Tobacco Documents Online

*1). ` • • . 0 ,., -a$7f ~,. Less Harmful Ways of' Smoking l Ernest L. V'b'ynder; M.D., and Dietrich Haffneann, Pb.D:, Ameriran Healt~h Foundotibn, New York, New York 1C1C)1'9 , all forms of smoking, therefore, we need to ac- IN THE 4R,EA of smoking, it is our hope that someday we will be a society of nonsffiokers.. Obviously, from a look at the social, pplitical„ and economic behavior patterns of today, such a perfect society is not within isiilnediate sight. So,, while keeping the ideal in front of us:as the ultimate objectiti•e, there is much that can and, should be done abong, the road we are following. Concurrent with our eftioru to achieve a society that rejects celerate our efforts to find and encourage less harmful smoking products. If successful, this approach would be particularly pertinent, to the problem because any improvement in the product at its source involves al!'smo}:ezs and will tend to ameliorase the: injurious efl'ects of smolcing, on the broadest possible spectrum. Numerous epidenziologic studies have shown a higher mortality rate for smokers-especially eigarette smokers-compared with nonsmokers (1);. For most diseases and conditions relatedl to 'smoking, the evidence is based on so many ,epidemioliogic and' experimental studies that it is belleved' no.further documentation need! be offered to:confirm the conclusions. In fact, fbr lung cancer, the case against cigarette srrtoking, was clear as early as 1950 when Wytd'er and Graham, (2) and Doll and Hfill (3) published their, fiist reports (texr-figs: 1, 2). More recent evidence continues to point to tobacco, usage as the major contributor to death and incapacity from a variety of cancers, cardiovascular disease, and chronic pulimonary• disease (1). There seems little doubt that, in the absence of slnoking, life expectancy would be significantly prolonged~ What can we do about this major public health problem? This cmm- aiunic3tion attempts to deal with the various facets of the issue. s.. _ ..o ` . r... I III 1II,1;; ,`. 11 : ,1, ~~ ~~ I~I NONt LIGMT MGCEAATELY MtAVY MEAVY txctsslvc GNAIN iCDMTROL iR1 LUNG CiU1eiR (MYNDER. C:L. AMD lGRAMAM. La. 14301 . TtxT-r7ctrnE 1.-Percentages for amount of smoking among, 605 1 malr patients with lung cancer and, 780 men without cancer. Ropnoduad,from Joarnal o/Uu .tmofine dYEdh mt.ltiocoalion 143:924-336,1950 (2). .ranr Cr..o..cco l cn a~1 . IQ..®w A.K,W..R.[S I Tkcr-etcuRS 2.-Percentagr of patients smoking different amounts of tobacco daily, Reproduced!from BriarA ldfe+diaal Jownal,2:'34. 745; 19J0.(d). t Presented ara workshop of thrSecond World Conference on Smoking and Health{ sponsored by the United Kingdom Health Education Council, held in London, Septern- ber 20-24, 1971 L 1749 1005052891

.. 1750 WY2IDLR ANtY Ht)FFtitANN HEALTH RISKS FOR' EX.SMOKERS While we know the risks incurred by those who smoke and those who have never smoked, the dam on individual$ who have stopped! smoking are less well documented. Appropriate analysis of ' ex-smokers is made more difficult because the risk for disease depends on the duration and amount of'smokirng of' each individual. For the individual who gives up tobacco after many years of heavy smoking, the risk for lung cancer appears to remain the same as if he was still smoking, for about 3' years (4•6) (text-fig. 3). For the next 14 years, the risk gradually declines to the 1eve1 of'risk for those indiwiduals who have never smoked. Clearly, certain irreversible patho- logic changes remain in, the lung of long-term, heavy smokers. ....~a,. l.. m TrxT-ncuxx 3.