Document EqXBoYwEB6Zqy1qjJDxV477mx

Volume 71, Number 2 F E IS III FA 15 Y 2 o a a S'ttbli.iltetl 187U COINTEZYTS h}>i<icmu>tu"V of leukemia- -- IMi K review................................... Cilcin F. Mradars* Glues lo suieiile....................................................................................... lulwin O'. Shttoitlninn and Xnrmutt Fnrheroiv. Cardiovascular tliwa**** ami public health................................. John W. Ferrer. The epidemic climate -1*11 It review............................................. If'iiiston li. / Vj'rr*. Unfiuisluti business in muicmal ami child nutrition............ Marjorie M. Hesehine. The tJ.*e of insecticide trrsiled cords for houseHr control. . , __ John )T". Kilfmtrick and li. F. Sehanf. Man** emergence toward heaIth~-The C.-K. \. \\ iiicinw lecture. 1955......................................................................................... Henry van Zifa Hyde. i*ul>Hc health tu IVnttM ivattia.............................................. lUruyn F. Maiiisan. . Kpidcminlogical t<*sl* of theories on lung eaueer etiology . . If iiliant linens zri. Health itm intt in ei\il defense- -A swmjhimuih....................... Hinlopral-medieal emisideratioo* in aloutie defense. . . Edwin (). If illintus and Samuel C. in^ndnun / /. iludiolngiru] defense. ....... . . ................................... Simon Kinsman. (ihctmenl -weapons.............................. Hurry l*. Krtnuer. ..................... IIT 105 )09 k 115 125 159 1 11 irl 100 165 175 175 1KI II t. Epidemiological Tests of Theories on Lung Cancer Etiology Sy WILLIAM HAENSZEL Questions n the etiologies of lung cancer cannot, bo resolved by a single. de cisive experiment or set of observations, ami a single study is not the natural unit for discuss* ingiheepidemiologv of this disease. 'Hie maj<ir concern must be with the interrelationships of findings and their collation to see. wiiether a ehaiu of evidence stronger than any single coiu|s*neut part can be forged. Information, on a pro|n*sed etiology cannot l>e assessed without relating it to both positive and negat ivo findings on alternate possibilities. drawn from animal experimentation and other lalxu-ntory work as well as from epidemiological studies. Also, with a subject under such active scrutiny as lung cancer, it is diflicult to attempt to divorce facts already established from interpretations which may either l>e stated explicitly or merely Ik* implied by investigators tlirongh their choice 'if specific studies to bo pursued. Under (huso conditions, a better pers|eetivc on epidemiolog ical issues in lung cancer studies may be gained liy viewing them in the jnjrspective of ]K>ssiido future, work. Since, interrelationships of indi vidual findings arc tlie. main concern, Uus pa(x*r /1/r. Haenzzei is head of the Biometry .Section, Biometry and Epidemiology Branchy National Cancer Institute, Public Health Serx'ice. lie pre* scaled this paper at a joint session of the Biometrics Society and the Institute of Mathematical Statistics, Chapel Hill, iY. G. April 23,19S5 deals more with the ideas underlying a set. of studies than with the tactical details of'auv single study. Increase in Lung Cancer If no real increase in lung i-am-cr existed. all associations 1k*(wccii lung ;.i--i*r and r.\j<o-i to atmospheric jmlhition, ciga.. ite smoking, or oilier exogenous agents rereui ly hurts hived into tiic environment would U* suspect as not point ing to |>ossihle etiological relationships. A few persons still maintain that, ttie recorded increase ill lungrnncer incidence ami mortality is lictitious and has resulted from improved diagnosis (7), hut- this is no longer a popular position (2). Among the reasons for accepting a real increase in lung cancer mortality are. I lie. sex diiferetices in rate of increase. and the magni tude of fhe current, continuing increase (./). One puqmso for itupiiries on the increase in risk won hi lie to provide more prtviw ijnantitalive estimates Ilian are now available from mortality data winch may later prove useful for reconciliation with uuuirrirnl measure- of cxjxisuiv to suspected agents. For this reason, anyone prnjmsing to study tlie. elfeet of possible misdiagnosis of lung cancer as luliennilosis t.}) or some, other cause should relate, these cihnts to data produced by the death registration sys tem. Samples of dentils from lung cancer and other causes taken from vital statistics sources could !* compared with autopsy protocols and other diagnostic, evidence. Evaluation of the VuL 71, No. 2, February J9S< i v\ < i ><inrci-> of /`iruf in official iIoni iis from '.iIht r:niN>s i,n,um,>ii>ly recorded :j> J ii (fir <" iul (//) lung r.HKTr deaths a!:* U'llllllcd to of i I I' raitJM'S, WOllfil !* rendered by (he l ct*i ive character of aufopsv i i>( irs. 1 unvm>r, Liu* e (fort. >hou M l>e made. A useful niiniinimi objective would be to deter mine limv many lung cancer c:im*s remain un diagnosed at present. Karher reported that in a -cries of 1,070 morphologically proved cases of bronchogenic carcinoma, 01 jxnviMit were not posh jvpiy diagnosed prior to autopsy (d). Un doubtedly, many of these deaths were not re corded as lung cancer on (ho. ilealh certificate. Environmental Agents Tim working hypothesis for most investiga tors has Itecu limit the. prohuhh* riliws of lung rawer ary to ) found :imoti^ environmental agents (). 