ML20126H287

Affidavit in Response to Ld Hamilton 810309 Affidavit Re Low Level Radiation Risk.Ld Hamilton Work Is Fraudulent. Affidavit & Ltrs Encl

ML20126H287
Person / Time
Site:	Allens Creek File:Houston Lighting and Power Company icon.png
Issue date:	03/24/1981
From:	Bross I J AFFILIATION NOT ASSIGNED
To:
Shared Package
ML20126H286	List: ML20126H281 ML20126H287
References
NUDOCS 8104080409
Download: ML20126H287 (27)

Text

-

O ATFIDAVIT OF IEWIN D.J. BROSS IN PE.SPONSE 'N)

AN AFFIDAVIT OF LEONARD D. HAMILTON (MARC'H 9,1981)

Non-scientists confronted with my scientific report, "A 1981 Reassessment of the Health Hazards of Low-Level Ionizing Radiation", and ,

the affidavit by Leonard D. Hamilton of March 9, 1981, might seem to be in the classic dil===a Lat to do when experts disagree?

Fortunately in this instance there are three simple ways to resolve the dilamma. The first resolution is possible because Hamilton and Bross are in complete agreement on one basic point: There are not two experts here, there is one real expert and one fraud. The only question then is, which is which?

.wong other compliments, Hamilton says that Bross's work is among the "few studies that have been roundly criticized for reasons ranging from dishonesty to poor statistical methods" (p. 7) and "have been largely discredited by the reputable scientific consnunity" (p.14) .

On the other hand, "A 1981 Reassessment of the Health Hazards of Low-Level Ionizing Radiation" reviews more than 40 scientific reports I

which have the following characteristics in commons (1) They are studies of human populations who were actually exposed to low doses (0.1 rem to 10 rems) of ionizing radiation (in contrast to the data on high 4ose exposures used in all of the reports Hamilton cites at the top of page 3).

(2) The studies are biostatistical-epidemiologien's studies involving the use of statistical analysis of data on ac'. .adity or mortal-ity.

810 4080Mf '

c -

s l

l (3) Most of the cited studies have been reported within the past five years.(in contrast to the long-outdated studies.used in the  ;

reports cited by Hamilton on page 14). .

Since the area of expertise required for a scientific evalu-ation of the current statistical evidence on low-level radiation hazards  ;

is training, experience, and demonstrated competence in carrying out t

studies with the al~.. ;.:ntioned characteristics, it is not hard to l distinguish the real expert from the fraud. Consider the factual infor-nation on Irwin Bross and Leonard Hamilton.

(1) One has a Ph.D. in biostatistics, has headed a Department i i

of Biostatistics at an internationally famous cancer research institute _

l l

(RPMI) for over 20 years, and is a long-time fellow of the American [

Statistical Association. The other has no statistical qualifications.  !

i (2) one has published over 300 scientific papers on epidemi- l ology, biostatistics, and public health, has just had a book (SCIENTIFIC  !

STRATEGIES TO SAVE YOUR LIFE) published by a reputable scientific publish-er (Dekker, New York, March 1981) on these specific issues. The other

~

has never published a single original and competent paper in biosta-tistics, epidemiology, or public health.

(3) On March 17, 1981 in Cincinnati, Ohio, both Hamilton and ,

Bross were at a public meeting of the scientific advisory committee for the Center for Disease control / National Institute for Occupational i Safety and Health (CDC/NIOSH) in the federal building in Cincinnati, ,

Ohio. Dr. Karl Z. Morgan, Dr. Thomas Mancuso, and Dr. Irwin Bross f had been originally named to the panel by Congress in recognition of their impcrtant scientific contributions to research on Iow-level .

t l .

-,_m._ . , _ . . , _ . , . . . _ _ . . _ . _ _ _ . . . ~ _ . , . _ , _ , _ _ , , _ . , , _ - , - . ~ . - _ ~ . _ _ . , , , . . . - ~ , . . . . . . , _ . , , _

ionizing radiation (Sea Serial No.95-179). Subsequently the Atomic Industrial Forum, the official lobby of the nuclear industry, insisted upon representation on the panel and Leonard Hamilton was appointed as their lobbyist.

(4) At the meeting of March 17, 1981 (which might be said to involve the reputable scientific community), Hamilton's repeated inter-ruptions of Bross specifically brought up*the queation of scientific credentials. On several occasions, Bross referred to Hamilton's counplete lack of qualifications (e.g. , " amateur statistician".) , but no one at this reputable scientific meeting (not even Hamilton) questioned this ,

assessment. Nor did anyone question Bross's statement that he was the only professional statistician on the panel. Bross's standing in the international scientific comununity is evidenced by the introduction used by the New Scientist for a recent article (Attachment A) .

On the basis of this factual evidence, it should not be diffi-cult to decide who is a scientist and who is a ichbyist, who is a real expert and who is a fraud.

. A second way to resolve the dilemma directly involves the Motions for Summary Disposition since this puts the burden of proof on l

NRC. If there is prima facie evidence of serious adverse health effects from normal operations of a nuclear installation under NRC AIARA standards, this evidence should be heard. It is not necessary at this point that the evidence be conclusive.

On page 3 of his affidavit, Hamilton refers to 'the fatal flaws in the Na$arian and Colton studies on the portsmouth Naval Shipyard" t

(Affidavit pp. 8-13) to deny the conclusions in the original Bross

(

l

_ -_ - . , , - . . ,, m _- -- - --. - - ,

4 affidavit. Since the shipyard was operating under NRC ALARA standards, confirmation of the Najarian-Colton findings would be direct evidence that the NHC permissible levels represent a serious hasard to the health of nuclear workers. Since the permissible levels for the public are set on a similar basis, there would also be hazard to.the public. This

, l would also flatly refute the reports relied on by the NRC Staff in their motion for summary disposition (Affidavit p. 14) and in the Gotchy affidavit.