-Relative risk of lung cancer by number ; of years of ex-smoieing, male Kreyberg, I lung cancer (ineiudes epidermoid carcimomas ~ and "oat"' cell earcinou mu of the lung): 196b-71. t. For sufferers from chronic cough; cessation of smoking results in a significant improvement of the cough afiter 4 weeks, but this does not appear to change the degree of emphysema praenr (1, 7) (text-fig. 4). This suggests that the effect of smoke on both productive and' nonproductive persistent cough is relatively acutq and thus reversible, while the damage to parenchymal lung, tissue which leads to emphysema is largely iirevezsible. The effect of cessation of smoking on myocardial infarction has not been investigated in much depth. Evaluation is complicated by the additionall important etiologic fiunction of blood lipids andl hypertension in this disease. Data from the Na- Tsxr-F=nx 4:-ChangF in eough, among ex-cigarette smokers from va:nous firms and hospitals, New York, New Forit. tional Heart and Lung Institute's study in, a controlled population: in, Fraaninghatn, hlassa- chtuetu, suggest a relatively quick reduction of risk among ex-smokers (8) ;the study by Hatnmond! (4)' indicates that, among long-term; heavy cigarette smokers who cease to smoke,, the risk of death from myocardial infarction is reduced by 1 year, but that the low rislk level of those who have: never smoked is only reached after 10 yean (text-fig. 5). If these data are confirmed, they cYllMt smWeTS' cq,retlK smttb 104.4 ~ 1 • t1 p l~,-sl.a... 1164 S109 101014 20. hon• ranuiE.-smaen suwtrrs I Nimmond I{fdrtnpL 1NEV1 , TaxT-rsctrna 5.-Mortality ratios ftoas coronary heart disease in men. 40-79 years old; by smoking stattu: , JOURNAL OF ' TFM NA3IONAL CA'NCZR' L*7S4TrUTE . 10050'S2892

. f LESS' I'IiAR!MFWL WAYS OF SMPJKllVO suggest that cigarette smoking, has both a chronic and acute effect on myocardial infarction rates. The chronic effect is perhaps due to the influence of' smoking, on atherosclerosis, while the acute effect tmayy be due to the possible influence of' smoking in the induction of arrh.,thmia and thrombus formation (1). Understanding the occurrence of smoke-related diseases among ex-smokers is essential for the proper evaluation of'less harmful tobacco products. At the same time, it is reasonable tu assume that any beneficial effect such products may offer is unlikely to be as rapid' or as definite as occurs whem an individual stops smoking entirely. HEALTH RISK AFTER SWITCHING TO~ C1GAR' AND PIPE SMO'K'ING The risk of disease to a smoker who switches from cigarettes to cigars and/oi pipes has never been fully investigated. Here, in addition to considering the duration and quantity of cigarettes smoked before the change-over, the amount of cigar and pipe smoke inhaled by the individual who used to inhale cigarette smoke needs to be evaluated. Individlnals smokirng more than 4 cigars andjlor 8' pipes per day have an appreciable risl: for lung cancer even if they claim not to inhale the smoke (9, 10). In any case, cigar and pipe smokers are known to have a relatively high risk for cancer of the upper alimentary tract (11, 12). While 11 or 2 cigars or, pipes after dinner do not seem to be associated with any appreciable increase of lung cancer risk and their effect on cancer of the ozali Catity is also limited, excessive cigar and' pipe smoking does represent a serious health hazard. IDENTIFY'ING' THE HARMFUL AGENTS V1''hen discussing, the work area of less harmful smoking, products-a subject summarized by us. previously (1.3) and' also reviewed in the proceed= ings of'a similar workshop at the First World Con- ference on Smoking and Health (14)-accuratee identification of the most harmful components: within tobacco smoke ia obviously vital. VOL. 48, NO. 6, JIUNE 1972' 1751 NEOPLASIA - Large -scalt fractionation experiments have shown 3' types of tumorigenic components in tobacco smoke: a) Tumor initiators.-tTostly in terms of poly. nuclear aromatic hydrocarbons (PAH); they reside in the neutral fraction of the smoke. The precursors of'the PAH and the alkvlated PAH are believed'to be present to a significant extent in the wax Iayer of tobacco leaf, primarily in the fbrm of'terpenes and phytosterols. [See text-fig. 2 in (15).] b) Tumor accrlrrators,-Tumor accelerators (such as 1'-methyliradole, 9-metltvlcarbazole, and' 4,4- dichlorstilbene) are also present in the neutral fraction. [Sae text-fig. 5 in (15).] They -are not cancer-producing in themselves but tend to enhance the effect of a carcinogen when given jointliy: c) Tumor jiromotns: -Tney reside largely in the tc•ea11y acidic fraction of tobacco smoke. While phenol! itself'is a weak tumor-promoting substance in this garoup, other acidic components still remain to be identif