'Hie association of the high pre valence of rancors for several other sites, par t irnlarly the moulh, with specific customs found only in certain parts of the world >1 lengthens tin* lielicf that direct. emit act with known or suspected carcinogens may ho involved (d). On the assumption that agents with opportu nity for direct contact with hmg tissue, are the most, likely candidates. efforts to establish etiological relationships U't.wcen lung cancer ami environmental agents have Ikc directed into three main areas: (//) airlxirne agents en countered in special occupations; (7>) attnosphcric- |K>lhtaut.s, particularly those resulting from eomhust-ion of hydrocarLms; and (*;) smoking of tohacco. To assort that an agent is etiologjeally re lated to lung rancor simply implies that its re moval from a defined |spula(i<m will result ultimately in a marked reduction of lung can cer. 'The [topulution must. lie defined in a maniht which safeguards against. an edeet. being produced merely hv the processof select ion and rlasMltcafitm. 'Hie concept does not. require de tailed .specification of a mechanism producing ` the elfind. GoidlxM'gvr's finding that pellagra may )>c prevented by diet was no less valid be cause subsequent work led to the identideation of the vitamin B complex anil, finally, to nico tinic- acid as the cUoct-ivc a'rout, This state ment does not imply that ideas on mechanisms are nut hclpfui; to rhe contrary, they mav >ug- gest shannons m wliicii data can be collected and tested. ' Occupational Carcinogens* Thou* is general agreement on tlje etiological significance of several occupational exposures, and such findings have been important in es tablishing the principle of multiple etiologies for htii"' cancer. Long-term observations of closed imputations have convinced most people that the Schneelierg ami Joachimsthal miners (7. /) and certain groups of cliromatc workers (0, 10) were. subjected, by virtue of their occu pations, to excessive lung cancer risks. 'Hie evidence ott chrouiale workers is purely epi demiological, not confirmed by animal experi mentation, and was first subjected on the basis of clinical observations. Although mauv hivestigutors have deduced from tlie Schneeberg and Joachimsthal observations that radiations emit ted from radon wore, responsible, Loren/, felt that the case had not been firmly established and that other possibilities, such as dust pneumoco niosis. had not been conclusively ruled out (11). Ivxcess lung cancer risks detected by exami nation of occupational mortality data, such as data ott workers employed in the preparsition of coal tar products and producer gas (72). have also been gentrnlly accepted as reflecting an etiological relationship between lung cancer and some suspected carcinogenic agent in these products. 2vor is there any great disposition to f question the. status of agents delected by clinical I observations and retrospective, studies <*f agents such as arsenic dust, nickel carl>onyl. asln^ios dust, and coke oven fumes {/>). When the study results arc reviewed to pick up occupa tional exjMwures whidi appear rejwitedly. rhe danger of being misled by a highly unusual re sult in a .single study is minimized. Additional occupational risks will no doubt Ik* demon strated in the future. However, because of the small mmdxjrs of workers exjxjsed, the further denuroeration of very restricted types of risks will not be crucial in working out. the epidemio logical details of lung cancer. This statement would not apply to the common and widespread agents contributing to atmospheric pollution, including hydrocarbon combustion products H.4 . . Public HecltH Report* ,1.-. juiur*. mui :ilT ; >J:tnC wastes rinci U<'! 11 > Ullli 111 ii " . , . / , /hi. ; ! 11 l ! i r r o Ml i (fill i O* :no(<ir \-i`iiic'ic exjiausfs. iogical work !i;i> ln-rii ii;;miJi-ci'i| ;>v ;he iiuiiiiitv f o cias>i fy l mi i i v i< i i iii i> iii (iir genera! population / Atmospheric Pollufanfs quantatauvvly witb respect l.o atmospheric ox* poMires. Menningt id fiiMm-ic?, arc hard to col- To date, atmospheric pollution theories have loot, ami in mobile populations, problems of not U*en well documented by demonstrations of classification by residence histories become com excess risk among workers with heavy exposures plex. Wlii'ii leads on specific tvpc*s of atmos to common atmospheric pollutants sucl\ as may pheric polhiiaufs are obtained from studies of exist in railroad yards, oil refineries, and con special situations, Ihr way mav then !*e ojamed gested urban areas. "With die advent of group for return to ihe problem of reclassifying the insurance and pension plans, the methodological general population with respect lo special ex* problems of such inquiries have become much posures with gi-calcr chance for success. simpler. Severn! cohorts can be located for study. The National Cancer Institute is fol lowing a group of 1,100,000 railroad employees Smoking of Tobacco who have been in the industry 10 years or longer, Although at least. l-l retrospective clinical including station, oflicc, yard, and shop cm* studies (-17) have reported <m the association of t ployoos, as well as train crews. Rather detailed smoking and lung cancer, these studies have occupational histories arc available and. with been criticized on a variety of technical ground.