For the estimates from the reports cited on page 14, it would have been impossible to find excess risks of leukemia and other cancers at the Portsmouth Haval Shipyard. If, as will now be shown, clear scientific evidence of such risks were found in the CDC/NIOSH data on PNS, this is prima facie evidence from a federal study.

The scientific evidence of serious health hazards can be found in CDC/NIOSH documents prepared for the March 17, 1981 meeting .in Cincinnati previously cited in connection with credentials. All state-ments can be verified from the CDC/NIOSH documents and tapes of this i ..

meeting. In particular the " fatal flaws" to which~damilton refers on page 3 were discussed at some length. Although Hamilton repeatedly claimed that the Najarian-Colton 1978 report had been specifically refuted, he was all g himself. None of the other'four panel h rs agreed with this axtremist position.

In an attempt to refute a statistical analysis by Dr. Colton, CDC/NIOSH belatedly did some analyses of its own data. The result a.s described briefly in an article entitled "A False Report on RaaLation

. . . - - . - _ - - - .- .. ~ _ . . - - . - .

l Hazards at a Nuclear Shipyard" prepared for publication by Bross (Attachment B). A photocopy of the analysis in which CDC/NIOSH inad-vertently confirmed the Najarian-Colton report on leukemia by getting a ,

statistically significant result (p = 0.0464) is given as Attachment C-and refutes Xmmilton's assertions. .

From the March 17 documentation it is clear that the original Najarian-Colton study of the PNS workers was essentially correct, that the citation of this study by Bross was warranted, and that the " fatal r flaws" claimed by Hamilton exist only in his own mind. It should be noted that this exchange brings out clearly the strong scientific evi-dance of serious health hazards from radiation exposures at a normally-operating nuclear installation that conformed strictly to all NRC regu-lations. This is far more than the prima facie evidence needed to throw out the motion for suusnary disposition by the NRC staff.

This evidence was produced by CDC/NIOSH, a federal agency, but ,

the reluctance of the agsney to report the true facts to Congress or the l

the public is evident from Attachment B. This or.ly emphasizes the noed 6

for a full consideration of the more than 30 positive studies of low-level radiation hazards cited in the original.Bross affidavit (and attacked in the Hamilton affidavit of Narch 9,1981) since none of_ this I important new scientific information has been used g the obsolete risk i estimates : relied upon by the NBC staff M their cost-benefit calculations.

i

?

The CDC/NIOSH lung cancer data, for* instance, shows risks approuimately l

200 times greater than the BEIR I and III estimates, more than enough to reverse the cost-benefit ratios, a point made in a recent letter to Presidaat Reagan (Ateme h t D)..

f l.

{-_._.._ _ ._ _ _ ...._ _ _ _ _ __ _ _ _ . _ _ _ _ _ _

l

_ _ . _ . . _ , _ . . . . _- - .. . m - _ _ _ _ _ ,

t I

_o.

Finally, the third way.to resolve the dileoma on disagreement  ;

i between experts is to look at the specific points raised by Hamilton in his affidavit. Because of the foregoing, this can be done as briefly as j i

possible.

l The claim on page 2 that "Dr. Bross has no.true facts upon ,

which to base his assertion" can be refuted by simply going to the list-of references in the original affidavit, noting that there are more than 30 positive studies. Each involves thousands of persons exposed to low- ,.

level radiation and each involves "true facts" that have been disparaged but not denied.

Namnton's claim on page 2 to speak for the " scientific com- ,

munity" is rsfuted by the March 17 meeting where-he wao the only member of the panel to hold his extremist views. l

' The claim that ?Bross has completely. misrepresented the data he has used" and the " critique" on pages 9-13 is based on nothing more than technical-sounding doubletalk. Hamilton expects to get away with this fraud because non-scientists cannot understand these technicalities.

However, the public can judge who is the real expert and who is-not and  ;

can find a full discussion of these matters in plain English in Bross's ,

new book, SCIENTIFIC STRATEGIES ~TO SAVE YOUR I.IFE.

i A

The same book answers in detail the rhetorical questions about F hypotheses and definition of doubling dose (p. 9) . Hamilton doesn't understand that adjustments are handled more. scientifically in a mathe-natical model than by his Mickey Mouse arithsetic (and theylhave been made this way) . All that is shown on yp. 9-13 is a total lack of sta-tistical competence on mentiton's part ~(a common ~1ack in the radiation k

i-L

,i=,.,,y ,,-~~n--. w- ., . , we n - -y-- , . , , , -,,,,r. .h%u%..,,..,....m...- _ . , , . w w.m'.e-e--n _,-.u~...E.6,. .,, . ,,, . , .w.-.-...-.m~ c .

6 protection casuunnity).

Finally, Hamilton says "By ignoring all negative studies, Bross has limited his universe and, therefore,-skewed his results greatly".

As was pointed out repeatedly (and can be easily verified), the list of studies cited in Bross's affidavit to Reference 34 were all of the relevant scientific reports cited in the Libassi list of references plus a few others omitted from that list. The selection was made by the Interagency Task Force, not Bross. References 41 to 47 updated the listing. Note that references 41 and 43 are negative reports. Clearly >

Hamilton made his fmise assertion without bothering to actually look at ,

the references. Once this claim of " systematic error" is refuted by the ,

actual reference list, it might be noted that Hamilton's admission still stands: " The ' bundle of frail reeds' argument is correct when the individual reeds suffer only from random error." Thus, Hamilton's affidavit, when corrected for misstatements, actually sups. orts the arguments in the original affidavit.

s m.u -

p,m+.m - g. , , . .p, =

..yiy. a.9%y.. . , ..e, , ,.,, . , ,ygsg,g,, ywg ,_7.,y. g%.,,, ,y9 my , ,.%,.9 -p-.g n e -,-w-- w. v-g- -g.,,-

1 LIST OF ATTACHMENTS i

Attachment A - Muddying the Water at Niagara (Published in the New Scientist, Vol. 88, pp. 728-729, 11 December 1980).