ed. CHRONIC BRONCHlTIS/,E1w1PHYSE'MA Some of the components of' the particulate matter are ciliotoxic, particularly the phenols, as is also true of a number of the volatile components, especially certain volatile acids, -aldkhydes, and hydrogen cyanide (HCN~ ', (16). Conceivably, smoke components which destroy the defensive characteristics of'the respiratory system may play a role in the induction of lung,cancer by destroying, or changing the cilia and mucus, thereby permitting the absorption of'tumorigenic components into the bronchial epitheliuffi. It remains to be shown that the componenm contributing to cilia destruction and mucous stagnation are thus - also responsible for chronic bronchitis and emphysema, Such a sequence of events appears to be a strong likelihood. CARDIOVASCULAR DISEASE Due to its effect on the catechoiamiior reitasee mechAnecm,nicotine is believed to have a chronic effect on atherosclerosis by increasing free fatty acids (17). It may also have an acute effect on thr. 10050528~3

1752 W'YA'DE.R A21D HOFP34ANN induction of heart attacks because of the relation- ship of catecholamlines to arrhythmias and thrombus formation. .. Aznong individuals who smoke, the carbon monoxide content of the smoke they inhale increases the level of carboayher.iogiobin in the blood. It has been suggested that a relatix•e!state of hypoxi& in heavy cigarette smokers accelerates atherosclerosis and also ~ adversely af3'ects, an already atherosclerotic coronary system (18). At. present, our investigations have led us too indict nicotine as an important culprit in athero- sclerosis, but more ! worh in this vital area is neces- sary to establish whether nicotine or carbon monoxidc has the greater infiuence. EXPERIMENTAL STUDIES TumoriqRnicity In the biologic setting, it goes without saying that,, if one n:duces the amount of tobacco-smoke condensate (tar) given to Ian experimental animal, the tumor yield will! also be reduced. This has, been exiensively proved in studies producing skin tumors I in mice with tobacco-smoke condensate ~ .(19-16). Findings in: the experimental situation, therefore, support the data from dose-response studies conducted on man in terms of d utation amdd amount of smoking (text-fig• 6). On the: strength~ of the combination of both human and experi- mental data; therefore, it is reasonable to advocate the lowering of the totali dose available in a given ; tobacco product. Ttx'r-t'tovm 6'-Tumor respnnse to difCerent doses of' eigarctae-smoke condensate in acetone, 1959 and 1'960: It, has been shown that, if natural tobaccos relatively high in nitrates are selected, tumori- genicity of the resulting tar can be lowered, which is apparently secondary to a reduction of' the initiating carcinogens iun the tar (14). It has also been shoti.m that, witen mad'e into cigarettes, tobacco stems practically free of terpenes yicld a smoke condensate with little : tumor-promoting activity (text-fig. 7): raem.: 1so.v arro. . .: - . hanaer: 30%SawmeuoW.nor., ;Eecn oraw , 50% •Si.PW cone.raaa . 9f3..r. 33ec,Sronaero tEeM prao - 37%'Stmi 'cano.raan J 3G nrc. a0 50% Sanaore r- x. ® 30 i x 5 0 wRarrom / - ~ % ,_. ~-r so x Sr«n eone.mer. % 0 ~~ . ¢. % aax'5,..~ ,. e 10 12 ' 14 , MOWM b/,owame18raooNCnran iwW oofMw+W Taxr-rtouu 7: Tumor-promoting activity of smoke condensates fromicigarettes made with standard tobacco . and, tobacco stems. DMBA ~, 7,12,dirnethylbenz[al ~ ` anthracene. Extensive experiments with tar obtained from cigarettes made entirely from reconstituted tobaceo, sheets have shown that the lessening, of complete carcinogenic activity (13, 14, 19) was primarily due to a reduction of' tumor initiators. This finding is eonsistent, with another experimental fiading that tumor-promoting activity is not noticeably reduced in the tar when reconstituted tobacco sheets are used [text-fig. 1I1 in~ (1S)]. We have also been studying tobacco substitutes. In gcneral, products with a high cellulose content give a: tar with low tumorigenic activity. For instance, cigarettes made of hay can deliver a, tar with low total tutnorigenic and, particularly, JOURNAL 0F' TFfE NATIONAL CsANCER IIV5TITtTTE 100505ZS94 .

