Tobacco Documents Online

-those are references number 10 .y in oj.her publications. d us a cop~' of your paper. ontinue ? ;ed state of knowledge, blaming ote that emph%•sema is largely a sed to the same environment, in- ver develop emph~•sema. Women laim that the cause of emphysema resea rch. I the most common cause of death ularl.• males, the oversreight, the in itiherited lipid diseases. The :ized do not withstand critical -noking have indicated that the :)r constitutional. A certain per- th a striving, time and pressure, isposed to coronary diseases. To= than cigarettes for the causes of rtion to nicotine and `'tar" would taken information. The original :, stated#hat nicotine is probably ict this generally held opinione ziRar6tte condensate (so-called at fresh condensate free of arti- producing quality. ealtli for Americans. and have . In mY considered opinion, the - ied. and if passed. R-ould gener- be Iikely to co-opt and to corrupt .rch effort. -opt - and corrupt our medical cher would be-inteted;tedan thftk_~,:: lv, a-hat-publications in tlu-bst- tmber of them. sir and I would Burch of England. In England nress and sense of fair plar, and .11 which is going on. including ing. xhat is his - R•hat is the thrnst !rences. X THIS DOCUMENT SUBJECT TO 277 CONfIDENTIALITY AGREEMENT. Senator liE~N-mr. Give me the name of the article. lir.,oxxEtca. Oh, vrell,1 cau give you some references, sir. That is aew jcientist, 1974. February 21 issue, and FeLruar% issue. The 1975 is in various issues of Lancet and British Jlrlcal Journal. ,enatorKE.NEar. Give me the name of the article. Dr. SoxxEEns. The name of the arti$je-is "jmoking as a Cause of " Lung Cancer. Senator IExxmr. All right. Now, that is one. ti1'bat are the others t . Dr. SoxxERs. In respect to emphysema, in addition to the articlcs I cited, there are a number of papers by Russell P. Sherwin. They af~- peared in the Pathology Annual, about ly6e, and again - Senator Iit\-xEaz. I am not asking you for lyGe, but the last five yea rs. Dr.SomMER6. Oh. in the pulmonary pathology seminat•, published i>> 1975, there were articles by Dr. Thurlbecl• and llr. Sherwin. reviei% in~ the subject, and with respect to coronarydisease, there are a nunil,er of articles. Senator hE-,-,Eny. Is this on efnphrsema and smoktng' Dr. SOxMERS. Yes, sir. SenatorliEx.,;Enr.The causal relationship between smoking and - Dr. Soxmmas. No, it discusses emphysema as a disease in relation to its development, and the possible causative factor. Senator I:ExrEny. And it has discussed smoking in that. too ? Dr. SommExs. Smoking is included in the discussion. Senator Ii.ExxEOr. And what conclusion did thev reach % Dr. SoxxExs. That since it is so hard to define the disease by pre,ent methods, it ishard to implicate any factorscausinR it. Senator Iitxxi:or. That is the conclusion of the Thurlbeck article ~ Dr. Som xERs. Of that particular article, yes. Also. the article cited in the references here. Senator IiE..Nmr. That is two. What other ones ? Dr. Sox•rERs. On emphysema. sir ? Senator K.c.~zor. 1o; on lungs and cancer, and oancer of the lung. Dr. SoxxERs. I have a pack of reference cards this thick in tny ba~. Senator fiENvEar. \o; you are here as an expert. I am just askit~" the last 5 vears•. Iwoula expect that you could be able to kickoff at least half a dozPii prettv quickly. You are here as an expert witness. and you ought to be abje to,since you have not done any research directl.,.• but are relYin• on other peoples' a•ork. Dr. Soat?fER:. Yes. article by Weir and Mitchell iu Cancer 1976. In respect to carcinoma- Senator KtxxEnr. A'hat is Weir and Mitchell's article on? Dr. SoxuERS. The '"'eir and Mitchell article dealt R-ith.lung cancer. and the difficulties in diagnosis and classification. It drew the conclusion that with the little we know it would be pre- mature to draw an etiologic conclusion. including cigarette smokin1g. Senator hExxEnr. That it-excuse me t Dr. SoxuExs. It would be premature to draw an etiologic conclusion from cigarette smoking.