-. (lie accumulation of mortality data over several The objections advanced cover such points as years, such specific points as the cfleets of diesel- interviewer bias (interviewers itt some studies izat ion, prolonged exposure to smoke in railroad knew the. identity of lung cancer patients atul tunnels, and metal dust exposures in shops may controls at tin* lime of* interview), and bias lx* looked into. The important feature is tliat arising from selection factors associated with work histories and exposures be specified in liospical admission (in most studies only hos some detail. Significant effects may be diluted pitalized lung cancer patients were inter and overlooked by considering only gross com viewed). However, the objection based on pany or industry experience. knowledge of the identity of patients and con The widespread use of petroleum products trols ignores t he evidence of Doll and 111 11 (18) makes it. very desirable to gather data on work that- males erroneously interviewed as lung can ers in that industry. Some medical departments cer cases, as established by later events, showed of the major oil companies have staffs well smoking histories characteristic of the control equipped to undertake such studies. The major Hither than of tho lung cancer series. Rceu use problem here, as elsewhere, is to secure access to of possible selection factors involved in taking company records and to rearrange data or up the-smokiug habit, some feed that the tobacco- ganized to meet other administrative needs. In Inng cancer associations reflect associations with areas where petroleum operations are coneen- other common, but still unidentified factors. (rated. data might be collected on a geographic Tho present data ou assK*iatious IkM.wccu rai her than an industrywide basis. At least one smoking and lung cancer are the most extensive study is under way in Oklahoma which may and afford the most opportunities for discussing vivid some information on lung cancer risks additional studies. For orientation, the tyjnj of among oil industry employees. evidence now <m band mnv 1m compared with Tlmse advancing atmospheric carcinogen (he epidemitdogu-al findings for cholera, pel theories have pointed to the sizable urban- rural differentials in lung cancer risk (24) - al lagra, and dental curies, all discuses in \ylnch animal experimentation played a minor role, at though some further assumptions are required least in the early phases of the investigation fn reconcile this hypothesis with the great excess . (19-Zo).- ' risk otaorved among males. The presence of On review, ihe basic core of associations on carcinogenic substancys-in urban atmospheres smoking and lung eaiu-er must apjxuir as im has been repeatedly demonstrated by experi- pressive as any produced in thoe 'other investi- VI. 71. No. 2, February 1956 gallons. i lowover, limy aunvr 'loos not present the features of ideographic localization charac teristic for cholera, pellagra, amt the absence of dental curies, which made it possible to elab orate in x rather straightforward manner the basic associations with water supplies suspected of contamination (cholera), diet (pellagra), ami fluoridated water (dental-caries) with corroIxm(ive detail ami combinations of isolated special situation#. Sndi detail is helpful in, ruling out some association as not significant rliologically; in this manner, for example, the associations between cholera and altitude ('within ` .(don) and lWivcvn i>cllngra and mill viiiagr occupations could Imj discarded as not -Igniliraiit. For lung cam*tr, there arc as yet no nHintrrpsirt.s, for example, (o the cholera patient in an isolated rural ai*oa w)io regularly sent for water from the famous Broad Street pump {10)) to the absence of pellagra among dtK*Xors, nurses, ami attendants in mental insti tutions, although the disease was common among patients: to the peculiar age distribution for pellagra in a children's institution where the disease was limited to children between the-ages of 0 and 12; to differential attack rates for dental caries by gradations of exposure to fluoridated supplies among natives and in-mi grants; and to the sequence of dental events fol)wi ng sli i fts in the source of a community water supply. Collection of such details generally results from the study of populations. This is one of the reasons why forward studies on lung anccr among defined [>opulatiou cohorts are so important. Furthermore, l lie epidemiological models for cholera, pellagra, and dental caries have been put to the test in the successful application of control measures. Fur jicllugm, only the aildi<ion of meat ami milk (< the diet was involved (-/>). Fluoridation effects could be observed under controlled conditions because of iho happy accidont of public water supplies; if private wells were the only source of water, the problem would have been complicated by sclf- Mdection of families fluoridating their own supplies. The test of Snow's conclusions on cholera (75), which required implementation through administrative decisions on methods for control of- public water supplies, occurred much later, more than 10 years after lie h;ui formulated them. The right combinations of circumstances'to permit such straightforward tests of most of the proposed lung cancer etiologies are not present now. 