.. . LESS xAPMFtQt: wr.Ys oF sMOt:mc! 1753 a tumor-promoting, activity. Sirnilarly, our current long-ter= biologic studies with a tar obtainedl from cigarettes made exclusively from oxidized celltalose suggest a significant reduction in total tumorig;nic activity to mouse skin, a finding which we believe to be due to a trductv.oni of tumor promoters [text-fig. 12I in (l.i)1''. We consider the addition of h such materials to tobacco products as another promising,approach to the reduction of turnorigenic activity of tobacco smoke. As these various experimenm have shown, thee ttunor-initiating, and tumor-promoting capacity of . eigarette smoke: ean be selectively redinced by various methods. ~ Cardiovascuiar Disease Ftelati.-ely little experimental work has been done om the effect of'd'ifi'erent types~ of' cigarette smoke on card'iovascular factors. A study by Kershbaum a at. suggested that smoke ~ from filter . cigarettes compared with smoke! from regular cigarettes was not beneficiall in terms of a reduced effect on catecholamines or free fatty acids (17), ` Similarly, studies on thrombus formation with, the use of different types of,cigarettes proved incon- clusive (20). Certainly, more work in this vital area is necessary. i . '-Chronic Pulinonary Disease Extensive experimental' studies on cats, rats, frogs, and clams have : shown that a reduction of particulate: matter and volatile components-in articular acrolein, formic acid, and HCN-will reduce ciliotoxicity of' the: smoke (1, 16). Con- clusive epidemiologic studies relating the effect of such: a reduction in the smoke to a reduction of chronic pulmonary diseases in man have not yet been conducted. . The identification of' substances im tobacco thought to adversely affect man is obviously a prerequisite for designing less harmful tobacco products to be smoked by man.. EPIDEMIOLOGIC STUDIES . ProductlvSiodificati on No amounu of knowledge based' on laboratory findings alone can replace that gained from VOL. 48, NO. 6,, JxJNE 1972' humans. Therefore„ individuals must be contin- uously monitored to assess the value of modi- fications introdidced' irs smoking products. These! studies are vitalc to achicve proof for prodiuct modi- fications introduced in smoking products and for product modifications suggested in the laboratory in: terms of'their effect on diseases in man. Needless to say, to expect a change in risk for a disease, one must have had a change in the tobacco product. Indeed, a significant alteration has taken piace in American cigarettes mainly because of more effective filters, the increased use of reconstitured' tobaccos, the use of tobacco varieties with rela- tiiveNy low tar andi low nicotinecontent, and the usee of paper which enhances the combustibility of the tobacco column. These changes have led to a reduct2om of tar andl nicotine levels in American cigarettes over, the last 16 years (text-fig, 8). Similar modifications are reported from Gerrnany and indeed throughout most of the world (21, 22) (text -fig. 9). ~ i r i ra ~ i ~120 C r Y' r = t r i i t 23 L I_10 1958 160 '62 'di''66 '66 - 1970 Yeor T=r.ncznes B.-Tar and nicotine content of'best,selling American cigarettes (8090 of totall sales: 7,20 brands), 1958-70., I'he average nicotine content of tobacco grown in tlie United Siates has not changed significantly in itself (text-fig. 10),. However,, the increased use of tobacco sterns and sheet in the blends used primarily for manufacturing filter cigarettes (text- fig. 11), has resulted' in a lower nicotine yield as a whole,, decreasing as much as 40% on a, sales- weighted basis between 1958 and! 1969 (text-fig. 8).- 100'505259S