THIS DOCUMENT SUBJECT TO 278 AMOMMUn AGREEMENT. Senator KxN'.%;ZDY. Would you say those are the Ieading three. or do you want to add a couple more, just so we got the last couple, and we hace your prime references? Dr. Soxxm. R'ell,sir, there are a4arge number. Senator liEx,%*soY..No; just the top.couple, that come to mind, within the last 5 years; studies that you are relying on. Dr. Soxxetts. In r+elation to the cigarette as a- Senator Kt:xxto:. E:actly. That is right. Dr. SoYxtxs. There has been cer~ recent interest in this. In term~ of the pathological lit.erature, and the epidemological literature, is re• viewed and summarized bv Professor Burch. Senator ht.rrtnr. S1'ef1, that is the problem, you see. it seems that you are referring to the old studies. We asked You what the new studies that support your position- you seem to have some difficulty in finding them, or naming them. Dr. Somutrts. There are a number, sir. Senator liExN-mr. Give me the. last two or three that you think are the leading ones, in the leading medical journals in the country. that support your position. Dr. Soxut:aF. In my position, there is not enough that is known about lung cancer. . Senator Iaxxzny. Well, that supports your position thit it has not been proved about the relationship between smoking and lung cancer. Let me state that very definitirely. very preciselr. You say ~ou ha~~c not done the research vourself in this area. You are relcinF on other testimony, and other publications. T o.r, name them. Dr. Soxxtxs. Sir, the basis of my understanding is readin: in the field for 40 t•ears, and investigations in the .field. and most diffi-utt thinq to prove in science is a negative. But the fact that no article= producing conclusire proof of a positive. causative efiect year aftrr year. indtcates it perhaps does not even exist. So.•ou are asking me to prove the negative. Senator Hatrr. Will the chairman yield ? Senator KEx%aaT. Yes. Senator Haxr. I think the chairman is asking the same qnection I am. The first sentence in .our prepared remarks sa' vs You are goinL, to review the newer scientific data. not analrze the Iack or absence. You state vou.riil discum newer scientific data. Now. I look over the references in your statement. there are 3 documents out of 13 that are dated 1975. .%1r potition, frnnklr. and I think to a certain defirPr. the chnirman'; is based on the 197s HEW reports. Senator '.Nlorgan said there was morf infnrmation than that. We want to be procided.rith that newer information. We are not talking about prorint?the negntire. We nrc talkina;thnut thic new scientific data that you refer to, and some pro~~lc thic mnrnine referred to in their testimonv. R'e have three article" herethat have 1975 data on it. N'on-, we are at a loss here.because there are no footnotes. You make Fomp cdternrical asvrtinnc ahnut the state of the infnrma• tion nntl the e.•idence. and list 1R articles. We do not knnn' whirh ar• ticleF--Rhich one refers to which statement. That would be helpful. V, CTR H~ ~~~~~~

e leading three, or do e last couple, and NTe come to mind, within rest in this. In terms .gical literature, is re- you see, it seems that ,port your position- or naming them. •ee that you think are : in the country, that nough that is known asition that it has not KinR and lung cancer. el}-. You sar vou have , are relyi.ng on other jing is reading in the Id. and most difficult--- ! fact that no articles itive effect year after he same qnestion I am. sars ~•ou are ~,+oinr to e lack or absence. You t tatement. there are 3 degree. the chairman's ` -an said there was more- ormation. ~. We nre talkinr about -ne people thir morning a on it. . )o footnotes. ne state of tlie informa- dn not knnR which ar- I'hat would he helpful. THIS DOCUMENT SUBJECT TO 279 CONFIDENTIAUTY AGREEMENT. if you would provide them. Three are dated 195;,. and then you go to the 1960's. 1 am talking about the HE1t' report. '•Health Con:rqitense~ of Smoking." %chere on page 15 it sav~--- Coronarj heart disease is the major cause of death in-tbe L'nited StatrS. alad la the most Important single cause of e:ceai morulltp among cigarette en,oler- I am referring to paRe 65, which saV-- Ctprette amoking ia the most Important cause of chronic obstructice pulmonary dlsrase, emphfaema, and broncIIlUs. I think it is, in a sense, with regard to this report, that we %cant the newer scientific data whieh you have promised. Dr. SoxxExs. In addition t,o.vhat.ras already cited there are three papers b* v the Leuchtenbergers, orho work in tiwitzerland. they point out that if vou use condensates, so-called tar, in their model system. it has no relationship to the'development of abnormal changes. whicln they regard as possibly cancerous, and if there is any part of the ciga,- rette that might be held responsible. it is not the condensate. Those are the three papers in 1974. Senator H..Rr. But those are neRer than the HE11' report ? Dr. SouxERs. Yes. Senator HARr. June 1975 ! Dr. SoYUEx.c. Yes. If `ou will look at the reference in the HEAV report, in certain se.ctions. about a third of them were within •% Venrs of the report, in others, about half of them within 2vcars of the report. So the references used in that report also go back quite n nttnlber of yea rs. Senator HaRT. But the Assistant Secretary for Health sipied tlti~ in Tune 1975. and I presume he agreed with the conclusions then. He would have had the benefit of the Leuchtenbergers. or N% hoever. Dr. SowitERc. No. sir. you do not realize there is a little ln involved between publication and receiPt of a journal in the United States. that in abstracting, and the use of it b~• the Government. Senator IiExxEny. In fairness. just-your point just does not hold up. The~• have list after list of pages of recent studies, and I.~ill just read tl,e year. Starting-I mean just at random, on page G•1. this is r,ue of them. December 19i3. September 1972-going down. December again. 1972. American Nfedical Arsociation,\o.•emberof lA;?.1Iost of these ref- erence-~--\farch of 1973. Jfarch of 1973. INfai v of 1A73..Tust taking, otie nace. And this is 1974 report. There is not one that is before 1Q7 1. Dr. SOxaERc.1 es. sir, but I referred to the 1975 report. Cenator I:ExxT.nr. "'ho does Dr. Burch work for ? Dr. Snu.rr:Rs. He is at the t'niversity of Leeclr. Senator fiExxEny. Is he a consultant over there f(ir nm organiza- tion4 Dr. SoatvERE. He statea in one of hif- articles. to answer ohwinu~ questions, he is, \o. 1. not a smoker: and No. 3. not in the pov nf the tobacco industn-. Senator KEN-xmT. Ofi. f The prepared statement of Dr. Sommers and supplemental letter follow:) CTR HN 014"791