Attachment B: A False Report on Radiation Hazards at a Nuclear shipyard. (Material submitted for publication).

Attachment Cs Photocopy of CDC/NIOSH printout, March 17, 1981.

Attachment D Letter sent to President Ronald Reagan, March 12, 1981.

---wu4

+ - P-y -- .-vm e- V --

9e wr1

ere!~ - _P _,.

~

a*

9

! J ini .1]

Muddying the water at Niagara $J"il %';A4""f/i, ;"Yln',,%

histories before and after moving to

% g 4 g,, Love Canal. There are no documented  :

birth defects in 57 live births prior to  !

T OVE CANAL, a suburb of Niagara Falls in New York State, is the scene p the move to Love Canal. v of the largest and most notorious of America's huardous waste Theve is a striking difference between J tragedies. It exploded once agala into prominence in mid 1980 with the what the blue-ribbon panel repon says first of a number of centroversial studiesel the, effects on the healthof the ;ocal about data and what it actually does with g cornmunity of the toxic wastes duniped there (New Scientist, vol 86, p 298). g ,,', than th act s e know t e co d*

• EgIIere a leading American biostatistician airs his views en the latest report. cerning environment-host inter actions, g f

Was there a serious public health hazard contamination would be expected here even if those facts constitute an incom-were about twice at Love Canal or was it merely much the three areas of probable lower con- plete body of knowledge and if they as many pregnancies in ado about nothing? The answer depends tamination as in the two areas of higher reveal the limitations of the sciena of 0 on e.hether you believe the opin!ons of a this field at the present time " ilowever, blue-ribbon panel convened by Governor contamination The contamination evalu-ation was made before the health survey the blue-ribbon report does not mention ft/

Carey of New York State or look at the was completed. any of the above facts (it offers only f actual evidence that this panel was cup- (2) in the two areas of probably high . opinions of the type already quoted). It 9

posed to consider.

The f acts come from an epidemiologi- contamination, there were 158 pregnan- might be noted that the IIcalth Depart-cies and 37 miscarriages. In the other ment report also does not present the cal survey of the Love Canal area begun three areas there were 318 pregnancies above facts directly, and it is necessary

( y J

in June 1978 and focused on excess mis. and 35 miscarriages. Tne risk' of mis- to do some digging to get to the key carriages, birth defects, and low birth carriage was more than twice as high in f acts in that report.

weights. ne panel did not question the the women in areas of high contamina- Finally, do the above f acts demon- f f acts themselves. Instead, the 10 October, g 1980. report of this panel says: "De tion as in the controls at Love Canal at Lovethat strate the chemical contamination Canal has probably produced a 4 from areas of lower contamination, results of the latter studies were and re-main inconclusive, owing in part to the miscarriage (3) The 113 per cent increase in the serious public health hamrd? This ques. #

relatively small population available for cant. (The data rate is statistically signife- tion involves assessment of scientific evi-  ;

have 95 per cent con- dence and there are two different a p.

study and the absence of a comparable, 6dence limits from 17 per cent to 319 proaches to this task. On one hand, one '

matched population of controls with can make a rigorous and objective evalu-which to compare the figures. The in. per cent.) This large increase is also sig-nificant from the stand-point of public ation of the qualig of the data and their o

vestigators (Dr Nicholas Vianna and implications for public health. An alter-others at the Department of Health) health and indicates a serious hazard.

(4) In the higher contam' nation areas native approach is to make subjective thought there might be some increase in there wyre 122 hve births and 14 docu- judgements on the basis of " expert

• miscarriages and inf ants with low birth opinion *' From the internal evidence of weight, but the data cannot be faken as mented birth deferis. In the control the bioe-ribbon panel report it is clear arema there were 280 more than suggestive." ne following are 15 birth defects. Again, the risk of birth that the panel exclusively relied on the live births and the facts:

(1) Using maps and aerial photo. def,ects was more than twice as high in latter approach.

What is needed for objective evalua-the areas at Love Canal with probable graphs, Vianna and coDeegues divided high contamination as in those areas with tion? One requirement is a careful re- '

the Love Canal area into Sve sub areas probable low contamination. ,, view of the methodology used in the de- "

where, on the basis of proximity and (5) The 114 per cent increase in birth sign, conduct and analysis of the Health drainage. different levels of chemical defects considered by itself is not quite Department's survey at Love Canal. By .

statistically signincant at the 95 per cent curr'ent " state of the art" standards in Fg QE.d level However, under the null hypothesis epidemiological 6 eld studies the method-

.=^

*g J=v N 4 of no effect from chemical contamination, ology is acceptable and the quality con-

,, ' , /* i T !,,Q the miscarriages and birth defects would trol ou the data exceeds usual >

'* 3 . 'y % e S1 be statistically independent. De chance eY

. JIM % One lona 8 wait i

that both would be markedly elevated is i A b 3 tQ r,k h

• f? jy

./** I[, . 4 extremely small, much less than I per

' 7"^ " E '" "'d c"'"' -

~

QL*/, k I

,",y{& f cent, if, however, the contamination had produced genetic damage, then under H b'oo'm^ing, floth Pan Am and British y *%jy this non-null hypothesis one would .ex- Altways have now been granted govern.

• 'v' , j, i pect both types of reproductive wastage ment approval to provide the first

Qn- q regular passenger air service across the .

j ' \,\ A g

y j

to be elevated.

