Tobacco Documents Online

of the Board' of Regents and its Advisory Committ.ee on Research and Deveiop- ment> we thank these individuali for their contributions not only to the R&D Program but also to the continued a&znt emenc of knowiedge in this state. Indlla Kilc:eate, PAD, Derscsor Ra.mrh and Det's1oP"uw Aag+ louuicna Board of Regertu The Board of Directors and saff of the Cancer Assodaaon of Lo+=s=RaInc. and'the Cancer Assodation of Greater New Qrleans, Inc, a United Way agency, are very proud to have been involved' in the planning and coordination of the Intrrnational Lung Cancer Update Conference held March Sb, 196l. Many ofthe papers presented in the monograph reviewed the smoking habits and' the epidemiologic trends in lung cancer incidence and mortaliry in the United' States, Europe. and Japan. TZtey all repeatedly emphasized the impoT- tance of dgareae smoking as the major causative factor in lung cancer. Environmental hazards (eg, air pollution and asbestas) and host faaoes (eg. geneacs and nuaition) play a small role in the overall etiology of lung cancer. The most important conclusion of the Interstational Lung Cancer Update Conference is that an intrrnational emphasis should be placed on smoking cessation programs aimed not only at high-risk adult populations but more importantly at aU adolexerna The only rational approach is to prevent lung cancer by graing individuah to either sop smoking or never to start to smoke dgaretses. RoMt G WeilaaaJrer., MD, Preident fawuf A'ssocianon of Greoler .ti'ety Orleant, Inc f:araes Auociation of Loui:iara, Inc With the high inddence of lung cancer in Louisiana, it was very appropriate forNew Orleans to be seleaed' as the host city for the International LungC.ancer Update Conference. The Cancer Association of Louisiana and the Cancer Association of GtraterNew Orleans are glad to cosponsor a conference that brings together some of the world's lung cancer experts • Personally, I have appreciated the opportunity to be involved' in a program that could help resolve some of the health problems of Louisiana's citizens. Ruth A SAnxaood Era+cvtiw Di+.csor Caraer Avooatioa oJCns,ater 1Veto Ort.anr, Inc Canar AssodQabR of Louiriasta, Inc

I C HAP7ER 7 The Causes of Lung Cancer In Texas PATRICM A BUFRER.` UNDA 1NlLlJAMS PIME•• 1HOMW .l: AM.SON:" ond CHARLES COllfANT`• ~ Epidemiolosy Rmeuch Utut. The Univenity of Teaa. Health Science Center at ]iouaon. Sehod of Publir Heilth. Hou.toe, Tesa. 77025 •'Environreentai Epidemio{ogy Eranch. tHationa1' Cancer Inuitute, Lndow 3C]S. Benhesda. Maryland 20205 ASSMCT A popularion•ba.ed case-comparison inte++riew study of lung cancer was conducted from 1979 to 1982' in six Texat coastal countiea-Orange, JeSerson, Chambers, Fiuru, Ga1- veuon, and Brasoria-to evaluate the asxxiation of lung cancer with ocrupuional'and other en.ironmental exposures. Lung cancer mortaliry rates in these counties consistentlv have ex- ceeded lung cancer mortality rates o`»erved for Texaa and the United States from 1950-1969 to 1970-1975 for both sexes and races (white and nonwhites). Fiistoibgically and tytoiogicslly confirmed incident cases diagnosed during the interval July 1976 rtto June 1980 among white male and female residents aged 30-79 years were a.eer- tained from partiaps<ing hospitals in the six-counry area. Both popul*tion-based and de- cedent eompuiaons were selected and matched on age, race, sex, region of residence, and vital aatu>t at time of aacertainment. The exposures of pnimary interest in the study of lung cancer are those auociated with oc- cuparion (employment in specific industries and ocrupation>t):in conjunction with tobacco, almaol„diet., and residencial exposures. Rey Wo.dr: Smoking history, penocbemical iaduatr7, &iatologic types, constrstetion wrken, chemical manufacturing, tnasporsation Introduction and Background Data presented by DoU and Peto (1) and related reports (2) indicate that respiratory cancer sites„ dotninated by lung cancer, show the most dramatic in- F creaus of all cancer sites over the patt 30 years~Tlte~role of:moking in the etiology ,~, pf respiratory cancer has been well docuaunied LIn addition, lung caneer is L qt verbp C'+.*r ~+wrvhv+a[ r+e t+n0 CCnesr Cass on0 PVM+w+ 83 N0T1CE Thts matetial may W protected by cOCyright in (TttA 17 U.S. Code).