'rims, the immediate lines for further work on the nature and meaningof the smokin'*lung cancer associations are to develop reason able facsimiles to tests by control measures and to assemble supporting, corroborative details from population studies, not overlooking any negative evidence which may appear. Substitutes for Direst Tests As a substitute for a direct test by control measures of the smoking-lung cancer mode! one may look for groups in which smokihg is either proscribed for religious reasons or does not otherwise fonn'part of the cultural pattern. Two of the four basic study elements required are at hand: data on the distribution of smoking habits in the United States population (col lected as part of the Current Population Survey for February I05f), and data on lung cancer mortality (from publications of vital statistics offices). The major problems would be to secure the cooperation of the groups selected and to develop procedures to obtain the coun terparts of these data for their memberships. To avoid artifacts which might be introduced by classification procedures, the primary test must be whether the overall lung cancer mor tality in the group studied was markedly less than for an appropriate segment of the popula tion of the United States and commensurate with differences in the smoking patterns. To this, other refinements may be added, such as observing whether mortality differences be tween the study and control populations dis appear when specific comparisons are made by amount of tobacco smoked. The limited number of countries which can provide reliable diagnostic data on causes of illness and death has discouraged similar stud ies of groups abroad which have unusual smok ing patterns. However, special study situations may be encountered. One illustration is cited, without judgment as to its intrinsic merit. The director of the Hadassah medical organization recently reported the absence of a single case 166 : Public Health Reports :i 1 o nr.st I.' j-virs. ;i:u; be further outhat, they did not smoko cigwetlcs but n>cd a form of Oriental water pipe {27). Israel is a country with western standards of medical care and this lead may be worth pursuing. In liu-se matters, the importance of reviewing Uio primary sources of population, morbidity, and mortality data should be stress!. - ------Other clues may lead Indirectly to population groups with unusual smoking histories. For example, Steiner (<7), on the basis of necropsy evidence in Los Angeles, lias found a possible . exception among Mexicans to the usual sox ratio for lung cancer; tho proportion of Mexican women with lung cancer at autopsy approaches the proportion for Mexican men. This suggests the need for further studies among Mexicans to confirm the facts and to uncover possible reasons for the aberrant sex ratio for Jung can cer. This might conceivably lead.to the finding of a group of unusually heavy smokers among women, a useful contrast to the experience of abstainers. Corroborative Detail* Additional details ou the nature of the smok ing-lung cancer relationship should be assem bled from large-scale forward studies. These possibilities were obviously in the minds of those who planned the American Cancer Society and the National Cancer Institute-Veterans Administration studies. Both study groups consist. mainly of men between the ages of 50 and 00, the ages of highest lung cancer inci dent*. One approach is to expand the evidence on the nature of the association by detailed crosselussUication. Epidemiologists have long ac cepted the principle tliat the etiological signifi cance of an association is enhanced if it can be shown to persist within a variety of sub-uni verses. The National Cancer Institute-Veter ans Administration cohort can be subdivided with resjject to occupation and industry. By lengthening the observation jxjriod to several years, further tests can be applied in both studies, including checks on the consistency of regression relationships between lung cancer mortality and amounts of tobacco smoked for ' .i s >;u ionn i rural ve-mienis. mi :"l- ::i .; :;. i Addil inruil Leila e;ui in* puisuml bv rui.uitig age :;C deal li lo '`.igc Mailed mink i n^" to ice if any consistent pattern emerges. Studies Among Women Lin fori anatcly fur study purposes, the spread of uagc started smoking" is not too great among men, and it may be. hard lo assemble sizablo cohorts for (he more unusual combinations of year of built and "age. started smokim*." Vomen would lie more promising subjects, since women of all ages began smoking in large num bers during tho lU.'Jo decade, and duration of smoking by women is not so closely tied to chronological age. Fur this reason, more pro spective and retrosp<*ci ive. si tidies on smoking and Jung cancer among women would be re warding. in addition. to the one scries of studios ou smoking among wmneti bi Doll and Hill (2b'), two i-i-t-UVC simile.-* are known lo lx* under way. The relative rarity of lung onneer among women imposes severe study handicaps and 1ms undoubtedly deterred other efforts. * Potential cohorts for followup would include employed women covered Uv group life insur ance policies. To minimize turnover an<l loss to observe! ion it would be desirable to restrict the cohort to employees with several years' service. - Studies of lung cancer among women should establish what part, if nnv. of (he sex diffctvn- linl in lung cancer mortality (or incidence) dis appears when comparisons are made specific for smoking class. The fragmentary Doll-IIill da!u suggest tliat x good pa it, but. not all, of the sex differential may bo accounted for in this manner. Indirect Checks Retrospective and prosj>cctive studies may lkj viewed as efforts to secure ctiologically mean ingful data, which official vital statistics sources cannot provide. since they are limited to classi fication by standard demographic variables. There is another transitional bridge by which vital statistics data on lung cancer deaths ran Vol. 71* .Vo. 2, February 1956 . 167 VT.' garamMM \f \\ jj ` 1A < i: % i [,,. unlived. Hv taking information on smoking !