Tobacco Documents Online

A, !Csw 11}taOC " 11112-M7 Lung cancer among women in north-east China Q NacmilJae Press lb,. 1990 A.H. VIVu-Williams'. X.D. Dai', W. Blot ; Z.Y. Xu', X.W. Sun', H.P. Xiao', Si. Stone', S.F. Yu". Y.P. Feng`, A.G. Ershow', J. Sun'. J.F. Fraumeni Jr' & B.E. Henderson' Drparrotrar of Prr.ennre Hedicinr. b'ni•ersiry of Sovrliern Cafijoraia Sckool of A/ediciwr. Coa Anrels. CA 9003l. USA: YanoolCaricer lnsnra.u. Srrkrtdo: MD 20892. L'SA: 'Harbra .Nedica!Sckool. ffarbia. Heilaqlfarrt Aorian. People x Repnbfic of Ckuta: and'Liaonint Public Healrk and Anri-Epidnnit Stario+t. SkeayaRg. L:ioa+ing Aorisce. Peopk r Rrpublic ojCGiwa. Srary A are-eontrolstudy of lung ancsr iovolrnns intarviews with 969 fenuk patienta and 959 controls in Slmnyang and Harbn. two induitnatiaties which have amonl the Aisbest rsta of lung onoer, ein China, revukd that cigarette smoking is the main aual factoraud aonotmud for about 35% of the tuetoun aniona ryoenen. Although the amount smoked was low (the aan averagrd agbt apnttr per day): the pereeau~e of smoken among women over age 50 in tlfee aua was nerty double the nauonal averaat. At, pollutwn from eoaa buramS stovea was tmpldaud. as raka of Ilntg anoa incterd is proporttoa to ysan of apownt to 'Kang' and oaher Aeaual dimaa indtgawua to the neyoa I. additm the sumEer of nrab oookad by deep frym{ and the fteqpenry of smokinew dunn= cooking were aeaoaurf with nak of lun6 aaar. Moes ore than eoatrab reported arortplacs e:pwursa to ooal diuet and to ~noks from burning fuet. EJeqrd neka were otrerved for smelter worters and dwrsaaad nrki for tnuh workerf. hwx csromt EroneheuwttaPlyten& pneumonia. and neeat tusen.vloau oontnliuud tiPbaatfy, to Iaq oeov rnk, as did a liatory of tuber- culoaia and lung anar in fataily members. HiVrr ibtake of nroeaewr>d Megetabis was aot pouetevs apunet lung cancer in this popuVtaa. The 6ndinf were qualiuteveiy rmdar aawra the malor all tyyr of lung anar. ncept that the auocuuom with smoking and prevtots tun6 diraas were stronger for t+quaeausr oat oail aneers than for adenoatonoma of the lue& The rate of lun6 cancer among Chinese fattalea is among the highest in the world. Elevated incidence. particularty of adenoaresnotoa of the lung, has been noted for Chinese females residing in different jeograplfic area, including Singapore (Law et L. 19761, Hong Kong (Kung at af.. 19b1). Shan6llL (Gao rt al.. 1998) and the Untted States (Hinds to al.. 1961). The &* tatea an umtuuaJ because few Chinese women smoke toDacso. Within China, elevated rates of fetnale lung cancer are found in urban areaa such as Shang- hat and in rural as well as urban atne of the nortEeastero provincea of Liaoeunj and Heilonpiang (National Cancer Control Ofboe. 19110: Xu er al.. 1986). Reasons for the po- SraPhic varuuon and dusterinS of hi$h ratas of lung canoer en the northern provincea an not known. We report Eere the results from ax-control uudia coudueted in Shenyang and Harbin, the two major industrial atin in Liaoning and He2lonpan= provinces, to evaluate the role of several poten- ual riat factom In 19E3-17. inveuiptorf from the l.iaoning Provims Public Health and Anu•Epidemie Statioe and the US Natioeal Cancer Institute condncteQ a laW tmj anrsr study iaciud- 4 1.317 males (129 ooa, 733 oootrob) and 1.073 famaLa (318 naea, $55 ooeuoG). During the tame time period, imf tipton at Hubim btedieal Couef said the Umvenity of Soutbern California conducted a aae-oontroi ttudy focvwd 0o tetoale hang eanoeir (4/6 easa, 404 ooswir). lavwtiptaa from both stttdin met dw" the pianainS phase of the study and adopted a uni8ad protoool' to aacertaia and adaet oaaa and eoovola, and a oommon questionnaus for tha imarviaw component of the study. Data on risks from =*king and ait poDution amoni man and womee in Shisayanf have baa published ebewhe:e (Xo as dt., 19f9j. Herde we npott Aab amon6 femaka aarodatea with a variety of fatton, inenaio0 rampie tiaa by nearly Io% by eambininS informatiba from that two eion. Cor*wqOada.oe A.H. M.MiI`aea Raewd 27 