Pafrido ~l 9uMSW. LnOc VWior+u Rlt:klr. thornos J. Moson sr cl recognized as possibly the most important work-rrlated' cancer. However, the in- teraction between smoking and octvparional exposures and the increased risk that may be attributed to an occupational exposure has not been very.vell~rharacterssed for a largr number of woricplace exposures. A population-based case-comparison interview study of lung cancer, obtaiaing detailed occupabonal histories, was conducted in six Texas coastal counties where ltmg cancer mortality rates were elevated (3). Figure 1 shows the location of the counties of Orange, Jefferoon, Chambus, Brazoria, Ga]Leaton, and Harris, a highly industrialittd area where Houston is located. Approximuely 25 %(3.5 million) of the total stsce populadon in 1980 resided in this southeastern coastal area, the majority (77. S 96 ) in Harris County.. Newfy diagnosed, histologically confirmed cases of lung cancer in white females (induding Hispanic) were aaceruined frotn July 1977 through June 1980 in Harris County (3 years)~and from July 1976 through June 1980 for the surrounding five counties. Similariv, cases among white males (including Hispanic) were ascer- tained for four years (July 1976 through June 1980) for the five less urban but in- dustrialized counties, excluding Harris County. Background lung cancer mortality rates for white males and' females were examined by Texas State Economic Area

Tn. Cms.m oa Lurlp Ccsnc.r n Trxos gs. ~ Top 10N.1Siqnof >U.S. ® Low 10w.rSlpnd < U S. Not ToC 10'h/Sipntt >U.S. ~ Not Low 10Y.1Spnt1 < UIS. O Not &qnificantly DiftfrMt Frpn U.S. Fi~urc 2. Lung cancer mortality, 1970-1975 for white mala. (SEA) for the time period (1970 to I975!) irnmedisteJy preceeding the case- comparison study. As shown in Figures 2'and 3, these maps consisuntly document the significantly higher lung cancer mortality rates observed earlier for both white males and white females in these Texas coastal counties. The dark areas along the upper Texas coast are the Beaumont SEA (Orange and JefTerx>n counties); the Houston SEA (Harris County), and the C`,alveston SEA (Ga1.•eston County). Age- adjusted mortality rues (adjusted to the 1960 United States population) in these areas are in the top 10% of rates for SEM in the United Stues and att signifiwntly higher thaa the white male or white female lung cancer mortality raee for the total United States population. For white females in Harris County, this excess was notahJe for both the tate and the tread' in the rsze from 1950 to 1975 (4). For a11 ages, combined, the overall exeas ia lung cancer saorsality in the Texas study ares is approximately 30-40 46 , but this is considerahly greater for some age groups. Occupational and industrial exposures of iuaporunce for residents of the Texas coastal area include those associated with shipbuilding and repair, chesaical and 'Fxciu"g deuAs for 1,M.

86 PCfrlCio A BUPRw. Lfn00 Wfiarns PICIW. ThornCS J' MCSOn *t at  Top 10N.15iqmt >U:S. ® {.ow 10Y.1Spnrt s U:S. 0 NOt Top 1'0'h'lSqnrnf -U.S. a ' Not Low T0'/.1$iynrt <U.S. 0 NCt Sqniticantfy Dittorent From U.S. 1<'>;uec 3. Lung c.ancer mortaliry 1970-1975 for white females. pec:.hemical manufacturing, petmleum refining, conatrucrion, and metal in- duatries. The largest T:nited States based chemical and synthetic rubber production fasilitiea art located in the study area, so a high proportion of the working popula- tion currently is employed or -hu been employed in dsese industries. For some of the smaller counties, such as Orange andjefferson, where a single industry is dominant, as high as 2796 of the working population reported~ being currently employed in chemical and allied products manufacturing compared with 2% for Harris County (5).. Methods Histoiogically confirmed incident caaes of lung cancer diagnosed among white male and female residents (including Hispanic) of the study counties for the designated tirne intervals ( Juiy 1977 througfi June 1980 for females in Harris County and July 1976 through June 1980 for rnales and females in other counties) were ascertained by review of hospital and state records. Hospitals in the study area that were not already participating in the Statewide Cancer Reporting Program