>;ii iwiis in i in- general populal ion (collected in Ciim-ut. I'ujMilntion .Survey foe February in;,;,) mul applying to (hem the daia on ab>o!iin*. and relative lung vancev risks by smokini; r!;js.*i as reported in ilu; preliminary results of the American Cancer Society study* (~3) and in several retrospective studies (-70), the "expoled" distribution of lung cancer deaths can U* computed by sox. urban or rural residence, geographic region, and broad occupational groups, on the assumption that risks by smoking class iu the groups studied hold also in the total population. Comparison of the observed and "expected" numbers of deaths would then diow how much of the prevailing variation in lung cancer morralitv can bo accounted for by considering smoking histories, hi this manner, some useful consistency checks on (be smokinglung cancer model would l>o provided. J*jueh consistency checks need not In' confined to smoking histories. The Current Population Survey material illustrates an economically* feasible, means for remedying some deficiencies iu standard population classifications. Al though many investigators have expressed the need for procuring information on (he populat ion distribution of characteristics under study, fiiev have often l*eh deterred by the cost. Probability sampling methods for gathering such data can provide a powerful tool to the epidemiologist. Other Possible Factors Turning from subjected environmental .agents with properties which permit direct con tact with lung tissue, several possible causes of lung cancer remain. Tlicrc is tho class of environmental agents which do not- come in di rect contact with lung tissue. Closely related would \x>. tissue changes iuduceil by virus ami other infectious agents. Constitu tional susceptibility might also be considered. I f some jjersons have a predisposition t-o attacks on lung tissue hv some morbid process, the rise in lung cancer might bo due to the suppression of "competitive" respiratory causes, such as tuberculosis and pneumonia. Finally, there is the possibility of the association of smoking with other physical, psychological, and emo tional factors which may Ik? engendered bv tin* processes iuiluenciug persons to tain* up smoking. ___ Tn dealing with lung cancer, where prusp*vts for tests of etiological t^lationslups hv trial of control measures are remote, other hypotheses, liowever unlikely, and the collection of evidence to support or discredit them should Ik* en couraged. The pursuit of other lines of in vestigation will be useful. If results are posi tive, another eject of lung cancer will have been discovered; if negative, they will buttress the interpretations applied to effects already known. Gilliam, in a general discussion of chronic disease epidemiology, alluded briefly to a va riety of items investigated by him in a study of lung cancer patients and controls, using the retrospective, case-history method (31). These items ranged from color of eyes and use of den tures to histories of illnesses and use of general anesthetics. As he hastens to say, such indi vidual findings cannot be used for any general izations. However, it may be worth while to inquire systematically into other finding?: col lected but never deemed worthy of riqxirt. Several of the other early retrospective studies were "shotgun" inquiries and collected in forma tion on a variety of sus|>ected agents and ex posures. Collectively, these results might prove useful for testing and discriminating lietween hypotheses and the planning of further studies. Several studies exploring characteristics dif ferentiating smokers and nonsmokers may l>e expected.'* The question of differences lietween the two groups arises because smokers select themselves; smoking is not a treatment which can be studied experimentally by randomized application to a population. There is a wide range of possibilities for retrospective studies on the physical*-psychological, and emotional attributes of smokers and nonsmokers. The demonstration of an item as being associated with smoking history is essentially a screening device. The further test- is whether the item, when applied to a population, produces differ ences in lung cancer risks equal in magnitude to those yielded by smoking history classifications (if smoking history is uncontrolled), or whether an effect is observed when smoking history is controlled by classification. 164 7CTfTiT<r*OmTTPITTTTr iiocianon ><inu.' diceries "11 (In' subsequent increased inns; cancer among persons with histones of ; i/Hpinicory illnesses such us influenza, pneu monia. bronchitis,'and tuberculosis, based nau- aily on ciinical and pathological impressions, have been reviewed by Doll {32), who continues on to point out some observations inconsistent with these theories. No associations of lung cancer with the presence or absence of an ante cedent- illness have yet been established by ol*-er- vations on populations through the use of retrospective or prospective studies. Retro spective studios are handicapped in this field, because patients with lung cancer ami other respiratory disorders apparently recall previ ous respiratory ailments more readily than do other ]*wsons. There are some possibilities for forward studies on cohorts known to have re- eocenul fmm specific illnesses, froo from tle recall** bins in the retrospective approach, which may 1k exploited. For example, records an lie assembled of 'World War I veterans wlio had influenza and/or pneumonia in 101$. These veterans can he traced forward in time to determine their lung cancer mortality expe rience. (A study of British veterans (40) lias been published recently.) From civilian life, one could draw on recovered pneumonia cases rejmrted to health departments in the 1030 dec ade, when these agencies had active programs based on pneumococcus typing and distribution of N'ntm. - If excess lung cancer mortality among such groups were demonstrated^ this would not neces sarily Iks a result of eifects of antecedent ill- nes.