s t 1754' 1.7 1.6 \ ~1-5 °`1.4 1: Year WYNDER' AND: HOFFNtANN 19M 25 fJ TiExT•nouRL 9:-rluerage wet total particulate matter (TA\i) and nicotine content in the mainstream smoke of' 6erman eiRarettes (90C,'e of'totalf market), 1961-70. _.~..~~.- . _ . ..\.. ...... 1961 1!96a1 1961 0t t ii11i i 19541 '56 '58' '60 '62 '64 '66' '68 1970 Ye2 r : uo-'..;.. ,- Tztr-ncmta 110.-Nicotine content of a tobacco blend (671a fine-eured; 33I~"'p' Biiuley ) and of ' the fil'ler in the best selling American, cigarettes ('80t,'-o of total sales:, 7-20 braads). During the same period; the tobacco industry reported a! 30 7c decrease in tar on a sales-weighted' basis and' pointed out that the sales of filter ciga- rettes increased from, 10~-80 7, ' of the total du'ring, the: period' 1954-70. Piedicted Rctes To the exent that tar is associated with the deveDopment of' cancer in manj we should expect a re'duction of cancer among individuals who have smoked! low-tar cigarettes for at least 110 years: We Stlms ~.~.-'+ _M-0- I 1I I I C 1 l ! 1 1! 1961 '62 '63 '64 '6S '66 - '61 '68 '69 1970 Year TExi-ncuRZ 11'. Use of tobacco sheet and stems,in A'meri- ean 1 cigarettes. mayy well not observe any significant difference before that time, jpdg4ng from'the results obtained among ex-smokers. If nicotine does indeed afl'kct cardiovascular disease, to the ~ extent I believe possible, then we should observe some redinctionin risk for the disease within 1 year of the stnok'er'S changing to filter cigarettes, based' on studies by Hammond (4). If' cardiovascular disease relates primarily to carbon monoxide,, no changes would, be observed after the smoker shifts to: filter cigarettes, since the carbon monoxide content of filter and nonfilter cigarettes generally does not vary significantly. If the par- ticulate matter is responsible for chronic bronchitis and' emphysema, we should' observe reduced rates of these conditions antong filter cigarette smokers; but if these diseases are due primarily to volatile acids and aldehydes, then' any reduction observed would only be noted among individuals smoking cigarettes with charcoal filters. Current data suggest a correlation between edia- eat2on, socioeconpmic status, and smoking habits (23),. In any population, the higher the educational and' socioeconomic status, the greater the number of nonsmokers and ex-smokets. Data also indicate that low-tar and lbw-nicotine cigarettes are smoked more by individuaIS in the higher socioeconomic: groups and are smoked more by women than by men, We can, therefore, predict that smoke- related diseases wU increasingly become conditions of' persons in the lower socioeconomic strata (text- figs: 12, 13). The continuing surveillance of the. JoURhAL OF THE ?LATIIDNAL CANCER IIds4IIUTE

.. LESS H,ARHFtJL wAYS' OF SMOKING 1755 smoking habirts of Btitish physicians, and habits concerned with cancer of the lung (25, 76)L Cur- noted noted among our own cases, shows tliat this is no longer jusra prediction (24). me r- Y " .L AO=CAftl1 tDnV 14111 t7n lYl . /01 (7n =) ,"a"~ = , °ou:ras . TExT-rteunx 12. Percentages for nonsmoker, ex-smokeryd and filter cigarette smoker by educational level: 436 male eonmoli patients, Memorial' Hospital, New York, New York. , - MQ~.0/N6f./131) uU PM 4171 = -0, = Wo ta/ Tixr-rzcuRt 13.