Maeef tllOt ad in tsriset fwf 23 )dX 1!!0. Cax attenawarM We sought to enrol all newly diap.waed pnmary lun1 eaaeers in fernaks in the study aresa Eetwaeo 19E5 and 1987. Ua6a- in, the caacar registries of Harbin and Shenyani, a system of rapid we axsrtainment was established with the ooopers- tion of all the major Aoapitala serving its area (about 35 in each city). In brief, the admitting psytiaans at each parti- ayatinS Eoepital completed a can abstract form whenever a lung cancer was diagnoaed. We reodwd thae abatracu on a bi •weekly baaia and selected as elipbk caas those with pritnary. incident Ittnj anars diagnosed among female reai- denti of the study area who were aged le+n than 70 yan at the time of diaposia. The Inng anoer diapow said adl-tyPer claaaibntion were veribed loatlty in each study area by a Panel of Pulmotiary specaabsta and patl,olopw. Conaoi aehetiae Controh were fimaba tandomly seheted from the 0eawd populationa of Harbin and Sbenyang. Conttob was fn. qucatly matched by s•yrt a0e Qoup to the espawd diatr'-. Dution of eaan, wh'ss was detarmiaad in advance uaii4 the number and ap diateiDatioa of female lung aaest ow reported in the two dtiar I. 1963. A tEtas-w0e armpling pnoaedure ras used to ialra out oeatrol The iaitial umt for flandomisalm wissJthis 00 ~ es' Harbis and ia Siasyan0. Committas WWI rnedomly silactad with nplaoeaaet aflw wdglitia0 by tlee popalatioa eisua Tba we rtaedoml)r elae a botseboid QouO dos Use approxasut)r 10-23 lomdloid FouRla withia arae- aietad neighbourUood oota®iwr. L the dnal stap. amoa0 an fimales I. the 3-year ya astr0ocr ritlis the Dotteadoid wap, M na raadomly aels.tid go -, I , A rtrtrttoad, pn-0oded qonooaraaus was eeaaQ by t=aiaad intsrviewrs Who oondoctad pe<sostal ina+niews with the par tiapaats is tseir 6omo at most eits or is the lospseai/dieic. The imav;w Rdlwad iatormuioa as deograi0aie 6aun, activa and pueiw t:sokx asporma, bfeama eaeideatid and oontpatioaal 6ietoeies, diw said ooob j ptaetits. peetnsial eioory at soe®allipaaa hte0 dbtw.e, fiiaary of wb.atloeia

LUNG CANCER AMOn+G WOMEN IN C)ifWH W (TB) and ancer in first deF+e rsiww, and reproducvm factorn. Quauom on smoking included the amoutu aad types of tob.csa products aaokad, ap when smoking start. edL and for a-tmokers. a0e wbaa' smokiag stopped. To asses pwive stsoka expoaut. we asked about lifetime tai- denual exposure to tobacco staoka from cohabtants, iecliid- ing the amount and duration of exposure from each smokini cohabitant. In additioe, we asked tf the subjoct was exposed to passive smoking at each work plaa. For each ntatdena in which a sub)ect lived for thtee or mots years. we asked in detail about heatlnt and cookin= practxxs, including methods for hatin= and cookmg and tyy.a of fttela used. Several questions were asked about 'MCan><. brick bsdt com. monly used' in the north<astern part of Chitt;. which ara heated etther dtrealy by a stove underneath them or by pipas connected to the cooking stove. To a»e» dietary habits 5 years prior to intervtew, we asked sublects to estnaate their frequencies of intake of 33 food itema including staple Qatns (ncc. wheat. maux). soya bean products (bean curd, fernent- ed bean paste), dned' psas and bsaas, aatmal protetn sousan (eggs. fish. shellfi>,li. liver. poultry. porkl, ferttteatadrsaltad foods. alcoholic beversgea, and fresh ve*etablis and fruits. Also included were questions on dinposit by a physician of previous each lung diseaxa. age at lung diseaae diaposia& and if hospptalisauon was required. Information on outcome of each pnpancy. age at menarche and at menoputae was also elicited. As a quality+control measurt. interviews were cassette-recorded for review by a Beld supervisor. SronsNcal nwrbodr Thr data were edit.d. coded. keypunched and submitted to computern•d ranp and eoe.xency checka. The stausttal analyses Waa Dassd on multivaruts techniques for case- control data ( Bresbw t Day: 19l0)• Uno- udiiiooa) logistic repssaion ana)ysa were used to eatimate summary relative risks (RRs) of lung