-<*-i. If constitutional predisposition of lung tissue were a factor, tlie rise in lung cancer mortality might be due to the suppression of lubcrculosis and pneumonia ns causes of death. Tf a genetic basis for constitutional predisposi tion to lung cancer were postulated, other tests could 1 < devised. One approach would be to assemble lists of relatives of diagnosed cases of lung cancer for followup to determine lung cancer mortality. Lists of familial contacts 'mild be assembled from records maintained in tuberculosis control programs. Investigations of this character might be undertaken more readily in some European countries, where pop- [>onu i;U ion riTnni '\-iciii' i -\ i>i . Retrospective and Forward Studies Tlic preceding sections iuivu emphasized (he wide. I'nngc of possibilities for the use of retro spective and forward studv techniques. Retro spective, case-history studies use as their point of departure diagnosed of a disease mid matched controls and com pare antecedent events in their previous histories. Prosp<*o.tive, or forward, studies start with the assembly at a fixed point in time of deli nod cohorts classified with respect to certain attributes, trace them forward in time, and note events occurring subsequently. Perhaps the major criticism of retrospective studies for lung cancer hiis U*eu I hat- diagnosed cases have generally boon drawn from hospital populations. .Positive association Udween two diseases, not present in the general populating, may be produced when hospital admissions alone are. studied, Iiecansc jiersons with a com bination of complaints are more likely to re quire hospital treatment, fdd, J.}). Smoking i* not an illness, and for lung cancer it is difficult to see how the smoking history could have, any inilucncc on hospital admission. Lung cancer is a serious disease, nonnallv requiring hospital ization. and roughly fonr-lifths of all diagnosed cases are hospitalized {>). The proportion is even higher when mirro-cnpically confirmed cases alone- are i*onsiilered. There would haw to be extraordinary differences hi smoking hi?-- torics l>et ween hospitalized and nonhospitalized patients and eontvols to tqefinfvrenres drawn from hospitalized cases (.#).' So far, the re sults of forward jaipulafion studies on excess lung cancer risks among smokers as compare*! to nonsmokers have agreed in geiiend with tho-c of the n*tros|K*etive studies. This suggests that biases entering into the selection of hospital ized cas<*s and controls studied rwrosjx*ctiveiy are probably not. the source of the. associations noted. The fundamental assumption underlying ret- rosjiective studies and the estimation therefrom of differences in risk.- U*tween poptihuion groups is iliai the diagnosed eases and controls each be representative of the universe chosen Vl. 71. No. 2, Febniary 1956 He; for investigation. Allhough there are precau/ ions which can by taken in devising a Sampling plan, this representative property cannot be guaranteed merely from internal examination of a single set of data. One must be guided on fids point by the comparison of results from several studies and judgment as to the possible biases operating in any setting. Forward population studies have been ques tioned on the grounds that the cohorts have not hvt'ji selected hr nrohabilhv sampling methods hh ;;.ut mdiviwn::; 'indy results cannot be gen'rali/Ail to too Uual population. Tlic expense involved in tracing cohorts drawn from the gen era) population by probability sampling meth ods would be great and most investigators have liven forced to look for populations followed routinely for other purposes, such as persons covered lv employment or insurance liencfita. The criticisms of unrepresentative sampling would appear to disregard the ex}xn*ience avail able from actuarial sources. There arc two rommon types of insurance: "ordinary" poli cies, which require applicants to pass a physical examination, and ``industrial*' policies. which undergo some underwriting selection hut re quire no physical examination. Policyholders would scarcely meet- the usual criteria for rep resentative samples. 'Hie'Metropolitan Life JtiHiranee Co. has published extensive data on Hie mortality ex|xrienoo of their indus trial policyholders and finds rather close agreennMit. with the mortality exjieriencc of the gen*' oral imputation, after appropriate adjustments for age. sex. and nun.* romjKisition (,?/). Data drawn from actuarial ex|>orien<i* nud rej>orted * in sn^h sourer* as la-ngth of T.ife. (.\W) and Thc Shf/it/ind ]}uttct)tt.