-Percentages for, nonsmoker,, ex-smoker, and filter cigarette smoker by educational, level: 328 female control! patients, Memorial Hospit.t4 New York, New York: RISK AMONG FILTER C1GAR'ETTE S1w10KERS Studies that have investigated the risk of diseases among, smokers of different types and brands of' cigarettes are limited and at present have only been VOL. 48, N01 6, JU'NE 1972 rent studies by the American Hiealth Foundation are.evaluating the effects of all types and' brands of tobaccos on cancer, chronic bronchitis, and!emphy sema. These studies have now, been extended to include myocardial infarction, cerebral thrombosis, and peripheral vascular disease. The information required by these studies includes the different brand names, cigarette lengths, and other para- meters which contribute to:the total exposure to cigarette smoke. At this time, we shall limit our report to a com- parison of the risks ~ among filter and nonfilter cig- arettes as they relate to, cancer of the lung. The findiisgs reveal! that, compared with regular cig- arette smokers, filter cigarette smokers of at least 10 years' duration have lowei risks of developing lung cancer. These data are: strengthened by the evaluation of an additiona188 11ang,cancer patients (text-fi'g: 14). Of 82' patients who, smoked filter cigarettes for 110 years or more, 73.5% smokedl more than 20 cigarettes a day compared with, 63.610 , of the lung: cancer patients who smoked, regular cigarettes only, as reported previously (26). The liang, cancer patient who smoked filter cig.-, arettes for more than 110 years will have to have smoked 'more of them than the lung cancer pauentt who smoked only regular cigarettes. V1re.have: nott observed a similar finding, among the controls. Thus the smoking, of filter cigarettes for 110 years or more: tends to reduce the risk of linng, cancer by Tc=-nouxc 14.-Reliative risk of lung cancer by number of dgarettes smoked per day, ma1e,,Kreyberg I„196fi-7l. 100505289'7

f 175'fi' .• WA^.vDER' AI+fID HOFFMANN about one-thirdL Our data suggFst that indiaiduals smoking filter cigarettes with a very low-tar yield have an even greater reduction of' lung cancer risk after 10 ~ yeats. I Ald eurrent smokers of filter cigarettes began smoking before the time filter cigarettes were widely availablk, and consequently alll started with non- filter varieties. The data suggest that individuals smoking 20 or less of r'lmerican.ry-pe filter ciga- rettes a da'y from the beginning of their smoking, eareers'would have a relatively lower risk for cancer of the lung (text .-fig. 15), ?Lxr-ftotJxz~ 1 S-PeKentages of patients by number of eigarettes per day and filter versus'l nonfiltcr. In agreement with data reported by Bross and~ Gibson (25), these findings indicate that it is the tar that is carcinogenic to: man, since volatile components are not significantly reduced by f lters'. com¢uonly in use at the present time. RISK AMONG CIGAR AND PIPE SMOKERS Results suggest that smoking of'eigars and pipes exelIIsively carries a lower risk of lung cancer than: smoking of cigarettes. However, very heavy cigar an& pipe smokers, even if they do not~ inhale, incur an enhanced risk (9, 10),- The risk for such smokers to develop cancer of the upper alimentary tract is at least as great as that for lung cancer, and most studies suggest that cigarette smokers have a somewhat greater risk for this cancer (Il, 12). The effect of cigarr and pipe smokiitg, on chronic pulmonary and cardiovascular diseases appears to be relatively low, except when cigars and pipes are smoked inn excess. The differences in risk for lttng cancer, chronic -pulmonary disease, and myocardial infarction between cigar' and pipe smokers and cigarette smokers appear to be reiated! mainly to differences in depth of' imhalation (1). The nicotine content of cigar and' pipe tobacco is higher than that of cigarette smoke, and the fact that the carcino- genic potential of the smoke is strong is well e%idenced by its effect on the oral! cavity (3). LESS HARMFUL SMOKING HABITS~ It is unlikely that there is any component inhaled less harmful to the human system than unpolluted air. We are therefore referring, to less harmful ways' of' smoking