THIS DOCUMENT SUBJECT TO DONFIDENTIAUTY AGREEMENT. 1(y purpose, in response to the inv4ation to pre'sent ap• perscn a: vie, of the status of knowledge of smoking and health, is to revirw ::._ nwer scientific data. This material is not included in the q~ctat:::.= frow thr 1973 vers16n of the USPHS Saroking and Hea;th rercr: c:te: Senator Hart as justifying Bill flo. S 2902. CoPni:ar.ce of the rre:e-- day knowledge by fairminded individuals casts grave douC; or. :'.e existence of any such 'overwhelming evidence' as -:ained by tne I am Sheldon C. Soeters, k.D., Direetor of Labora:ories, Lencx r::: Hospital, New York; Clinical Professor of Pathology, Coluabia U:.iv. College of Physicians & Surgeonq, New York, aad Univ. of Sou:ne -. California, Los Arigeles; editor of the Pathology Arvwa: and Pa:.".o:ag. Decennial series; Chairman of the Scientific Advisory Bcard, Co zr.c'_: for Tobacco Research, U.S.A., Inc; with about 260 pub a=atior.s. curriculum vitae and publication list are available. First, in respect to the reported overall higher dea:- rates c'' smokers than no:rsaokers, the results were obtained by statist:=a: epidemiologic methods now recognized as erroneous ar.j use of is no longer regarded as Justified. One to one comparisor.s of s:.:te:s and non-amokers ignored mary confounding variables. As berxs:-. :_ dicted the result ;roved to be a statistical artefact. :...r res:::e_ f: x+ igr.oring three serious sources of error: (1) Nonrandoa comparisons, sx aning that smokers and non-scokers a,e CTR (~~~ 0 14~t='~~

. .tion to present rky personal view ; and health, is to review the is not included in the quotations ioki n: and Healt` re'.ort cited by i290,-• Cognizar.ce of the present ,:s casts grave doubt on the :vidence' as ::aimed by the dis- :or of i.abora:ories, Lenox Hil: Dr of Pa:hology, Columbia Univ. e York, aad Univ. of souther.^n he Patholopr Annual and Pathology. ientific Advisory Boarty Council ith afu: 260 publications.' : are available. era a higher deatt rates of : were obtained by statistica: ~. as erroneous and use of which is le to one comparisor.s of smokers iding variables. As•BerRson pre- . I:istical artefact..;.'t;ir resuited of error: :hat smokers and non-saOkers are -2- ;~: .°?f ' ` . 281 THIS DOCUMENT SUBJECT TO CONfIDENTIALITY AGREEMENT. not randcro subsetS of the population being testFd. Smokers ch=se to smoke and others choose not to fmoke, and both groups are se:f- selected, apparent:y by complex processes not presently unders:oo:. The crucial error is the use of statistical methods to test nc-.- random groups, desigried for and applicable only to comparinb groups. Ro11inQ the loaded dice statistically gives arn ir.va-:' answer. (2) Dependence on limited statistical ttsts for proof. INher. Rose a.d Bell wrote their monograph on predicting longevity, they tes:e: factors that might be associated with early death among a gro,:; cr Boston MW II veterans, followed careiLlly, for 30 years. A one-to- one comparison of cigarette smokers and non-smokers gave smok:ri the nuAber one place as a predictor, like in mary other studies. However, when multiple factors were included statistically, cigare-:e smoking dropped to somewhere beyond the 30th most significar.: Yre- dictor of early death. Dissatisfaction with job becaWe the nu.:`-er one predictor. (3) Secularity, meaning the change in habit patterns of the poplla:::-. from about 1910 to 1960, with increased proportions of cigare::e smokers in each decade. A model population (not a real popula:`_on; with such changes over equal time will show a statistically signif- icant increase in total death rate as a minority changes torart a majority. This secularity factor or change in population pa::e:-.s has no health implication. For example, in the original Surge=- General's Report a table sho.ed pipe smokers to have a lower death rate (0.8) than non-smokers. While encouraging a belief that pipe 10.011 0 - 76 • is Y:.+.--.A-Xm0s CTR HN 014,179-.~