anar associated with various factors while adjusting for other factots. RRs were calculated for aU lung cancer combined and for spealfc cell typa. We praent results for squamous cell and oati3mall cell anoen combined because we had too few oaUsmall cell anoen to conduct separate analysis and because these two odl types of lung cancer are more strongly associated with t:nokin= than adenocaranoma of the lung (Lubin ! Biot. 19iI). Our analysis for adeaoarciaoma of the lung did not include large cell anan. There were too few IarBe cell attcsrs fdr ia- cltuion by cell type. In the ana)yss iadtdinB all wbjaW, the rtgression modeb contained tsrme for age (Yaa tb.a 30. 50-59. 60-69 ysan), eduation (no formal •dtsratioa, pri- mary or secondary scbool, h* rclool and hider)k smoking (non-smoker. tmokod 1- )9 cipr.nr pr day ad 1-29 ynn. 1-19 dprettea pa day and 30-39 yn. 1-19 ciprsuss per day and 40 + y.ara. 20 + dprataa par dsyy and 1-29 yeata. 20i cipntte par day aad 30-39 y..ra, 20 * apsetw per day and 40 + rean) and study oaaaa (Harbin versus Sheay><v4LX We aJw conducted aaulns rsstrtct•d to nontmokan, dehti6B the awkeB variabis ia the repse.ion model aad adjuatibB oaly o. aBtr. .dtastioa, and centre. (n - 310) admoaranoma., 23% (• - 201), squamous oi(l caecinotaas. 16% (n - 117) oatlsmall oall atvnomas and'the rematndtr wsre larlr cell arcnoma., mi:turw of other ocU types or the ca(1 type was tat ktsows (e - 6d). A tou( of 959 controls (404 in MtarbiaL 555 in Shenyani) we" interviewed. Casse (meatt age 55.9 years) and controb (ntean ap 33.t years) were closely matchad on age but asa were less educated than controls. Relative to thoee with no formal ediration. the RRs for women with prtauryr)untor school high school technical school or college education was 0.9. 1.0. 0.3 rssprctively (RR for fioe.r trend 0.9: 95% CI Oa-1.0). S•tokiq AoDin Table I shows the perunups of wotnen by 3-year age group who smoked cigarettes for 6 months or lonpr.. The preval- ena of smoking in the Ptteral populatwa (i.e. among con- trob) van•d with a0e. being much hrgher (approatmacs(y 40%) atnwto women 50 or over than among women below 30 (smoking rate 24%). but tnereased ruks wets seen In smokers at all aVs. For all lung artars combined. smoketl~ sapertenood a 2.3da;ld (93% CI 1.9-2.p iaetnsd risk of ,hatB tattoer: The a0e-; editation-'auti aty-.djiisted'RRtfor tmokittg wsre 4.2 (93% Cf 3.0-3A) for sqttaretoua cell aota3. 2.2 (9S•e. CI 1.4-3.2) for oat.small eell antxn. CS (9Sy. CT 1:1=1.9) for .deaoaraaoma of the lung and 1! (93•ti fl t.9-3.3) foe the 'othv` ate0oty-wlach included thor ditposad cliataihr,, large cell anaera, and those with mixed or unknown cell type. Moa (S7v,),ases began smok- in$ before they were 20 years old. compared to 40% of controM: the avera0e age when subjav bepn to smoke was 19.9 for aan and 24.0 for controls. The women were not heavy smoken. Few subpets (9% ase, 4% controls) smoked 20 or tnorr cigarettr per day. and the tnan daily number of ciptrettss smoked was 8.1 for wrm and 6.1 for controls. NeMerthelear, tbem w.a wmoeat varunoa in amounts smoked to show that rtab of 109 aarra stptt- (lanrty (P<0.0DI) inasaaad with ittcna.ej* numbers of aprstte smoked per day and with ittaeaao{ duration of smoking (Tabk II)'. C1ear indepeadest eRaen wen st+en with each mea.urtt of smoking eapowre wtthii aterorits of tde other. with th• a»oaatiom uraw for squamovuoat odl areinornaa than for adenoar+tiaortst. At the t:me levd of tmokiaf . 2• to 4-fold diffsream in the maptitudt of the risk bst.~wa the two oail typw were typtdly obavved. raseii,ir nr..k:w Table IU dbowe the R1Ra arooand with p.ive smoke esponar.- Orst ataoet aW atb~ .Mr adiaatiaB for personal smotinf atd thea among noe-smota>< Ei0tr,ipt per a.t of an aa aad coatrolt reportrd havieB Yvtl : at Irate o.e of thair n.daats wiJ a eQhabiraan wbe .aa a t~okar. TLan wwe ao si;ai6eaet aae-ooarol d'idrn.os is svw harieB Iind' .iti a took.e. eunpt for .otemokas wbo Y..d .ith a Wotrr t.Uo ts.okad: Aan tsa risk .n tdtsad (RA 0.7; 93% C10.6-0.f). The Ivwaf•d riak arotdaaad with a tapoam wlse tawoktd waa r oNy, r Mnla 60% of aoasmoici.B eaoaoir aar M% of .oo-taoby ta.r i. H.rbia t.qor.d tW tlr tpo.r siwr Nook.d, aottprt.d to 32% of aoaeorv.