% a monthly publication of the Metropolitan Life Insurance Co., indicate that, the relative patterns of mortality with resjM-ct. to cause, age, sex, race, and geographical region in insured imputations have their coun terparts in general mortality ex{>ericnce despite some differences iinnaguitiulc of rates. The co horts assembled for smoking and lung cancer investigation** do not apjiear to have been sub jects! to more rigorous selection than that en countered by industrial policyholders, nor is there reason to believe that the selection effects which did exist discriminated between smokers and nonsmokers. Even when stronger selection c(frets on mortality experience exist, as in the ..ease of ordinary policyholders, it is well known that the selection effects of physical examina tions on mortality experience wear off quickly: very conservatively, in 10 years or less (.7.9), usually in al>out 5 years. If there is any ques tion about the prospective study results on the association of smoking and lung cancer being affected by selection, the conservative course would be to discount the first few years' experi ence and to require that the early results be confinned by later experience after selection effects have worn off. The collation of experience from several studies is also a safeguard against error in gen eralizing from forward studies on lung cancer. Data from prospective studies may be sub classified by other population characteristics and further checked for internal consistency. With these precautions, the dangers of drawing inferences from forward and retrospective studies inapplicable to the general population seem rather remote. Summary . ..The size of the lung cancer problem, as indi cated by the number of lives at stake and the economic implications of any potential control measures which might later be advanced, should not blind investigators to the many possibilities for studies simitar to those on other chronic diseases which have yielded someof their secrets to the epidemiological approach. Diseases' showing pronounced variations in risk among population subgroups arc the more amenable to epidemiological study: and lung cancer falls into this category. Studies on human popula tions should continue to play an important role in delineating possible etiological relationship* for lung cancer, the mechanisms for which could then be elaborated by animal experimentation. Progress must come by the cross-checking of Hie several results of epidemiological studies, animal experimentation, and other laboratory findings. The issues will probably be settled by an evolutionary process, as was the fluoridedental caries relationship. Gradually, the meaning of certain, associations will become accepted. Dissent will die off and the debate will shift to Hie many other points in any 170 ' , Pablie Health Report*  > >r;; U<<u. :;ioao;s >L; 11 v e 'i u : r : :t *JFE*6NC$ {!) Higdon. R. H.. and Kifcbofl. II.: Stuokiug and ameer of the lung--Lot'? review the facts. Texas Rep. Biol. A Med. 11: 710-727 (1933). (2) Cletnmesea, J- Nielsen, A., and Jensen, EL: Mor tality and iocidenco of cancer of the lung In ' Denmark and some other countries. In Cuncer of the Lung; A symposium, edited by Dr. J. Oemnicsen, reprinted from Acta Unio Contra Cancrum. Paris. Council for International Organizations of Medical Sciences. 1953, 210 pp. (J) Ounn. II. L.: Lung cancer in the twentieth cen tury. J. Infernnt, 0-*11. Surjr. 23: 329-3-12 I ]!."). U) Ollliiiiu. A, G.: Some aspects of the lung cancer problem. MIL Med. 11G: 1G3-174 (1055). (J) Furbcr. S. 1L; Lung cancer. Springfield, III., C. C. Thomas, 1954, p. 4. (d) Steiner, l*. EL; Cancer: Ilace and gettgrnphy. Halt imore, Williams & Wilkins Co., 1954, .UJ3 pp. (7) ItoKotoski, 0- SaupeyFL, and Schmorl, G.: Die IxTgkmukheit der orztxwgleute in Scbncelicn; in Sachsen. Zeitschr. F. Krebeforscbung 22: 300-384 (1923). (-S) I'HIer, S.: Lung cancer among mine workers in Jou'-himstbaL Human Biology 11: 135-143 (RYU)). (h) Jlriuton. H. P,, Frasier, E. S., ami Korea, A. L.: Morbidity and mortality es;>erience among chromate workers. Pub. Health Rep. G7:S35*47 (1902). fc (J>>\ Madile, W., and Gregorio.*. F.: Cancer of the res piratory system in the United States chromateproducing Industry. Pub. Health Rep. 03: 1114-1127 (IMS). (//) Loren*. E.: Radioactivity and lung cauccr; A critical .review of lung cancer io the miners of Sdmeeberg and JoachimsthaL J. Nat. Cancer Inet. 5: 1-15 (1944). (!2) Keimawgy, E. and Kennaway, N. iL: A fur ther ctt>r 4 be Incidence of cancer of the lung and larynx. BriL J. Cancer 1: 250-298 (1947). {12) llucper. W. C.; Recent developments in environmemnl cancer. Reprinted, with additions, from the A. It. A. Archives of Pathology 53: 3G5:m. 473-023, W5-GS2 (1054). U4) UiH*per, W, C.: Envirenmentnl lung cancer. Indusc. Med. and Sorg. 25: 49-02 (1951). US) Leifer. J. Shtmkui; JL B- ami Shear, M. J.: Pro duction of subcutaneous sarcomas In mice with tars extracted from atmospheric dose J. Nat. Cancer lust. 3:155-103. US) Kotin. P- Falk, IL 1^, Muder. P,, and Thomas, M.: Aromatic hydrocarbons. Arch. ImL n.vg. - and Occup. Med. 9*: 153-1G3 (1954). i "t:i, ii. Till- IVMI lung .uni iiif -i-c ni> i L* Si fttlsticiiiu ^ : 7-15 (F I). ( M) i 'oil. It., niul 11 ill. I: 'iru 'K mg .; im| . :i ivi <*f the lung. iiru. M. . - : 745 i UlAi). U'J) Snow, J.; Snow ou cinders*. Ncsv York, The Ciumuyjjwcuith Fund. in:a5. ini j<f>. {29) G-ddbcrger, J.. Wheeler, G. A., and Sydctmriekcr, FI.: A study of the relation of diet to pella ' gra Incidence In seven textile-mill communi- lies of South f'arollna in 15MG. l'ui). Health Rep. 35 : G4S-713 (1920). (21) Goldlvcrcer. J.. Wheeler, t*. A., ami Sydcnstrickcf. E.: Pellagra incidence in relation to sex, age. season. occupation. ami ''distillling sickness" in seven Cottou-itull villages of South Can>ltnn during 1910. Rub. Health Rep. 33: 1G5&-1GG3 (1.02111. (Ji) Cohibcrger, J.. Winkler, G. A., and Sydenstrieker. E.: A study of tlio relation of fuctors of u sanitary character in ielhtgm incidence in seven cotton-mill villages of Smith Carolina in 10!