Tt,. Couss ot Lunp Conc.r n T.xcs 87 ,,erc contacted and'asked to partitspate in the study. Populuion-baxd and dece- dent comparibon subjects were sdected from state and federallreccrds and matched to naes on age, race, sex, vital status at time of aacertainrnent, and county of taidence (Harris County or other five counties). Hispanic study subjects were identified systematically by use of an algorithm to identify Spanish surname. Jv[edical~ records were abstracted by state-trained abatracton to obtain relevant disease and demographic data. Following contact with the family physician (for caser only), personal interviews were conducted with study, subjects or with the next of kin of decedent caaea and comparison subjects, using established criteria for selecting the most appropriate next of kin respondents. Interviews were conducted by trained interviewers in the bome using a standardized interview prococol. Detailed inform:tion regarding the primary exposures of interest was collected, specifically smoking history, work history,, residential history, and drinliing history. Industries of employment were coded to the Standard Industrial Classification (SIC) (6) and occupations were coded, using the Diettcnory of AuYpationa! Tit/v (7). The Mantel-Haensze] summary rhi-square and odds ratio ttatistia were caiculated (8). Confidence intervals (9596) were calcvlated using the metfiod~ of Miettinem (9). Results A total of 56 of'the 67 hospitals in the six-counry Texas study participated in the study, including all of the seven large hospitals (300 or more beds). Ten of the 1I mmaller hospitals that did not participate were located in Harris County. Tbertfore we were able to ascetvirt 92.2% (1520'caaes);of the total 1649 incident white male and female lung cancer caes (including Hispanic) estimated for the 3- to 4-year in~ ternl (mid-1976 or 1977 to taid-1980). The number of incident cases was esti, mued by adjusting age-race-sex-county mortality rates by population growth and an incidence: mortality ratio of I.3S:1.0: Case ascertainment was higher for resi- dents of counties other tlian Harrie County, 97.296 vs 82.196 (Table 1). A toal'of 766 female and 754 mak cases were ascertained representing, respectively, 88.7 and 96.1 % of the total estimated incident cases ascertained Hispanic females ap- peu to be pootay ascertairted~(38.196)„ but this may be related to the clavification based on Spanish surname which may not be an effective technique for ascertain- ing married Hispanic females. All ascertained eaaes will be used for determining age-race-sex and county lung cancer iacidence rates for the study area. A total of 88.9% of the ascertained cases were included in the interview study. Some cases (110, or 7.2 °J6 ) lacked histologic or cytologic confirmation of husg cancer and were ineligible for the case- comparison study. For the majoriry.of tbese cases (79, or 71.8%) the basis of the lung cancer diagnosis was radiologic or clinical evidence. There was inttufficiertt diagnostic information ava0abk on the remaining 3t cases. Additional lotssa of study subjects in the casr-comparison study were related to race and residential eligibil;ty tsiteria; unable to locate; moved out of interview area; physician,