{ onaroi aad 32% o[ so.4aoio.B AIt interviews ..n ooadttol.d i. Iri3-i7. At tb dor al aw neTUitmeot, 1.049 elipbie patida bed bees idetisad by the Harbis and Sheayang aowr rnpauia KmNtalt4nd and tiay-four (91.fX.) were intrevi.w.d. 32 (3.I%) d(.d before our att.mpr.d coetaa, 30 (6.1%) wn not 6osad and three (0.3%) rto p.rtidO.u. Forty-two per aat (a - r0S) of tM ar .wtt di.por0 by ti.u, bioper. 32% (a - 309) by c7no(op. aad 26% (n - 351) by radiology. AhhoaBlt the peetwanOra of pabol- iolb and cTtobpolly ma0rmnd aant wn hidar ia Sheoyang than ie btarbis, the at1-type dit+tribuoom wn similar. In the combined a of aatat Nsim wts 44% Tdr I r,evarts et taoto.R by sq.r aV Peeip at me+.rRoer dial; raiuw rinb far Lq e.et taooarl .af taok'ta, C.. C.r.ar AP fpesl • 96 aw.Y.e • 96 awtlre BR I!]% Cf)' <!M 30~ 34 1lU 3a 1.1 (1.0.1i) l0-34 7t13 ti0 196 3! Z7 (1.L 1.0) 33-5! 232 62 141 43 20 (1.s.3A) M-M IiM l;• 191 70 3.2 (i t. !.0) •3 + 137 N 161 ei 22 (1.4.3.3)

N4 A.H. WL:WILLtAMS rr a!. Tail. Il RR aad 95% C1 for lung eanoer auoeuud .itb iotmrty of ®okin4 by =l1 type CtU q+pt Cirarrrn per dQ.r Alllnn4nnoer 1-19 ;9 20 Squamous ou oill 1-19 ~1 20 Adenocarcinoma 1-19 ;0 20 Drwnt oJswabq lrerrl 1-29 30-l9 >40 113'b1.0, 1.7p f11i, 123N 2:6(1.9.3.3) 4,146/13) 3,2 (2.4, 4-3) (I17i103) 1.9 (09, 3:6): (I9r14) 3.3 (1.i.6.2)' (33J1S) 3.7 (2.9. 11.5) (36lI I) 2.0/1.3. 2.9) (49112.5) 39(21. S.9) (S6/t3), 4-7 (3• 1, 7.1) 1647103) 2-0 (0.7, 54) (6,14) 3:6 11.7. t.1) (10/13) , 120 (5.3. 27.0) (17111) . 08(0.S.1.3) (30~125) 1.7 (1.1. 2.5) (37,13), 2:0(1.3.3s0) (43l103) 01 (0.3. 2.6) 14,16) 3/ (i f, t.0) (ISiIS), 2.8 11.0.74) (7ill). '95u% confiderns intervals. 'Numbm of wes,controH am ut pueathaa. Taiie 111 RR for lung cancer aaocrated .ntR pauwve smoke T.Ne IV Relauve rrtsk of lung unm asaoeiatad with yan of uae of exposurs spanfic lwung devraes 411 nbjeru yow-opmrrr wJy .SOIIre.Of /IC[f7N rdU1 ve pAlptre Calrr (:6rN sswke eapwr rtponre coRrrols ItR trownob R!1' Any aohrtirtant w 112 111 74.117 yes 1" :$42 0.8 343,515 0.7 Spome no 398402 212:'271 ya !S! 351 ' 0.9 .05 331 0.7' Mother no 543 595 :91r410 ' yes 413 353 I.0 1119,192 0.9 Father no 41341,513 233-352 ya 472433 1.0 IE2;2S0 1,1 Wortplia no 403 sa1 187 301 ya S63•SI3 1.2 " .301 1.1 'id3ustsd for a3e. eduauon. personal smokrn6 and study area "Adjustad'for ap. eduauoa. and s+udr ua.'T<0.03. cases in Shenyang. There were no signi5caat ttenda in rsak with intensity (i.e. number of cigarettes smoked by faaily memben) and duratioa of exposure (i.e. years of smoking by cohabitanu): saotpt for an iacrsada, riek •arodatd .nt1P mammas mumry af fatler's ttmokini in the pnuat:e of cbe ~ ~ There wu a small atoas risk associated with pauive smoke exposure at the workplace. For all subjects, the smoking•adjustad RR was 1.2 (95% CI L0-1.4). The r.wlt was similar for nonamoken (RR 1.1; 95% CI 09-1.6). There were no tigutbcaat doae-reapome trends aaociat.d with years of pasuve smoke exposure at work. Kror6y atd cookinl procr(cet Tabk IV preaents RRs assooated with duration of ure of Kaag and other heaung devioa. Elevated risks wsrs ob>ternd for merasinj years of uwe of IKaag (particvLriy wbeo hrtad by stoYp utldelDeats), h0ittld brlCt wallt; or 1)o0ltti (4R beated by pipes le.d'ing from the satow to the wall or Aoork coal stova and coal burners. On the other hand, d.ez..rd risks were obeerved for ina.uinj yeans of uer of no.