(6) An informal statistical argument on the abeve point can be based on the Pacific for 30 years. But once inside China visitors must still fly on aircraft N operated by the Chinese Civil Aviation bv ', . . .i . (i#',Q[ ' '

i

! fact that there are 10 ways to ditide the Ave sut> areas into a two area combina-Administration which has a monopoly on internal air travel A group of

- . . v8 tion and a three-area combination. There Japanese tourists recently told a horror p, is therefore one chance in 10 that the story of CCAA bureaucracy. Their flight, r { k 7 highest differential in miscarriages would occur in the rombinations suggested by from Chengdu to Tianjin tria Peking was

/ Tgg the maps. There is the same chance th.it much f aster than scheduled because fog at Peking prevented the airliner from

.E] ,n C

. N this would also be true for birth defects.

nos the probabibty o. both events oc. maldng its reutine halfway stop. But Chinese bureaucracy works is s&ange j

l$;l  ;"'n f.*YT)" / curring is 0 I X O l =041- <me chance ways. The CCAA rule book says that j in 500. passengers bound for 'nanjin have te i ,_

(7) Methodological details in the Vianna report establish the validity of fly via Peking. So once the fog had ,

' e the controls. De women in the higher cleared the Tianjin-bound passengers had to fly back to Peking and then return and in the lower contamination areas are to Tianjia before they were allowed at c T*N .

simnar with respe<2 to age, race, and O

( f,. i ~e */Mj -- m~, M-. .j other factors. 'Ihe women in the higher abe plane.

- . wml

~ --- - - - _ ~ . . _ _

l

l. _ _ __ _ _ _ _

l I

i i

\

i i

l A FALSE N.a' ON RADIATION HAZARDS .

AT A NUCLEAR SHIPYARD i T

On 31 January 1961, The Lancet published a report from an agency of the United states goverrament (Center for' Disease Control /

National Institutes for Occupational Safety and Health, CDC/NIOSH) on

" Cancer Hortality at a Navy Nuclear Shipyard" (1). . In reporting on the results of a follow-up study that cost over $1 million and took almost three years, CDC/NIOSH reached the following conclusion concerning hazards to nuclear sta=arine workers at the Portsmouth Naval Shipyard (PNS) in Kittery, Maine

" Finally, in PNS radiation workers, we found no positive dosage response relationships between ionizing radiation dose and mortality from any cause reported."

At a belated meeting on March 17, 1981 of the scientific i

advisory casanittee on the CDC/NIOSH PNS study in Cincinnati, Ohio, the ,

i reason for the negative findings reported in The_ Lancet _came to light:

At the time when the report was submitted, CDC/NIOSH had not carried out 7 a single statistical analysis of the dosage-response relationship for any cause of death, s In response to a January 26, 1981 memo from me which pointed out a clear and strong relationship between the reported lifetime radiation doses and the cecurrence of excess lung cancer in the radiation-exposed workers and a letter from Dr. Tha

• re Colton with a statistical analysis of the dose-response relationship for lei *===4 m and related

- ~

s- c -

r . --- , . -- - - - - - - - - - - - . . _ _ - - . - - - - - _ . - _ -

diseases (Table VI,1) , CDC/NIOSH finally did some statistical analysis of the PNS data for the March 17, 1981 meeting. These were reported at this open public meeting and the proceedings were taped by CDC/NIOSH so that all statements in this article can be verified by documents or tape.

What CDC/NIOSH found after doing their own analyses of their own data were positive relationships between ionizing radiation dose and mortality from a number of different causes of death, a flat contra-diction of the false report in The Lancet for 31 January 1981. The positive relationships range from weak and iffy to strong and clear.

As always, there could be different " interpretations" of the results but this in no way can justify a false report of the actual results.

Scientists or other readers who find it incredible that an agency of the U.S. government would put a false report.in the scientific literature must consider the political context of this and other studies of the health hazards of low-level ionizing radiation. The 31 January article notes that "This study was done at the request of the Subcom-mittee on Health and the Environment of the United States House of Representatives"... "to determine whether the previously reported excess l

l risk of leukaemia and of other cancer mortality among nuclear workers 1

could be collaborated and to evaluate dose-response relationships between radiatica and cancer." The "previously reported excess risk" refers to the testimony of Dr. Thomas Najarian at the 1978 hearing and his report (with Dr. Colton) in The Lancet in May 1978 (2) .

l At the 1978 hearings Congress heard extensive testimony on how federal agencias, particularly the Department of Energy and the National i

l

. . _ _ _ e . . _ _ _ _ _ . _ _ .._. _ .

cancer Institute, had suppressed t.he studies of low-level radiation hazards that found positive resvits (3) and named Dr. Karl E. Morgan, Dr. Thomas Mancuro, and myself to an oversight committee to keep this from happening at PNS. The Atomic Industrial Forum, the lobby for the nuclear industry, insisted on three panel members to represent the industry position, and eventually there were nine members. At a meeting of this committee in August 1980, it was agreed that tha follow-up of the PNS workers was adequate fcr purposec of analysis. However, as the only professional statistician on the oversight comunittee, I was concerned that CDC/NIOSH did not present any plan for the statistical analysis.