88 potrtitq A. eum.r. Ur,do wftrnr aticlcl.. Thomc.+ J naoson .t a Table 1. Lung cancer case axertainmene in Texas study , by mx, ethnic group, and ara, 1976-1980 NMbv 1'taafbv :Voaber .ri.ae.d' .eeasas.d` ( S. ) caees inter.i.+.d~ White Femake Angio E22 750 (91.2) M9 Spani.b suuname 42 16 (3l.1) 11 Total ~{ 766 (8a.7) 460 W6ite Males Aado 767 730 (95.2), 460 SpeaiaA aurname 1! 24 (133.3) 15 Total 785 754 (96.1) 475 Area Hu*i. County 567 468 (t2.1) 275 (kmaJfa trnlj+. 1977.1960) Other counties 1032 1052 (97,2), 660 Total 1649 1520 (92:2) 935 t alnefude t 20 e.aas..atAout hmoioqr c onfirnuuon aaid e n add,uoftel I! cae c+tuemed te b e wni¢bk. mtatea ol ~r..acr~-•~udrendene: mterv •ude+.cast inefipbW n«locned. refua.Ubyphvaac b ue hoqna J!.or.rudr.uDyV . and caaenime+ne.adatd mubaquentlv 4rnttfied ar kndipbFk. a..dawto tieo!'pc or 9u.iny. .. Table 2. Texas lung cancer stud,v population by sex, study group, and ethnicity I 3-d7 i-P Caees Csatrol. T.tal. Total Female fli0 482 942 Male 475 166 941 Totai 935 948 18d3 SpaniaA surname Female 11 20 31 Male 15 19 34 Total 26 39 65 hospital, and subject refusals; and poor quality interview data. Overal) study sub- ject refusal rates were 7.7% and 10.7% for decedenl caxs and controis rrspec- tively, and 13.546 and 20.646 for living cases and controla, respectively. A total of 935 interviews was completed with eligible cases (460Amales and 475 males) and 948 interviews with frequency matched comparison subjects (Table 2): Included in these totals are 26 Spanish surname cases and 39 comparison subjects. Separate ana]yses are not ptzsented at this time for these study subjects. The average duration of time study subjects resided in the county of diagnosis or in the six-county srudy area is over 25 yean for all study groups. The majoriry of both male (86 %d ) and female (82 9fe ) ~ cases were decedent cases and were slightly older at time of diagnosis than the living nses (Tables 3 and'4), The distribution of age at diagnosis is compared for male and female study groups in Figure 4. A higher proportion of the female caxs was diagnosed ~ before age 60 (45:4 4b ) ~ tAan male cases diagnosed' before age 60'(34%). I ftJ

Tn* CoLss ot Lurq Conca h T.xos 09 Table 3. Number and percentage of male lung cancer cases by age at diagnosis and rype of respondent, Texas, 1976 to 19E0 T7Pe of respos,des+t Toal Sdt Nost of lia A8e .e Caoe Ce.vola C..a Coasals Car. Co.trels doPosi. (Yov.) No 96 No S No S No 9. No % No S 30-39 1 1.5 1 1.6 3 0.7 2 0.5 4: 01 3 0.6 40-49 5 7.5 7 10.9 26 6:9 34 8.5 33 7.0 41 8.8 50-59 23 34.3 22 34.4 102 25.0 98 24.4 125 26.2 120 25.7 60-69 31 47.0 ~ 23 35.9 165 40.4 164 40.8 196 41.3 18.7 40.2 70-79 • 7 101 11 17.2 110 27:0 104 25.9 117, 24.7 11.5 24.7 ~ 100!0 100.0 100.0 100:01 100.0 100.0 t Towll 67 64 408 402 475 466 Table 4. Number and percentage of femak lung cancer caaes by age ar diagnosis and type of respondent, Tecas, 1976 to 1980 T7Ps of r.opo.deat Taal 1 8df Nest of kia Ags s,t cam ConwL Coies Cosuoi. Ca.e. Coeaol. "8- (rean) No S No S _ No S No % No S No '1< 30-39 0 0:0 3 2.6 6 1.6 5 1.4 6 11 8 1.7 40-49 9 11.1 12' 10.3 40 10.6 110 13.7 49 10.6 62 12.9 50-59 36 44.4 55 47.4 118 31.11 104 28 4 154 33.5 159 33 0 60-69 24 29.6 34 29.3 153 40.4 135 36.9 177 38.5 169 35.1 70-79 • 12 14.8 12 10.3 62 16.4 72 19 7 74 16.1 84 17.4 100.0 100:0, 100.0 100.0 100.0: 1W0 Totalr 81 116 379 366 460 482 Proportions of male and female cam and compariaon subjeas using tobacco, cigamtes, alcohol, or who "ever lived' with household member who smoked regularly" are compared in Table 5: Ninety-seven perctat of the male nses and 9196 of the feraaTe casa reported,ever smoking cigarettes but a higher praportion of the female tltan male cases reported smoking cigarettes currently, 68% vs 5496. Proportions of heavy unokeri and use of alcohol (ever) were higher for cases tflaa comparison subjects and for males than females. An exvandy high proportion of both female cases and' comparison subjects report having lived with a household member who amoked regulariy, 93% vs 88%. Althouglt the patterns of risk difTered for males and females (Table 6), the odds tnrios for all srnoking variables were statinicaIlv signiGcant at the p•.05 level. Among males, ex-smokers had a risk higher than current smoken, whereas in females the risk was lower in ex-stnokers. The liighest odds ratio for fcmales was observed for current smokers, 7.9 vs 5.0 for ex-smokers. Odds ratios for the ac- I