<>. IMtb. Health Kep. 33: 3701-1714 (1920). (2.1) C.ddberger. J.. Wlmvler. U. A., and Sydcnstricker. K.: A study of the relation of family income ami other economic fadnrs (< jieilagrn in- * cidonce In s`viut cnttm*ntill viliagi?* nf Smith ('arollnn in 191d. I'nii..ll.-aith !{*. 2073 2714 ( 3920). (-'}) Rihby, R. G and Pntdevnld. F.: t... rsicrital action of ihioridt*H mul oihor agciUH on the teeth in the prcventlmi of tomh decay. In Fluorldnlinn as a public hcnitli iuensima, edited hy J. n. Stuiw. Wasidngton, 1>. C. Ainerlran Ass*ciiition for the Advancement ( Science. 1934, pp. 143-17S. (22) U. S. Public Health Service; Epdcuiioiigi<^tl studies of thmride waters and denial caries. Cnlhwtion of reprints from I'uidlc llcaitli Re ports. inss-w. (2G) Gddbercer, J.. Waring, II., and Tenner. W. F.: lMtattm pn*vcnUon by diet niiinnc insti tutional inmates. I'nb. Health Uep. 38: 1TJ51- 23CS (1923). (27) New York Tlim*s. Feb. i, 1955. (28) Doll, IL, and Hill, A. B.: A study of the aetiology of carcinoma of the lung. RrlL M. J. 2: 1271- 12SG (3952). (29) Hammond, VL C., and Horn, D.: Tlie relation ship bctwis'n human smoking liabits and dcatli rates. J. A. M. A. 155: inp;-132S (19341. (39) Cutler. S. J., and Loveland. D. R.: Tlic risk of developing lung cancer and its relationship to smoking. J. Nat. Cancer Inst. 35: 201-211 (1954). '. Gilliam, A. (5.: Opi>onunUh*s for nppUcnthm of epidemiologic methoil to study of ennccr. Am. J. Pnh. Ilenith 45: 1217-1257 (3953). Doll. IL: Rroociiiai ninniimiift: Incidence and aetiology. Brit. M. J. 2: 321-327 (1953). VoL 71, No. 2, February 1956 m ,-u5V-r. , IUtKwhi. .1.: I.uitilii'vnos u( ;h.* <if r:i)!< .11j:j10 data, mol rir> Hull. 12: -iT-.Vf (liWU). -I',i While. f.: .m. -i. No. in nuslioul re-csurli. Brit. i><*. rj. iw-in.--...... - '-.'.m S. 1`ni'Uf Ilvnilli Servlet*. National (\uievr Lt>UiuU*: I'wnwe illness In leu urtraji urea* <>( UnilcO State*. Cancer Myr!ii<!i(v Series N'os. i-Hi. Kraus. A. 5.: The use of hospital data in sludyinc ihc .is-ex-int ion liotwocn a characteristic ami a disease. I'tth. Health Itej*. dO; V-ll-- 12U (1334). f./7) Dnliliu. l,, .uni .'picjtetiuajl. Health omavv** .unoiic imin.-frml jxdicyhoLlers 19-th to 1'jrA se. Actuaries Tr. 3: 2M-0^S (1331 . (3.S) Dublin. L 3.. LiXk.i. A. J., nltd SpiecMwan. AI.; Lriicih of He'*. I. .\w York. Ronald Prows Co., li) 4*1, p. (.<!>) Jordan. 0. W.: Lift* comioaonclew. New York. Society of Actuaries, 2952. p. 30. (}0) Case, It. A. -M.. ami Lett, A. J.: Mustard mix ji- .somiiff, chrome bronchitis, anti lure <-am-f. An iovewtijmtioo into the possibility that po- soninc by mustard jrns in the 291+-1S war mieht lie a factor in die production of neoplasia. Brit. J. Prcr. A Social Med. 0: C2-T2 (WKil. Ninth Annual Seminar on Seafood Sanitation Tim ninth annual Seminar <m Seafood Sani tation. sponsored by the Virginia and Mary land Stale Health Departments, uas held in Solomons, Md., .September 27- 2V. 1935. More than 50 |K*tx>.* attended (lie 5 sessions, which included ait in.-jKi'tj'ou of an u\st<ir-paeking Unuse ami a visit by boat to a sob-dam harvestin? area on the Patuxent River. The first Seminar on Seafood Sanitation was held in 1946 because seafood sanitation spe cialists found that ibeir work required broad up-to-date technical knowledge. In addition to the common food plant sanitation problems, they v\erc concerned w ith the construction and o|>eralioii of municipal and private sewage disposal works; ihcy had to make bacteriologi cal. hydrographic, and sanitary surveys of shellfish-growing areas; ami lln-y were obliged to apply the results of these studies to decisions * mi the suitability of an area for shellfish cul ture. The seminar has provided a forum where bacteriologists, xaniuriuus. and sani tary engineers working with seafood sanitation in the CJiesapeakc Bay area can exchange in formation on >uch prolilcuts of mutual interest. Many seafood sanitation problems en countered by county. State, ami Federal jcr-onnei were discussed at the 3-day meeting. These included sanitary surveys of shellfish growing areas, general food plant sanitation, sanitation of oyster-shucking plants, sanitation problems in the crabmcat industry, sanitary problems in the new Chesaj>eake Bay soft-clam industry, preservation of food by radioactive sterilization, and a discussion of the relative merits of State certification of processors of breaded seafood. Included in the agencies represented in the discussions were State health departments >f Maryland. Virginia, South Carolina, Georgia, and the District of Columbia; New York State Department of Conservation; Maryland county health departments which have seafood sanitation programs; the Food and Drug Administration and the Public Health Service; and the Fish and Wildlife Service. The seafood industry was repre sented by officials of the National Fisheries Institute, the Oyster Institute of North Amer ica, and the Oyster Growers and Dealers Asso ciation of North America, Inc. William Ballard, president of the Oyster Growers and Dealers Association of North America, Inc_ and ojKiratot of one of the world's largest oyster-shucking and packing plants, told of his industry's dependence on sanitation and expressed the opinion that (he future of (he seafood industry rested with the accomplishments of the fisheries research scientists ami with the seafood sanitation specialists. ' The participants recommended that the health departments of Maryland and Virginia sponsor another seafood sanitation seminar in 1956. 172 Public Health Report* s