Tekniska Forskningscentral helsinki:
Finland 1
71
Dr. Rosalyn S.
Yalow Montefiore Hospital and Medical Center 111 East 210th Street Bronx, NY 104u7 Dr. Frank von Hippel Center for Environmental Studies Princeton University Princeton, NJ 08544 Dr. Thomas Fearon National Councail on Radiation i
Protection and Measurements 7910 Woodmont Avenue, Suite 1016 Washngton, DC 20014 Prof. Richard Wilson Department of Physics Harvard University Cambridge, MA 02138 H.
Arnold Secretary of Health Commonwalth of Pennsylvania 802 Health and Welfare Bldg.
Harrisburg, PA 17120 Dr.
H.
L.
Gjorup Health Physics Department Riso National Laboratory DK-4000 Roskilde Denmark Prof.-Niel Wald Graduate School of Public Health University of Pittsburgh RC510 Scaife Hall Pittsburgh, PA 15261 Brian Grimes (10)
Office of Nuclear Reactor Regulatory US Nuclear Regulatory Commission 545 Phillips Washington, DC 20555 J.
Schutz-Larsen Arvebiologisk Institut i
Tagensverj 14 KD-2200 Copenhagen N Denmark i
72L 1
---~r
-n
.,m v.
---g
.n
I Bernard Cohen University of Pittsburgh Faculity of Arts and Sciences Department of Physics Pittsburgh, PA 15260 Pfn. Dade Moeller (3)
Department of Environmental llealth Physics l
Ilarvard School of P 'lic Ilealth 663 Iluntington Avenue Boston, MA 02215 l
l Dr. Bernard Shleien (5)
Bureau of Radiological llealth 11FX-4 60 5600 Fishers Lane-Rockville, MD 20857
~
Dr. Eugene L.
Saenger Eugene L. Saenger Radioisotope Laboratory Cincinnati General llospital 234 Goodiaan Street Cincinnati, Oil 45267 Dr.
F.
IIorsch Projekt Nukleare Sicherheit i
Kernforschungszentrum Karlsruhe Gmbli
{
Postfach 3640 i
D-7500 Karlsruhe 1 West Germany Mr.
T.
Iijima Division of Reactor Safety Evaluation Tokai Research Establishment Japan Atomic Energy Research Institute Toka i-lau ra Ibaraki-ken 911 13 Japan Jim Martin Probabtlistic Analysis Staff US Nuclear Regulatory Commission Washington, DC 20555 Dr.
G.
N.
Kelly (3)
National Radiological Protection Board lia rwell, Didcot
-Oxfordshire OX 11 ORQ England 73
4 Mr.
C.
Devillers Chef du~ Service d' Etudes de Surete l
Radiologique et des Sites-
. Centre d' Etudes Nucleaires BP No. - 6 F-92260 Fontenay-aux-Roses France I
E. C. Watson dat,telle Pacific Northwest Laboratories Battelle Boulevard.
1Richland, WA 99352 Dr. Ian Wall r
Nuclear Safety and Analysis Department Electric Power Research Institute 3412 Hillview Avenue PO Box 10412 Palo Alto, CA 94303 Pfo. Yasushi Nishiwaki Jagdschlossgasse.9.1 A-ll30 Vienna Austria Dr.-Jan Beyea National Audubon Society 950 Third Avenua New York, NY 101?'
ilarold Collins (100) i-Office of State Programs.
US Huclear Regulatory Commission 7713 MNBB dashington,- DC 20555 i
Dr.
G.
D. l'aise r j
United'l;ingdom AToinc Energy Authority Safety and Reliability Directorate Wigskaw Lane i
Dulcheth Warrington WA 3 4NE ~
l Chesnire England v
l 74
\\
1233 J. M. Taylor
. 4000 A..Harath 4400
. A. W.
Snyder
- 4410 D.
J.'McCloskey Attn:
J. W. Ilickman JU.'B.
Varnado L.
D.LChapman 4413 N.
R.
Ortiz 4413 D.-C. Aldrich (100).
4413 N.~ C. Finley 4413 J.
D. Johnson 4413 L.
T.
Ritchie-4414 D. M.
Ericson, Jr.
3141 T.~L.. Werner-(5) 3151 W. ' L. Garner (3) for DOC / TIC (Unlimited Release) 3154-3
-R.
P.
Campbell -(25)
.for NRC Distribution to NTIS 8266' U.
A.
has d
i f
i i
1 f
D -
a 75 i
._