90 PCfrIC~o A&1f'M1K: L'r+CiC WiQnns PICk*. ThpMCS 1 AAOsC, _-r• Xw~ au~ ua nr r.o 41.7 I ta 0+0. \ a0-3. .a.. sa.. .oe. rn.. ..,,rw _ ae. . o~.....r fignre 4. Age distribution (age u diagnosis) for male and female studv subjects. Te:as luns cancer study, 1976-1990. Clear columns. cases; shaded colurans, controls. curnulated lifetirne agarette dose, expressed as pack•yean, were higher for males in the low and moderate categories but associated with a similar gradient in both males and females. No difference in risk was associated with the use of filtered dgarettes for either males or females. The role of "patsive smoking" in contributing to risk of lung cancer was exartt- ined (Table 7). In this analysis the crude ior unadjusted) odds ratio are increased and signi6cantfor both males and females. 1.4 and 2.1, respectively. However, when the confounding effecn of individual'subject smoking was controlled' bv stratifying the Table S. Proportion of cases and controls reporting use of tobacco; cigarettes and alcohol by sex„Texaa lung cancer study; 1976-1980 kales remales Caan CesrreL Ca.er Garrol. Tobacco (ever) 0.99 0.90 0.91~ 0 59 Cifaresta (em) 0.97 0.80 0.91 059 Ciprettea (current) 0.54 0.47 0.68 038 Ciprettes (light) 0.08 0.10 0.05 0: 17 Ciptettes (heavy) 0.45 0.29 0.34 0.13 Akdsd (erer) 0.86 0.81 0.78 0:63 Lived with a smolKr 0.76 0.70. 0.93 0;88

TTr Caw.w of I:unp Corca r Te:os Table 6. Odds ratioi auociated with smoking variables 91 for malet and females, Texas tung cancer study, 1976-1980 Male Ie.ale Ever .moked 10.12' 6.89 Carrent trnoker 9.99' 7.89 E,.®oket 10.85 5.00 r,tk•reAr+ Low (t)-3s) 6:24 3.21 Mcderace (36-63) 9.39 7.96 HigL (b4 • Y 13.05 13.35 Fihend cigartnes Yes 9.39 7.11 No 10;23 6.06 Doth 12.27 7.09 'MD', oddt ntwo .pJcann r p <. .05 male and female study groups into smokers (ever) and nonsmokers (never)~andex- amining the adjusted odds ratios, there was no significant incrxax in risk a»ociated~ with passive smoking. In fact, the odds ratios for nonsmokers living with a regullv smoker were not increased for either malts or females, 0;52'and 0.78„respectivdy. However, odds ratios for smokers living with a trgular smoker were increaaed, al- though not significantly, 1'.28 and 1.80 for males and females. The overall odds ra- uos (adjusted) associated with passive smoking were only slightly increased and not signifitint for either males or females, 1.2 and 1.3, respectively. When the possibility ofa "passive smoking"efiea was examined among nonsmokers by number of years lived with a regular smoker, there was very little difference in risk for females who lived with a regular smoker for 0-32 years (Table 8). Tbe odds ratios for rrtales sug- gest an increax by are based on smaller numbers tfiart the analysis in females. Table 7. Odds ratios for pssive smoking (household member smoked regulariy)iin Texas male and fena)e lung cancer studies, 1976-1980 Yes N. 0" 9s !. Ca.e Contrd Ca.e Coavel eatio Coafideace iistererall xe Males Crude 363 329 93 119 1.41' 1.04, 1.92 4.8 Sdfever tmoked No 5 56 6 34 0.32 0.15, 1.74 1.2 Ye 357 273 87 . tS 1.26 0.91, 1.79 2.0 Overall (MOR) 1.20 0.87, 1'.63 1.18 Females Crude 429 425 24 51 2.12' 1.29, 3.50 9.05 Self ever naoked No 33 164 8 32 0.78 0.34, 1.81 0.3 Y" 396 260 16 19 1.90 0.92; 3.58 3.0 Orerall (MOR) 1.30 0.78, 2.18 1.0 . p < .03. I