-co.1- burninj uovu and aatral batini. The patterne wae pm.r- a0y :aimilar for smokers aad noataDokas, and for aquamoal. oat a9 cardaomas and adeaocarciooma. We also esamiaed the risks uwooat.d with yesn whee coal wood. and aeetrtal 6eatinS aeev.d as the maia fsrJ for baatinf The Us Irad.d to rise with ieeree:e~ aee of oo.l aad datior with iecreasie0 we of wood aad aentralhatinz , but oooe of the ttmda wae ZiPiBeaot. Casia mors oRess reportd that tLdr boma Eecator staoky dtuW oooitng aod that they more fr.qveatly hrd ialitated eya dnria9 cookinl (Tabie V). Then abo was a asnisnet aeod in risk with ioeraaetg mrmba of malt cooked Iry deep feyia=, altbo~ this method of aookiaR .ro not frr queatly used. The reulti were aimiiv for itqnamo0e/oat ae8 caaaer+ aad adenootanotna. and for amoketn aad eoa tmotan. OcCNP+dat Subj.cu wen asked about all jobs in which they, had wortad I mc ma. yean, with wa atd controb compnd m tamr f.:poxve , rrar» Caurcaerrots RR/9J% CI) . Kang 0 25/40 1.0 1-39 3i4s376 1.4 (0./. 2:4) . 40-49 1321144 1 1 (0.6. 2,E) SO + 4131393 1.6 (0.9. 2:6) . Sununa Kanp 0 677?40 1.0 1-20 106/91 1.2 (0.9. ,1.7) 21 + I7I/122 1.1 ( 1. 1. 2.0) Cod stora 0-20 192/226 1.0 21-40 Slti4iS 1.2 {1.0: 1.6) 41+ 253/242 1.311.0. 1.7) Nooaal asoMa 0 212l113 I.0 1-20 3671340 0.1 (0.6. 1.1) 21-30 2!91295 0 7 (0.5. 0.9) 31 + 11//I3s 0.1 (0.3. 1.1) Haud br+ek walW(loors 0 516,6SI r.0 1-20 12719/ 1.5 {1'. I. 2:1) 21 + 2.3f204 1.4 (1.1. 119) Coal bmnes 0 323/S13 1.0 1-20 23t/202 1.2 (1.0. 1.6) 21 + 173068 1.1 (0a. 114) Gottal h.a 0 602/37J f.0 1-20 213/200 1V (0.t. 1.3) 21 + 1»r110 0.8 (0.6. 1.0) 'Adjua.d foa ye, iducsuoa pesaoe.l nnokm4 and study anL TaM6V Rdative rvk of hm4 eamr a..oca.rad with fraqomsy oCdop feyin4 tud eye nr+tauo. wlr oookse0 C.rs/e.ur.4 /tR' rro396 Cr) D.ep try (psn p...oaA) 0 32Me03 1.0 1 32H360 1.2 (1.0. 13) 2 170I107 2:I (1.3, 2.8) 3 + 121f41 1.9 (1.4, 2.7) Ey' irrftiao. tww/enMy N7j,1J2 1.0 teerta~ 21i1103 1.6 (1.2, 1.1h (ieqirt 19IS6 1.1(1.3. 26) 'Adjwra fa ask .dooooa peno..l amotiq aad stady ua of tJxv smp~ioymem ii 29 job otqoeia Xioat (77%) ..omea held at kret atae job outade the bom., but t;illp. 6nntly iea.ard roks wen obarv.d ody for tewl stodtieg .rork (RR 1.3; 93% Q1.0-2.1), .hiie a aRedkSetly decreased risk wr obrrv.d for te;mY worka. (RR 0.6; 95% C1 0.3-1.0). Tb wromes wie alw aakad if tbsy weQa expoe.d to 12 ~edse dvua, t~ots or 11>mr at wock. with 6om I to 16% r.porti" oe.th.job tTowtne to the 12 poDutioa itema. Caam reported esposi:re to coal dsrt (RR 1.3; 9S% Ci 1'.1-20) aad to amoka frao btreatK hrl (RR 1.6; 9S% CI 1.2-2.2) tssni&aetyr mon oflia

Prior hutt disease Table VI lisu RRs of lung cancer assodated with spertfic prior chronic l4n8 diseases. Lung diseases that wtre first dsa8nosed within three years of lung canca dia8nosts (and a comparable time period for controlst were excluded from the analys+s. After adjusting for smokin8. history of any pnor lung disease was associated with a 30'i% increased risk (93•/. Cl 1.:-1.8). The excess was greatest for pneumonia (AA :.I1. An increased risk was found for bronchitis and,or emphysema. but the association was limited to squamous,oat rrll cancers ( R R 1.6); and not found for adenoarctnoma ! 3tR 0.9). Wt investigated whether risk of lung cancer varied aecord. tn8 to the ta8 ttme following the diagnosis of prior lung disease. Earlier detection of chronic bronchtus,emphysema conveyed greater risk. Relative to those with no history of chronic bronctiitis(emphysenu, the RRs were 1.3. 1.3. .nd 1.7 respectively for conditions detected 4-10. 1l'-20. and :1 + years before lung cancer diagnosis. On the other haod" the RRs were hi`her for moro recent diagnoses of pneumonia and TB. The RRs were 2.7. 2.5 and 1.8 respectively for pnetunonia. and 2.8. 1. t. and 1.2 for TB first detected 4-10; 11-20 and 21 + years pnor to lung cancer diYBnosis. The elevated risk associated with TB du8rtosed 4-10 years prior to lung cancer was si8ntficant: it was observed for both squamous: oat cell cancers and' adenocarcinoma of the lung. and among non-smokers as weU as smokers. fanil.r hLstors• of TB and cmtccr We observed a significant 60•i. (95% Cl L2-2.1) increased risk associated with TB in a household member. with similar risks for squamotwoat cell' arttxr3 and ademoaranoma. The familial associatien was seat in smotets and twa- smokera, and rematned unchanged after adjusting for petsonal history of TB. The risk assoctated with family his- tory of TB increased with decreasing age when the index subject was first exposed. After adjusting for smokins, exposures at age < 21. 