In September 1980, the comunittee received a set of data tables for what was now called a " final report". There was no analysis and no text. For dosage-response analyses the critical tables were wnat CDC/NIOSH called " dosage x latency" tables. For every combi 14 tion over a set of dosage categories and " latency" categories, these data tables give the observed deaths and the expected deaths (on the basis of U.S.

white male vital statistics) for each cause of death. Apart from leukaemia and a few related causes (Table VI,1), these critical tables were missing from the final report.

l l

I repeatedly requested these tables and they were repeatedly promised to me. CDC/NIOSH assured me that the only reason they had not been sent out to the advisory ceamittee was because there were too stay and there was nothing of interest in them. My set of tables finally arrived in late December 1980, weeks after the final report was released i

I in eOly becember 1980. '1he rest of the panel did not see these hMa=

until March 17, 1981. >

i r

.e opeuuo me

i. ~-. . _ . _ _ _ _ . . _ _ _ _ _ _ _ .. _ _ _ _ .. .. _ . _ ._ _._ , _ _.._....._.. _ . _ . _ _ _ _ _ .. _ ._._

I l

l 1

Y  ;

In October, a draft of the final report was mailed out. It  ;

i had the original tables, no statistical dosage-response analysis, a l

brief text, and a completely negative conclusion. Despite my strong I l

objections that this was an incompetent report and not even an adequate l preliminary analysis (and despite other objections raised by Dr. Morgan, Dr. Cobb, and other committee members), CDC/NIOSH released its final report to the media in early December and suhaitted an article to The Lancet without even sending a copy to the committee. Thus, the con- 4 clusion of the final report had been predetermined by CDC/NIOSH long before any statistical analysis of the data was :nade, a practice not uncommon in U.S. science and health agencies when official policies are involved (4).

When I finally got the data on the missing causes of death, I naturally turned first to lung cancer. This is not only known to be radiogenic, but it also had the largest number of cancer deaths. Since it is well known that there is a long latent period for solid cancers, I looked at the workers who had 15 or more years " latency" and then at the simplest dosage categorization. At doses less than 0.5 ran for lifetime exposure it would be difficult to detect any radiation effecta, but for dosages over 1 rem thare would be some chance of detecting such effects.

The expected numbers were similar in the higher dose (6.56) and lower

, dose (7.40) series, so the observed deaths can be directly compared.

Since CDC/NICSH had said there was nothing in the tables, I was a bit surprised to find 13 lung cancer deaths at the higher dose and only 3 at

(

the lower dose- a difference which would be hard to miss even with the

" eyeball" method CDC/NIOSH had use- to " analyse" its data (as acknowledged t

-~ - . . . - - . . . , , , , . . . . . . --

l f

on tape). There was a clear dosage-response curve for lung cancer.

Using either a. sign test or Chi-square or non-parametric (rank) tests or 5

parametric regression tests, ' the dose-response relationship .was significant i or highly significant by standard methods. I alerted the comunittee and ,

CDC/NIOSH to the lung cancer relationship in a January 26 meno. {

on March 4, 1981, Assistant Surgeon General Anthony Robbins,

.the head of NIOSH, sent me a letter defending the CDC/NIOSH statements t

in _The_ Lancet and enclosing a written reply by his staff to my January  ;

26 meno. While conceding that the lung cancer data table in the mano  !

was "a legitimate use of the latency-by-dose chart and is the sort of i i

thing our epidemiologists might choose to do",.and that "the penhahi14ty of the observed excess h= = =45 by chance alone is less than 54*, ,

CDC/NIOSH continued to insist that "Dr. Bross's assertions are not supported by this data."  ;

The argument used here'(one often used whenever data fails to j support a predetermined conclusion) was that I had not fully specified 1 lung cancer as an "a priori hypothesis" to test. Yet the 31 January article spec.ifically mentions lung cancer (citing two papers), and a 15 f or more year latent period (and only this. period) would be consistent f with the March 4 statement by Dr. Bobbins that "the report reserves i

judgment"..."because of the small cohort sise and short latency experienced ,

by most of the cohort relative to the cohort sise and lateocy periods necessary to study organ cancers."

To show the danger of doing statistical analyses ot' dese-response relationships instead of "eyah=114ag" the causes of death, the CDC/NIOSE staff ocumented: "Te demonstrate further that patterns ednish i

r

will occur in the data by observation of them in the absence of an a priori hypothesis leave the meaning of their existence questionable, one can examine the category " cirrhosis of liver, ICD 581" and apply the same dosage and latency requirements that he used for lung cancer. For this ICD, a statistically significant increase (pc.05) can be oLverved in thoss with greater than 1 rem and greater than 14 years latency.

Similarly, disenes of the digestive system are significantly elevated."

Indeed, there are several other suggestive or significant causes of i death when an appropriate analysis of the dose-response is made. One of ,

the more important scientific findings of this study may eventually be that there is a broad spectran of disease from low-level radiation rather than excess cancer only.

It might also be noted that the CDC/NIOSH statistical analyses in the March 4 letter and at the March 17 meeting were not only made long after the paper had been submitted with a flat negative conclusion, but were made for the purpose of " refuting" the communications-received from Dr. Colton and myself. The story abounds with ironies. CDC/NIOSH did an analysis of "All malignant neopinama of the lymphatic and hematopoietic tissue" (mislabeled " Repeat of Colton Analysis") and emerged triumphantly with a "very nonsignificant" p-value of 0.9546. However, to nail down the point, they also ran leukaemia alone. This gave p = .0464, thus inadvertently comf4reiog the 1978 Ma$arian-Colton report. CDC/NIOS3 disposed of this awkward result by dropping the case of leukemia in the highest dosage category to get the p-value up to 0.64. After all, as reported in the January 5,1981 Deston Globe, "the NIOEM team said their findings specifically refute the 1978 study.*

ee=ennae a q ,-. ,-m- .,-,v,-m- - , * - w,s. ,. . . - -- - , - , - . - - =

. - - - - - . ~ - - . .-. ~ .. - - - . . - . -- . .. .. - . ..

I l

l Readers who might wish to draw some amral~from the sad-but-true story of how U.S. government agencies do science have a choice of several. The first is: In normal scionee'"A theory must fit the facts" whereas in U.S. federal science "The facts uust fit the policy." ]

Another moral is: There is no such thing as a " safe" position.