282 THIS DOCUMENT SUBJECT TO -3- CONFIDENTIALITY AGREEMENT. smokers were the 'sturdiest humans,  better e:planation would be the secularity affect. There were fe.er pipe smokers in ea:h decade, the converse of ciprette smokers. These data are published and available for all to read. Why have the proposers of the,bill not heard of them? It is because the annual supplements to the Surgeon General's Report on Smokir4; and Health are linited reviews and interpretations of the medical and scientific literature, from which are screened out articles that do not agree with, call into question, or destroy arguments and conclusions supportive of the official government position. Aa an aYample, in a presentation prepared for hearings on a proposal by Sen. lfoss (Utah) " liait by law the smounts o!' nicotine and con- densate (so-called tar) in cigarettes, I pointed out in 1972 that ap- proximately 1790 articles published since 1960 were not cited. Lists of these •ere supplied. On occasion, publications not supportinQ the official position have been suppressed. The epidemioloQic problems, shortcomings and errors fere aired at a panel sectinY of the AAAS in 1971, so the above is not secret information. One regrets that the proposers of the bill have not been sude aware of the serious defects, obsolescence, bias and incompleteness of the information on which they appear to base the current proposed bill. liow as to lung cancer, there is a statistical association between cij- arette asoking and lung cancer. But at present the nature of the CTR HN 01417c.-14

a better explanation would be feser pipe smokers In each smokers. for all to read. Why have the s' It is because the annual eport on Smoking and Health are the medical and scientific t articles that do not agree with, ts ard conclusions supportive red for bearings on a proposal e amounts of nicotine and oW I pointed out In 7972 that ap- nce 1960 .ert tat cited. Lists .nQ the official position have •oblems, shortcomings and errors 1 tAA.S in 1971, so the above is not the proposers of the bill have 'eets, obsolescence, bias and"- rhich they appear to.base the tistical association between cig- at present the nature of the 283 -<- THIS DOCUMENT SUBJECT TO CONFIDENTIALITY AGREEMENT. association or whether it is causal are not kno.n. The text of ae original Surgeon GeneYal's report deal# with the difficulties cf assigning causality, but the utmrry and conclusions brush these aside, and assign a causality not demonstrably evident in the text. It is widely known that a stitistical association is not by Itself proof of causation. A statistical association may point to ezperixr,:s that will help to determine whether there is cause involved. Animal experiments to apr knowledge have not succeeded In the production of so.called human type lung cancers (pathologically termed sq_s-a~s cell earcinoaxs) in a sianificant percentage of any species tested. This includes inbred &lce erposed to heavy cigarette saoke fnhalaa cr. over practically their entire life•span. Certain false alarms like the Auerbach beagle dogs, and other studies of rats and itiice, have stirred hope that a model had been achieved, but no cancers that grew, spread and led to death like human squamous cell lung carcinocas have been reported. After 45 years of inhalation research, and a::hocg= expensive efforts continue, no success has been achieved ir, ;rod ci^g experimental lung ca.ncers in any reasonable or even small n.unbers of experimental anima:s. Skin painting is cited in the bill proposal as part of the evidence c' carcinogenicity of tobacco smoke condensates. Suffice to say that in the past 5 years, skin painting of anisnls with tobacco condensate has yielded so fe+ tumors, either benign or an lignant, that the practice has practically been given up. Only a few governmental agency erperixe-.:s continue. Why tumors were produced ye:rs ago following skin pai^tir.g CTR ~~~~ ~~~r~~~ ~