21-30 and > 30 conferred risks of 1.7, 1.3 and 1.2 when compared to those with no household TB exposure. Famdy history of lung cancer in Mt deQer relatives. reponed' by 4.Y/% of the ases. was associated with a signi- ficant 80•/. (93•/. CI 1.1-3.0) incrsased' risk. There was little diRerence in risk by cell type or smoking status. T6s risk of lung cancer was somewhat higher among those with a family history of other cancers (RR 1.4; 95% Ct 1.0-2:0), with the excess risk bein8 higher for sdenoarcinoma lRR 1.8) than for squamous(oat cell cancers (RR l.l). .Mrnurstd and rtpr+aAtqtie forran Table YIl presents risks of lun8 cancer by variouf menstrual and reproductive factorLL Thers wers little or no asaoeiuion with a,e at menarchs. parity. hysterectomy. tt;fontsneow aboction, prepnaocy rR;wJtiIIt ' tn dimcult labottr, and use of oral' contrauptivet;. Ttiae was a sipsi5nat !0'A (95% CU 1.2-1.8) increased risk associated with history of misarriaele.. and am tended to have a tater ap at tratural meaoplur although the trend was tae smooth. LL'NG CANCER AMONG WOMEN IN CHINA 1rS TaW VIF Relauvc rtrkks of lutt8 onarasaoctat.d ntk exmtrtu) ard reproductive facton Cawrer"rro4 R/C 91% Cli 4P at mertarche 18+ 1!4 192 1.0 16-17 427412 1.1(02. 14) i4-13 2115,276 1.1 (0,1. 1.4) <14 5364 09(06.1s) N(ttttber of ehtldeen < 3 193 :03 1-0 3'-4 319 300 1,1 (09. 1'3) 5-6 :73:72 10011. . 1.1c 7t 169 174 to(07.1.3) AV at natural metwpa,us <43 n 112 1 0 45-49 373 303 1.7/ t._' 2.41 30-54 273 327 111 10.9. i tl 1S+ 31 22 t.7t1.0.3.21 Ponttve history of Hystenetomy %% 1010.6. 1.61, MiscaTrt+r $2,126 1,50.2. 1_81 Spotrt. altvrtwn _39-21t I 1(09. 14) QiAkuh taEiour 76 61 1 3(0 9:, 1 .91 Oral 0ontraeepuve Ss 68 0 t(0.3. 1.21 'Adlutt.d for ap. afuotbon. personal~unokkn8 and study arsa., Dirnnn Jactos The diet of the subjects was dominated by staple grains (median intake among controls - 1.095 times per yari, fresh ve,8etables 1i;188 times per year), fermented salted foods 1730 naxs per year), and soya ban products 1365 times per year). Less frequent was consumption of antmall protein sources t231 times per year), fresh fruits (52 times per year). and peas and btsns (12 times per year). Risks of 1urt8 cancer in relation to dii;tary intake an shown in Table YJIf. Higher frequencies of intake o(vesetab(es, either those rich or low in carotene content were not significantly protective apinst lung cancer. The three foods with the highest carotene con- tent in this study population were dried hot red peppers (16.9 m8 of' carotene per100 S). dark lesfy Sreens (2.7 tn= of carotene per 1008), and carrots (2.0m8 of carotene per 100 8)'. Carrots and dried hot red peppen were consumed less often by ases compared to controls, but thess ttema ,(tere na fnequently consumed (mean tntakr among controls was 41.4 and 70.0 tuoes per year re:partiveiy). On tht other handt aaa had slightly higher intakes of the tnorn oommody conswoed dark leafy Sretea (avenp intaka among controls was 163.5 tima per yearl. Caaa reported higher fr.atanscia of intake of animal pro- teie and' fresh fruits. Few women (12`/% car vetsut 8'/% controls) drsak alcohol nsore than once a ywr., but they showed a sipnfiant taookini-adjuuad 30% increased risk of lung aacsr comp.r.d to those who did not drink at alL However. tbeR was oo clear trend with increasin8 abobo( cotnumption. Thers were no appr.aable dsfferenas in dia- tary pattsrns for sqnamouuo.t odl aaam vwsttA atlsaoear- coom., twr for snwkss nastu ooe-smokan. TaYr 17 Rehaivs rtsk for tttttg easrtw aradatrd witl pr..iw li.y d...... rortive butory of: chroete broeelu or aud/or aspiyt+wa pKymom. tubacvlori. AY Ar4 Sprtw/arr Aiiw.o.cw.w.. Caesl' RA' nwve4 /9!% Cl/ tr RR )y' RR' 2101137 1.4 (1 .Z 1.8) 79 t.r 46 0.1 66/2a t1' (1'.3. 3.3) 23 2T IS 1:6 103/83 1.3 (0.9. ).7) 33 1.2 33 1.1 •Adjiawd for aw .dtttsoos, pswaal.eoku8 ud rrdy awaNr.bsr of ca...fds factor. 93% coa66rane iatw.ab eachtdM 1.0.