Perhaps to mollify me, NIOSH noted in The Lancet that there was still L the possibility that at some future time positive relationships might start appearing (because of longer " latency") . Unfortunately this did not please M=4ral H.G. Rickover, the father of the nuclear' navy who had demanded total " vindication" of PNS from the start. His tirade against i NIOSH to a congressismal committee hit where it hurt. .

Ironically enough, the navy also turned the negative statement ,

against NIOSH. CDC/MIOSH had been hoping to follow the epidemiological study with a large-scale cytogenetic study at the shipyard. On February 23, 1981, vice Admiral E.B. Fowler refused navy section of this proposal  ;

with:

"since this conventional retrospective epidemiology study has demonstrated with reasonable statistical confidence that there are no discernable health effects of radiation exposure in the Portsmouth Naval shipyard egioyees, it is concluded that additional studies such as the -

proposed NIOSH cytogenetic study are not necessary nor justified in the light of available scientific information.'"

Finally, shortly after vriting the cited letter, Assistant surgeon c==aral Anthony Robbins was susmarily fired by the Reagan adminis-tratism. (5) .

1 1

1 ut Ly:ptuLw anu Menetopoletic Tissue Jtepeat of Colton analysis l AWACEENT C

. p

.' Plot of rate by.015a.0001371 dose.

midpoint, rete .065,0; .30a.000157s .75,.000097s *

a i

3,.0003511 10,.000004 N.9546 .

L RATE ^

0.00036 +

a g@ /

s

~~

0.00033 + }

a - .

A

^ - -

O.00030 +

A s e  !

A A - '

O.00027 +  !

• ,1 L

+ -

A '

O.00024 +

s

• l A

A 0.00021 +

^

s .

O.00018 + .

^ .

~

^

Rh -

0.00015 + -

. ^a0 A

0.00012 + s..

A

• A *

.O.00009-+

~

Q _

^ '

^ .

^

i 0.00006 + - >

.a C.00003 +

.+ .. .

'A 6.06000 +R@ ,

.- --+- ..+_.+___+%  :=___ +_._+__ . _ _ + _ . , . - -

1 ~E 3 4 5 6 7 S 9. 10 11 12' 13

/

DDSE

2eukemia Plot-of rate by dose 'All 1 cat.

.30o.0000411 .75 1.0000971 midpoint, rate .015,.0000651 .065o01 P=.0464 3o.00016i 10,01 20,.000to

' ' ' ~ ~ '~~~ '

RATE ^

R-1.40040 +^ .

i

^ .

,- 5 A -

- - .. .f

).00035 +A ,

8

.I A - v 4 r A .

^ .

O. 00030 + < . ,

^ ' , .

^

A . .

s.. .

.A .

0.00625 +^ -

A

. + .

-" --- .1,

e. . - .

O.00020 +s ., .. -

-' s i.

. t.

e, . ,

^ A- -

0.00015 +. - -

A a

+ 1

[

I 6

A g e e }i g' .-

0. 00010 +a A - .-

^A A 9 4.00005 +-

e k

A

^

... - s

1 C.00000 +A ) A

- e_ .e. - - - ,.- - + - - - - - - + - - - -+- - +-----+---~~+-- .

14 16 18 20 22

\ 9 2 4 6- 8 10 12 e i

' I DOSE 3 ..

. o - - - -, -e- -c-- - = . , .- 4y +w-y v y g--yv -, i-.

Leukemis .

Plot of rate by dose Drop cat. 7 ll

. i aldpointo rate .015,.000065; .065,01 .30,.000041; .75,.000097; \ 4

~

l' 3a.00016i 10,0 P= 64 t

RATE ^

O.00016 +

A

-h '

A 0.00015 +

^

. 4 O.00014 +

^ .

A 0.00013 + ,-

A

^ .

O.00012 + -

A A .

0.00011 + -

A .

0.00010 + -

^ . -

A 0.00009 + .-

A .

O.00008 +

A A

O.00007 + -

A ,

^R @

0.00006 +

A

^

O.00005 +

A .

0.00004 + R ,

A .

^ -

0.00003+} . .

A A

0.00002 +

A .

A 0.00001 +

A -

e 0.00000 R I 0 1 2 3 4 5 6 7 8 9 10 11 12 13 i DOSE .

>. of rale by uuse cu.._ _o sidpointorate 015o.0000651 . 065aos .30,.0000411 . 7So.0vu.17s '

3a.00016i 10a.0000a4 P=.5179 i i

, , , , , , . .. _ _. - - 4 RATE ^ '

0.00016 + Ah ,

i i

g a

• l 0.00015 + ,

^ .

A' e

.).-

O.00014 + .

^ . .

.A 0.00013 + *

^ . .

A 0.00012 + .

A A

0.00011 +

A 0.00010 +

A A

A9 -

-0.00009 + * .

A -

^

0.00008 +

gg A .

A O.00007 + .

A

^A h 0.00006 +

A '

A 0.00005 + "

A A

O.00004 +A A'v9 -

A

• I 0.00003 + _ -

^

A 0.00002 + -

A * '

A 0.00001 + .'.,

a .

4 &:

e G.00000 +A :) -

_+__- w___+____+____+-___+___ + ___+ ___+____+____+____ w ___+____,____

8 '1 2 3 4 5 6 7 8 9 10 11 12.- 13

\ . .

DOSE

,q - . , , -.-. -

9 y m----- w

e mn u a soss.Pn u.