in relative rather tltann absolute terms. Less harmful ways of smoking would include the following: . Ioninhalatipn Fewer puffs Longer butts Low-tar/'low-nicotine cigarettes Moderate use of! cigars and;/or pipes Definitely,,one less harmful wayof'sxnoking is to inhale: the sm'oke as little as possible. Other ways include puffing on each cigarette less fteqtnently, extinguishing, thC' cigarette with a long butt, the moderate use of cigars and/or pipes instead of cigarettes, and'the switching to low-tar/1ow-nicotine cigarettes from those with high values. Some of the differences in lung cancer risk among various' population groups relate to the frequency of puffs and length of butts to which cigarettes are smoked. For example, the English smoker, who has tio, pay a high' price for his ~ tobacco„ smokes his cigarette far shorter and puffs more frequently than his Americancounterpart (27). JOU6tNAL 0'B TPTE IifATIONAII CANCER LYS'rMTrE 100Si052t598'

I LESS HARbiFTJI. WAYS OF SMOI:INiG. To encourage better smoking habits, therefore, our task is an ob%ious but not an easy one. Firsty we need to improve and widen our efforts to prevent young people from starting to smoke and to persuade: current smoken ~ to give up the habit. Secondly, we need, to instruct individuals on how to smoke in the, least harmful manner, if they cannot stop altogether. Thirdly, and from a public health, point of view,, perhaps the most effective approach overall is to make certain that all smoking products on the market are rendered as harmless as possible. In concik2sion, this communication has suggested introducing ways of smoking that may be less harmful than ever before-ways primarily base& on tlie principle of bringing less smoke in contact with lung tissue which wili', result in less absorption of'tobaccq smoke. We have shown that cigar amd'd pipe smoking is generally associated with a: lower risk for smoke-related diseases, largely because cigar and pipe smokers rarely inhale deeply. Whaty from a total point of view, scems to be most im- portant and most practical is that we have shown that certain types of cigarettes are less instrumentaSl in the induction of'lung, cancer than others and are also likely to offer a liower risk for other tobacco- related diseases. Continuous monitoring of human smoking habits is v3ta1 to the appropriate evaluation of :-tobacco producu presently available as well, as changed and modified products that may be offered in the future. This information should provide appropriate guidelines and incentives for tobacco industries and governments interested in producing products designed to make tobaccoo smoking,as harmless as possible by setting maximum, permissible levels of the di(ilerent' harmful com- ponents in the smoke. ~1"hile in some countries the tobacco industries are already moving in this direction on their own, appropriate governmental directives to accomplish this end are favored where IIeceiSary. As stated at the beginning„we need'to be practical in ~this life as well as idi:alistic. The gwal of physicians is to profect the healith of the public. Since tobacco, usage has such an adverse effeevon health, we see VOL. 48, NO. 6, JUNE 1972 1757 ini the reduction of this risk one of the greatest ehallenges in, the area of public health. We need to overcome the scientific, legislative, and social obstacles for the sake of' the people who have placed their, faith in the hands of research and medicine. REFERENCES (J) ULS. Public Health.Scrviee: The Health Consequences of 5moking, A Report of 'the Surgeon General: 1'971. Washington, ULS. Public Health Serv. PA:bL No. 71- 75'13, 1971, 458 pp _. . ., _ (2) WYNars EL, Gm"wsc EA: Tobacco smoking, as a possible