I 00 ooc o00 --= o00 000 00o e-- oo-- oraoa-oo. aoooe.no...vaaoo+o.-oaoe.•+onon

LLNG CANCER AMONG WOMEN IN CHINA 11167 affected firstde4m famtiy members was smnll (Ni.): Recent case-control studies in Great Bntatn(Ayesb er af-, 19g;{) and the United States (Caporaso er ol_, 19E9), however, suggiest that genetic traiu may mfluena susceptibility in a sizeabk portion of cases. These tnvesttptions revnkd sipnifKSndy increased risks of lung cancer associated wtth the genetiaUy controlled ability to extensively metabolise the drug debnso= qume. a trait aRectml St•i% of the control population studied in the United' States. We found no strong support for a roie of boneoeal facLors for lung cancer overall or spenfsa)ly tor adenoearpnotna. The cases did tend to experience menop.use at later aEes, but the trend in risk with age at menopause was not smooth. History of prolonged labour or hysterectomy, which had been suspected as risk factors for adenoarcinoma Eerause of the potential for trauma-assoctated lung embolism, occurred more frequently among our ases, but the excess risks were tsot sitittsfiant since relattvely few women were affected. Risk of lung cancer was recently reported to be increased amonl Chinese women with short menstrual cycle length (Gao er d., 1988): but thts variable was not assesxst in the cttrrent sttdy. In other countries the risk of lung anoer is generally reduced among those with higher dietary intake of carotenoids IiZteEler, 1989). but our findings are !ess clear. Cases had slightly higher rather than lower intake of dark green leafy ve8etabks. the most commonly consumed rich source of carotene. Moreover, in our analysis using a com- bined index of all vegetables rich in arotene, high frequen- R.(ir..es AYESH: R.. IDLE. J.. RITCHIE. J.C.. CROTHERS.,MJi & N'E7ZEL M.R. (1981): Metabolic oxidation phenotypes as twrken for susae(. nbdity to lung ancer. ,VOrur,r- 312. 169. tIRFSCOW: N E a DAY. '4.E (19EO). Srauriraf MerAodr in Corrr Rrseorch: dir AAalvJir of Casr-Coerrol Srr&rt. IARC: Lyon. CAPORASO. N.E.. FALK: R.T.- ISSAQ: HJ. a! others 119d9): Ltuy cancer risk. occupational exposure, deErisoquute metat+olic phenotype. Cancer Rn.. 19. 3675. CHAN. W C.. COLEUORNE. M.l.. FUNG. S.C. a NO. H.C. (1979): Bronchial cancer in Hong Kong 1976-1977. dr. J. Corw. 39. 182. COHEN. R.l.. DIAMOND. E.L.. GRAVES. C.G.,a 6 others 61977)_ A common familial 'component in lung cancer and ehrome obstrYo- tive pulmonary dtsela. lerrn. 4 523. ERSHOW. AG a CHEN, W K. (1990): Clun.e food eompotroon tahtes: a translation with EnElitlh common nuna. Lann faamAc names, and Pinyin romanuRd tr.mtiuntiom, i.+I C..y. Aw.L (ie the ptsa). GAO. Y.T.. BLOT. wJ.. zHE»a.,w. a 5 ot4en 119E7). Lry cancer among Chinme women. llu. J. Cwrer. M tiGt. GAO. Y T.. nLOT. vb.J:. ZHENG. W.. FRAUMEM, l.F. a NSU. C.W. (1M). Lung canctQ and smoking i. Si1yl.i. ir- J. Frieli.rrL, 17. 277. HINDS. M w,=TEMMERMANN. G.N.. YANG. N.Y. a 3 olt+as (1981). LrRerenrss ia lung eaaatr tfios t+sotiq awooE l.psar. Chinese snd Hawauae woenn in Nwa'r- br.,J. Cwnr, !7. 297. KUNG. 1. SO: K. t LAM. T. (19EY). Ltay omr rR RartE KoeE Chinae: aroeta6ty tud' hrtolope typn 1973-19ti2 i. /. Csrvr. A 3E t. lAw. C.H.. DAY. N.1L a lNAN1pJGAtaATNAM. R. (1974 IROdras ntes of IpncAe hutdbOcst typ, of hW oeos i. SiafPo^ Chinese dialst pvups, and t6rr astioblled sipAotta- Ast. J. Ce+rc.F. 17. w.. LURIN. J.N. a ELOT. Wi. (19i4). AariRamit of htp aror risk factors by histolbpe eatqory. !. Ned Cwes ttur., T& 3R3. Mt1MFORD. J'.L.. H'E X.T- CHATMAN. RS a 9 MMf (19E7). LttRE tamer and' indoor .u potluoora i. Xa.. M., ClIews. Seiwnr. 