Derecto, of Boostateca Roeweel Park Memorial institute 946 Sm Sweet sunato. N.Y.14203 n e w e.n.o .a.c.c a. see.._ _-e flees.A Pesa essansses emeteen. es of me. e T St.se res.9 Os.esuused March 12, 1981 President Ronald Reagan White House 1600 Pennsylvania Avenue Washington, D.C. 20500 Dear Mr. Presidents Although you campaigned against regulating co.porations to death, you are letting the Nuclear Regulatory Commission (NRC) do just this to the utility at Three Mile, Island Unit 2 (TMI-2).- What is far worse, NRC overregulation will cause the death 'of many human beings by insistence on an unnecessarily costly and dangerous " clean-up" plan.

While NRC claims that the " clean-up" will not even cause one death, this estimate is based on out-of-date hayh-dose data and faulty extrapolation. If instead the latest scientific data on the nuclear submarine workers at the Portsmouth Naval Shipyard (PNS) is used (reported March 17,1981) the picture is entirely different. There will be around 100 lung cancer deaths in the " clean-up" workers.

Round numbers will be used to keep the calculations simple and the common-sense approach clear. Assuming a 15-year latent period for lung cancer, there are 16 lung cancer deaths among PNS workers with lifetime exposures of 0.5 rem or more. Calculations from U.S. vital statistics by CDC/NIOSH show that only 8.23 deaths would be expected in this series. The approximately 8 excess deaths occurred in roughly 5000 person years. This is about equivalent to 1000 nuclear workers exposed during 1959-l%2 getting lung cancer in the 5 years 1973-1977. Since the doubled risk of these workers would continue for at least 10 more years, there would be at least 3 x 8 = 24 cancer deaths in this series.

Since probably 4000 " clean-up" workers at TMI-2 would be exposed to radiation doses as high or higher than those of the 1000 PNS workers (whose lifetime exposures averaged only about 5 rem), at least 4 x 24 = 96 lung cancers would result from the " clean-up". Note that NRC permits 5 rem annually. In addition to these deaths, there would be radiation-induced deaths from other causes among the " clean-up" workers as well as a substantial number of deaths in the general population from environmental contamination when the radiation is removed from the containment. All of these deaths and disabilities will be caused by MRC regulacions because there is a much better way to decommission TMI-2.

utrcn 12, 1961 Page 2 The method is called " entombment" and consists of immobilizing the radioactivity in THI-2 in concrete ano leaving it in the containment.

This would keep the radiation out of the environment. It would also keep the workers out of the containment at TMI-2 since most of the job j

can be done by remote-controlled equipment. The israobilized radiation isn't likely to hurt anyone until the whole installation crumbles to dust in a few thousand years. The costs of entombment would be about one-tenth of " clean-up" costs.

Unfortunately NRC is deterni.2ed to decommission TMI-2 by the rule book even though this is going to cause the very public health disaster that was so narrowly averted in the original TMI-2 accident.

Ver' mcerely yo s, WBross, Ph.D.

I 1 .J.

1 ctor of Biostatistics IIX7B/mak CC: Open

's

{

s

- * *

• we 4.-+w .m ,, __

arwin U s oross. Fri u.

Dwector of Bioetstestice Roswell Park Memorial Institute 666 Dm St,oet Buttato, N.Y.14263 e aa. muw h=i.a. ora w a..ee.u- -er A u I p.e. e at.r e en.m.t.ee af th i$ T se.g.M Mh 9.ugs e4 March 12, 1981 President Ronald Reagan White House 1600 Pennsylvania Avenue Washington, D.C. 20500

Dear Mr. President:

Although you campaigned against regulating corporations to dea:h, you are letting the Nuclear Regulatory Commission (NRC) do just thi.s to the utility at Three Mile, Island Unit 2 (TMI-2). What is far woroe, NRC overregulation will cause the death'of many human beings by insistence on an unnecessarily costly and dangerous " clean-up" plan.

While NRC claims that the " clean-up" will not even cause one death, this estimate is based on out-of-date high-dose data and faulty extrapolation. If instead the latest scientific data on the nuclear submacine workers at the Portsmouth Naval Shipyard (PNS) is used (reported March 17,1981) the picture is entirely different. There will be around 100 lung cancer deaths in the " clean-up" workers.

Round numbers will be used to keep the calculations sisiple and the common-sense approach clear. Assuming a 15-year latent period for lung cancer, there are 16 lung cancer deaths among PNS workers with lifetime exposures of 0.5 rem or more. Calculations from U.S. vital statistics by CDC/NIOSH show that only 8.23 deaths would be expected in this series. The approximately 8 excess deaths occurred in roughly 5000 person years. This is about equivalent to 1000 nuclear workers exposed during 1959-1962 getting lung cancer in the 5 years 1973-1977. Since the doubled risk of these workers would continue for at least 10 more years, there would be at least 3 x 8 = 24 cancer deaths in this series.

Since probably 4000 " clean-up" workers at TMI-2 would be exposed to radiation doses as high or higher than those of the 1000 PNS workers (whose lifetime exposures averaged only about 5 res), at least 4 x 24 = 96 lung cancers would result fross the " clean-up". Note that NRC permits 5 ram annually. In addition to these deaths, there would be radiation-induced deaths from other causes among the clean-up" workers as well as a substantial n=har of deaths in the general population  ;

from environmental contamination when the radiation is removed from the containment. All of these deaths and disabilities will be caused har NRC regulations because there is a much better way to decr===.4 aston TMI-2.

• - mL t- --ee- c -m e r e + y- *sy- +

.utui ., 1981 Page 2 The method is called " entombment" and consists of immobilizing the radioactivity in TMI-2 in concrete and leaving it in the containment.

This would keep the radiation out of the environment. It would also keep the workers out of the containment at TMI-2 since most of the job-can be done by remote-controlled equipment. The insnobilized radistion isn't likely to hurt anyone until the whole installation crumbles to dust in a few thousand years. The costs of entombment would be about one-tenth of " clean-up" costs.