ctiologie factor in bronchiogeaie caneinom& A study of 684 proved cases. JAMA 143:329-336, 1950 (3) Dot.L R, F3tu. AB: Smoking and carcinoma of the lung; Preliminary report. Brit Med J 2:739-748,,1950 (4) H..kxoxu EC: Smoking in relation to the drath~rates of one million men and womem Nat Cancer Inst Moooqr 1!9r127-204, 1966'. (S), hAm HA: The Dbrn.study of smoking and,mostality among U.S. veterans: Report'on 8yj yean:of obser- vation. Nat Cancer Inst Momogr 19:1-125, 1'966 (6), Dot:r. R, HtLL AB: Mortality in relation to smnring. Ten years' observations of Btniish doctors. Brit Med J 5395:1399-141Q; 1964 (7), Wrnaast EL, Kwvrcm PL, Lzssast RL: A short-term follow-up study on e:.cigarctte smolten. With special emphasis on persistent cough, and'' weight gain. Amer Rev Resp Dis 96:645'-G55, 1967 (B) DovLz J1T, Di.waxx TR, Kwr+NtL WB, ct al: The reiadonship of',cigarette smoking to coronary heart disease. The second report of the combinad expetd- ence,of the Albany, NX. and Framingbtm, Mass. studies. JAMA 190:886-890, 1964 (9) 9mLiN T,, GQU. OR: Relative risk of pulmonary cancer in cigar and pipe smokers. Cancc, 20t1288- 1296, 1967 - (10) Wrnezst EL, M,asumu K: Lung eancer, among cigar, and pipe smokers. Prev Med. In press (11) Wvxmr.n EL, Baoss IJ, FcLnxAx R:.M: A study of the eriologocal factors in cancer of the mouth. Cancer 10t1300-1323; 1957 (17), Wrnnrx EL, Buoss Ij,: A study of'eriologjcal factors in eaneer, of the,esophagus. Cancer 14:389~-4',113, 11961 (13) Wrrroest EL, Horrvmx~ IDt Reduction of tumori- geaicity of cigarette, smoke: An experimestal a~ppnoacii. JAD4A 192:88-94, 1965. (1! )~ Wsrt.ror.m , EL, Hsa rnwx D, eds.: Toward a, Lesa Harmful Cigarette. Nat Cancer Inst Monogr 28,, 1968, 282 pp (15) HorrxkNN D, Wmntse EL: Selective reduction of tumorigenicity of tobacco smoke. Experimental ap- proaches: II. J Natl Cancer Inst 48:1i855-1868,;1'972 1005052899

e ' 1758 WYNDER AMD I3/9P'F!'IANN (l6)~ WrnDrat EL, HornaTtx D: Tobacco and Tobacco Smoke. Studies in Experimental Circinogenesis. New York, Academic Press Inc., 1967, 730 pp (17) KERSas,a,nt A,: Btt.raT' S, HneoawrASSt' r1, et al:' : Regular,,Sl!ter.-tip, and modified cigarettes. Nicotine exeetion, free fatty acidimobilization, and catechol- amine excretion. Ji',\La 201:545-546, 1967 (18) Asratur P: Effects of hypoxia and of carbon monoxide exposures on experimental atherosclerosis. Ann Intern Med 71:+26--427, 1969 (19) DoNTLxNaL' W„EtxeYHORsr H, Ha'Rrs HP, et aI: Experimentelle, Untersuchungen Qber die: taimor- eaeugende Wirking voD. Zigarettenraueh-Kanden- saten, an der AI'3usehaut. II. Einzelvergleiche zNi.chen den Kondensaten,modi6zierter Zigaretten. Z Krebsforsch 73:285-304, 1970 ('I0.). SCHOLNDDRrTH,, WILR'DYtN4J,. CLtrrfoN EE: In9itence of'cigarette smoke on some blood coagula- tion. JI Tled' 1!:'117-128„ 1970 1 (?1) Ttuu J: Einige Tiendanal.ysen zuta Problem, des Zigarettearauchens in der Bundeaepublik Deutteh, land fur die Jabre 1961-69. Beitr Tabakforseh 5:193-197; 1970 (22) Ttsut Jj MtsrEtD JI: TrendanalyseD1 zuta Problem des Verbrauches an Iitikotin und Rauehkondeasat in der, Bundesrepublik Deuraehland fur die Jahre 1961- 1970: Beitr Tzbakforsch 6:5 1-55, 1971 (23) HznRSa Ja.: Facts on Smoking, Tobacco and Health. Washington, D.C., U.S. Govt Print Off, May 1968, 134 pp : (?l)' Royal College of Physicians: Smoking and Health, 1Wow.,Londoa, Pitman,Dirdical and Scientific PubL Co. Ltd., 1971„ 148 pp (?.9)BAoss IDJ, GmsoM R: Risks of'ltsag cancrr'in smokers who switch to filter cigaretteu Amer J Public Health . 58:1396-1403+ 1968 (26) WyxDrx EL, MASucm K, Bzwrrn: EJ!Jn: The epi- demiology of lung cancer. Recent tncnds. JAMA 211:22211'-2228; 1970 (27) WYxDEa EL, Fwztecass;D P: Unpublished data