TNIS DOCUMENT SUBJECT TO ~ QONFIDENTIALJTI( AGREEMENT. . ~ but are now rarely produced is unkno.m. M English book reviev of a work on tobacco and carcinoWesis stated that if in the 19511's the author had not reported «S skin tumors .hen other laboratories were I finding 3 to 5S, this field of research might have developed in a I I more orderly .ay. ~ ~ On what basis is the claim made of 90,000 lung cancer cases per year? It is based on death certificates, which are not scientific doc.inen:s, and in general are not corrected by findings if an autopsy is performe!. 'rhen analysis has been undertaken of the accuracy of a death certificate diagnosis of lung cancer, when compared with autopsy findings, slight?y less than 50 percent were found to be accurate. Pneumonia and tuaors spreading to the lungs from other parts of the body, so-called metastases. are most often mistaken for lung cancer. Regrettably we lack a scur.d data base by which to establish the true incidence of lurg cancer. :he figures appear exaggerated two-fold at least. Lung cancer is not an entity either pathologically or etiologically. There are some 10 different varieties of lung cancer, each of .hich is likely a different disease with a different causation. One type called oat-cell carcinoma recently was reported in a group of chemical workers s»st of whom were non-ss»kers, and it appeared that smokers in the sa.:.e plant were in some .ay protected. Time, scholarship, and money are needed to work out these difficult problesss. Chronic pulmonary disease, also called bronchitis-eophysema or chronic obstructive pulmonary distase, is a serious public health problem. Whe-. CTR HN 0 147`96 I

. An English book revie of a ted that if in the 1930's the s .her, other laboratories were h a:ght have developed in a D,,)J lung cancer cases per year? ch are not sci" fic documents, ndings if an autopsy is performed. he accuracy of a death certificate j with autopsy findings, slight'_9 accurate. Pneumenia and tumois s of the body, so-called srotastases, .. Regrettably we lack a sound je incidence of lung cancer. The least. thoiogically or etiologically. ',Df lung cancer, each of which is erent causation. One type called ` ed in a group of chemical workers appeared that smokers in the same e, scholarship, and money are blems. bror.chitis-emphysana or chronic rious public health problem. r'hen 285 THIS DOCUMENT SUBJECT TO -6- CONFIDENTIAUTY AGREEMENT. a disease has several names, it is an+[ndicatlon that not very much :s knom about it. Dnphysema of the lungn is difficult to diagnose clinically. Also pathologically it is difficult to establish its presence, type and exteat, even when a panel of experts examines w.'.c_e lurtY sections. These large sections are only available in a few re- search centers. Experts examining such sections, it was repor:ed, could reach no agreement on whether, how much and what type of e.:,`,v- sema was present. In the current limited state of lmowledge, blaming cigarettes as a cause is premature. Note that emphysems is largely a disease of .:`_:e aen. Blcak men exposed to the same environment including cigarette smoking, scarcely ever develop esrphysesv. Women likewise are uncocr=-.:;• affected. 3luch investigation, including animel models, Is needed. A claim that the cause of amphysema is 3mo.n, if anything, tends :_ _-.- hibit research. Coronary heart disease is considered the most coamon cause of dea:. :-. the U.S., affecting particularly msles, the over.eight, the diabea :. and individuals with certain inherited lipid diseases. The Framingna data originally publicized do not withstand critical evaluation. S:ui'_e_ of twins discordant for smoking have indicated that the strongest factor is evidently genetic or constitutionol. A certain persoraii:;• type haa been discovered with a striving, time and pressure life s:y:e, evidently particularly predisposed to coronary disease. Today ac:ive researchers look elsewhere than cigarettes for the causes of coronary heart disease.

286 THIS DOCUMENT SUBJECT TO . GdFlDENTIAUTY AGREEMENT. _7_ . A decision to tax cigarettes in proportion to nicotine and 'tar' .ou:d De based on douDtful, specious or mistaken inform.tion. The orig:r.a: Surgeon General's report,.for e:aiaple stated that nicotine is proba::y not a significant health hazard. No new persuasive studies con;rad:c: this generally held•opinion. Ftiologic conclusions based uporn ciga- V rette condensate (so-called 'tar') suffer from the demonstratiorn t`.a: fresh condensate free of artefacts of storage has virtually no t=cr producing quality. In suaraary, I aish for Detter health for Americans, and have*.orie: Aard•in the field for 35 years. In sd considered opinion the Di:i pro- posed is scientifically unjustified, and if passed would generate suc': Duge sums of money as would be likely to coopt and to corrupt eos: cf our medical and scientific research effort. , ~ JtvE •.J CTR HN 0147"96