Unfortunately NRC is determined to decommission TMI-2 by the rule book even though this is going to cause the very public health disaster that was so narrowly averted in the original TMI-2 accident.

Ver cerely yo s, I 1 . J. Bross, Ph.D.

i ctor of Biostatistics In7B/mak CCs Open

\

i

e. - mmo* m -mm- ee- -m em . o e s 5 eW= = -3 4 '-m'r- p+ ptevwc-+ -' - e w y *'m-'- -i- *- 2ew-,w- - '

l i 1 l

UNITED STATES OF AMERICA NUCIE.AR REGUIATORY COfetISSION BEFORE THE A*IOMIf, SAFETY AND I.ICENSING BOARD e

In the Matter of )

)

HOUSTON LIGHTING & POWER ). Docket No. 50-466 COMPANY ) ,

)- .

(Allens Crack Nuclear ) t Generating Station, tJnit )

No. 1) )

l AFFIDAVIT OF IENIN D. J. BRISS  !

i Caffalo, New York ,

I, Irwin D.J. Bross, of lawful age, being first duly sworn, upon my cath certify that I have reviewed and am thoroughly familiar with the statements contained in the attached affidavit answering Leonard Familton's affidavit of 9 March 1961 and that all statements contained therein are ,

true and correct to the best of my knowledge and belief.

1 M ia 6."J. Bross - f~

subscribedandsworntobeforemethis.2 day of .'I & b 1961. i

/ .m

? d l k (i l$ - / ! Lt vf-8

.)

St*tPA M. DApesRT My h 4maion empires: h (..W5. W

• Si Mw TM Osanteed ie trie Ceems,  ;

W Gamamassen Es, emes w,sh as, 39.f.;.

l n

3 6 e-mi. _w-- . - . - . - . - ---,.-v--r,, . ,. . . . - . _ . - ,. -- .m-. - . - . , . - - .-.y,-- #g

...inu a on.ena o.es .e -

Meoweit Park AdemonalinsIMute ses sem me,est~

tedeces. M.Y. MSEs

,,, _ - -. m. e 1 ammmes,e.n espumpe, ame ,emme se er se . y esas summen eenaame. -)

march 12, 1981 2  !

P /, s s

< n to:~~-

.3  ;

.3 9

President Ronald Reagan -

- 'APR 31981 * ..J -

l White House k- CM ~

' *~

l .}

1600 Pennsylvania Avenue fac -

l Washington, D.C. 20500 g 'g j Dear Mr. Presidents .

O '

Although you campaigned against regulating corporations to l death, you are letting the Nuclear Regulatory Constission (NRC) do jus t - l this' to the utility at Three Mile Island Unit 2 (TMI-2) . - What is far worse, NHC overregulation will cause the death 'of many human beings by insistence on an unnecessarily costly and dangerous " clean-qp" plan.

While NRC claims that the " clean-up" will not even cause one death, this estimate is bcsed on out-of-date high-dose data .and faulty extrapolation. If instead the latest scientific data on the nuclear.  ;

suh= marine workers at the Portsmouth Naval Shipyard (PNS) is used (reported March 17, 1981) the picture is entirely different.- There will be around' >

100 lung cancer deaths.in the " clean-up" workers. i mound nusabers will be used ".:o keep the calculations simple and the common-sense approach clear. Assuming a 15-year latent period for lung cancer, there are 16 lung cancer deaths among PNS workers with l lifetime exposures of 0.5 ren or more. Calculations from U.S. vital ,

statistics by CDC/NIOSH show that only 8.23 deaths would be expected in  ;

this series. The approximately 8 excess deaths occurred in roughly 5000 person years. This is about equivalent to 1000 nuclear workers exposed r during 1959-1962 getting lung cancer in the 5 years 1973-1977. .Since the doubled risk of these workers would continue for at least 10 more i years, there would be at least 3 x 8 - 24. cancer deaths in this series.

Since probably 4000 " clean-up" workers at "MI-2 would be [

exposed to radiation doses as high or higher than those of the 1000 PNS {

workers (whose lifetime exposurtes averaged only about 5 rem), at least 4 x 24 = % lung cancers would result from the ' clean-up". Note that .l' NRC petuits 5 rem annually. la addition to these deaths, there would be radiation-indeced deaths from other causes moong the

• clean-up" workers ,

as'well.as a substantial number of deaths in the general population from envira====tal contamaantion when'the. radiation is removed from the  ;

containment.- All of these esaths and disabilities will be e====d by MRC

{

regulations becauses these is a much better insy to decomunission TMI-2.

D

~_  % .__ . _ _ . - - _j

!$dx c - , 1961 Page 2 The method is called "entomoment" and consists of immobilizing the radioactivity in TdI-2 in concrete and leaving it in the containment.

This would keep the radiation out of the environment. It would also keep the workers out of the containment at TMI-2 since most of the job can be done by remote-controlled equipment. The immobilized radiation isn't likely to hurt anyone until the whole installation crumbles to dust in a few thousand years. The costs of entombment would be about one-tenth of " clean-up" costs.

Unfortunately NRC is determined to decommission TMI-2 by the rule book even though this is going to cause the very public health disaster that was so narrowly averted in the original TM1-2 accident.

Ver* cerely yop s, r

/ J. Bross, Ph.D.

I 17 .

iptorofBiostatistics

. /

IEATB/mak CC: Open 9

l l

t l

l l

l l

l i

i i

~. .. . - .. . . - - - - - . - . . . . . - - - - _ -

+-*-,--ar g m w--e - igi- , -wq.p%-. g y-w , e- -_-em,.,4sn,-a, y-.g.94,p. gy y- e. g -.y-p- g- y .----p--r ---. 3 m y g + - -g -va--