276, 217. cia of intake did not confer a stintbcant protective efect. Rawn>< for the absena of protes'uve eaects art not clear. A posstbk explanation is that three-fourths of the study population ate vetaabla hijh in carotene content at least twice a day so that the nearly tJniformly high intake of carotene<ontaintng foods limited variability and hindered detection of an effect. Data on plasrna carotene levels from this study population will be important as a more ob)ecstve measure of thar dietary, intake. Misclaswfiation of intake also may have dampened trtmds. We did not have inform- ation on portion size and the highest arotenecontauunE food in this population is dried hot red' peppen. usually used as a condiment. In addiuonL recall of pau diet may have been influenced by recent dietary improvemenu. perhaps more so among cases who may have been given pteeferenual dietary trsatment because of their tllness.. la stunmary. thiu invauption rttvaied thu contrary t a prion expatatioR iR Chinam aprette smokinj is the ma~or pme of IwtB asttxr among +romes ie sortAtast CAiba aud f 1e+omnbuta to the ttna't hyb ntss of mortality from this riesor. PrsventaR activities should emphaxse smokinE csssa- tion. while additional study may help clanfy the role of indoor and outdoor air pollutton: chronic non-mah`nant lung diaase- occupational exposures, familial susceptibility and other factors in the atuoloEy of lung catxer.. We thank Joan Howiand' for preparation of the nunu.cnpu NATH)NAL CANCER CtNrTROL OFFKE 11911d1. YayinE lwutw» e/ CieotraFhy ANm of C.wrw Mon.liry n rAr hoRk r ReFrb/r oJ CAa... Cstni Map hsr: inpnE. 001. wL.. ELSTON. R.C.. CHEN. vw. EAtLEY-wILSON: J.E. a ROTHtCHILD. H(19ti6). Increased fasalial risk for lung cfnar. J: Nmr Cwrw bvr.- 76. :17. OOr. w L.. ELSTON. R C.. CHEN. v W. aAJLEY-wt13ON: J E, a ROTNSC1/1LD. H(19E6) Fanahal IitnE ranosxeorr.cnnE an error in otkrolatroe. J Xal Cmrn hta.. 77, 990. QU. Y H. XU. G:X.. HUANG: F. FANG. J C'_a GAO. Y T 119E6). An Anr teu on otLe bf-produas o( tw IrtusE of cooking otis. Tww. 6. 59. SKILLRUD. D M:. OFFORD. K F a MILLJSR. R.D. (19E6): HiaAv riak of lung cancer in cAroac obwtnatt.e pubtson.ry duas: s pr¢ sp.cuve matched controlled urQy. Am. huow. Afef.. N6, 503. SURGEON GENERAL (19{6). Tlrr NndlA Crurilrrwrra oJ l..ohw- rov Sn.ek'.K. Dap.rarnu of H.eJeh tud Hrne. Servtae (CCIC). PubhSnoa Muwtstr E7-3E9E. Go.rrmeest hsnunE Oms: w..tti.poa D.C. MENG, 7LZ, HONG. LO. a CHEN, DY. (19t7). Smoking prs'rtaMea iR Clire ap/ 13 and .lsw- CAa AAt J., IM, Bi. tvU: A.tl_ YU, M.C- TNOMAL D.C_ M1SL M:C. a NENDERlON. R.L (19E1}. Flrsoaal a.d faanly, hiNOry of luy mtrr as nsk factors for sd++oeatdaoma of t!e hseR Cwrn R.s.. r, 72T9. xUlO. N. a xU: LY. (19Ef). Air FoYrtioW and ItaeR a.er in Liro.- inE Rtvvittae, Ropk's Repyhlir of Mae. MCl AFwsg.., M. 53, XU. LY.- SLOT. MJ.. XIAO. H.P. a 7 otbrn (19tM9). Lsokuy, rbr poRution ad ttr Riok nnm af lty o.tor r lM.ya.R, Claies. I: Nol Cwns JRw., Ei. IE00. DItNG, M.. Et,OT. wJ, tJAO, M.L a! oeMrs (t9{7). Ltog a.ar and p>Ier twYa~xtloer i.fieoon i. Ji.qf.i: & . J: Cwru.. !L. 501. a1tcLJla. RA. (HN). A t.vi.ar ef spih.ioio.e aid.oe that orole.oie. radtto tr t;tk of o..e.. J Jvrtr., nlt. lJK.