Tobacco Documents Online

/-, t/ l~s PJASSIVE SMOKING AND LUNG CANCER ELIZABETH T.H. FONTHAM, DrPH; PELAYO CORREA, MD; VIVIEN W. CHEN, PhD c < C . 132 JOURNAL VOL 13: al'RiL J La State "ted Soc 145(4):132-6, 1993

The findings of a multicenter national study of lung cancer in lifetime-never smokers are reported. An increased lung cancer risk of approximately 30% was found among women whose husbands smoked. Elevated risks of 40% to 60% were also found associated with workplace exposures and other exposures outside the home. The findings of this study have been included in the recent Environmental Protection Agency assessment of the health effects of passive smoking, and have implications for regulation of environmental tobacco smoke exposure in Louisiana. T HE US ENVIRONMENTAL PROTECTION AGENCY has conducted an assessment of the health effects of passive smoking, specifically lung cancer in adults and respiratory disorders in children. As a result of this assessment, passive smoking has been determined to be causally associated with lung cancer in adults and will be classified as a Group A (known human) carcin- ogen. This determination will have a wide ranging impact in terms of regulation of exposure in the work- place by OSHA and in public places by local, state, and federal governments. The largest and most recently reported study in- cluded in the EPA assessment is an NCI-funded multicenter study which was planned and coordinated by Louisiana ',3tate University Medical Center.' The findings of the first 3 years of the study will be summa- rized here. METHODS In late 1985 a multicenter US study was initiated to evaluate the association of environmental tobacco smoke (EI5) with risk of lung cancer in nonsmoking females. The study was designed to minimize some of the methodologic problems which have been dis- cussed in reviews of ETS-lung cancer studies by the National Research Council,Z and the International Agency for Cancer Researchs among others. These in- clude misclassification of smoking status, accuracy of case diagnosis and cell type, recall bias, ETS exposure from sources in addition to spouse, and control of, potential confounders. This study is a population-based case-control study of lung cancer in women who never used any type of tobacco. Tobacco use was defined as 100 or more cigarettes smoked or any other tobacco used for more than 6 months. Geographically, the study in- cluded five metropolltan areas - Atlanta, Houston, Los Angeles, New Orleans and the San Francisco Bay Area - representing a population base of approxi- mately 18.5 million people. Rapid case ascertainment procedures, which in- cluded weekly or biweekly review of pathology re- ports from study hospitals, were used to identify po- tentially eligible lung cancer cases. Upon completion of case ascertainment over 17,500 potentially eligible cases were screened. Eligible cases included English-, Spanish- or Chinese-speaking females ages 20-79 whoo had a histopathologically confirmed diagnosis of pri- mary carcinoma of the lung made prior to death, had no history of previous cancer, and were determined to be a lifetime nonuser of tobacco. Two control groups were selected in order to eval- uate response bias which is always a consideration in comparisons of i.ll cases and healthy controls. The first control group (population controls) was selected by random digit dialing and supplemented by random sampling from the files of the Health Care Fi.nancing Administration for women aged 65 and older. Controls were frequency matched to cases on age (<50, 50-59, 60-69, 70+ years) in a 2:1 control: case ratio. They met the same eligibility criteria as cases for age, residence, language, and tobacco use. A second control group was selected from females aged 20-79 with a diagnosis of primary carcinoma of the colon who met the same language, previous cancer, residence, and tobacco use criteria and were matched by 10-year age groups and race to cases. A multistep procedure was used to determine life- time smoking status. The medical record of each poten- tially eligible lung cancer case and colon cancer control was reviewed for information on tobacco use. Patients identified in the medical record as current or former smokers were considered ineligible. The physicians in charge of patients with lung and colon cancer who were still considered potentially eligible were then contacted for additional information on tobacco use. Potential study subjects with lung cancer and colon cancer identified as current or former smokers by their physicians were considered ineligible. All remaining ~ C C C JOURNAL VOL 145 ArR,L 133

potentially eligible study subjects (cases and both types of controls) were contacted by telephone to elicit information on tobacco use. Women who reported ever smoking 100 or more cigarettes or using any other form of tobacco for more than 6 months were consid- ered ineligible. At the time of interview, these same questions were repeated to confirm each study sub- ject's reported nonuse of tobacco. About 4% of other- wise eligible women with lung cancer were found to be never smokers after this multi-step screening proce- dure. At the time of interview, and not prior to that time, a urine sample was requested from each study •subject by the interviewer. The specimens were stored at -20°C until shipment to the American Health Foun- dation for analysis of cotinine and creatinine. This bio- chemical determination was used as a verification of current smoking status, to exclude study subjects likely to be current active smokers. Overall about 2% of the study subjects were excluded from further analy- sis with cotinine/creatinine levels of 100 ng/mg or higher. This level was chosen to eliminate persons most likely to be active smokers while allowing for the possibility of very high ETS exposures.' Determined to be ineligible by this criterion were 0.8% of lung cancer cases, 2.6% of colon cancer controls, and 2% of population controls tested. The lower proportion of cases with cotinine levels above 100 ng/mg in our study as compared to colon cancer controls and popu- lation controls is likely to be the result of prior medical record reviews and physician queries used to screen out never smokers from the lung cancer and colon cancer control series. The information was more com- monly available from these sources for lung cancer` cases; therefore, a larger proportion of potential cases were excluded as ineligible prior to obtaining the urine specimen. The extensive procedures taken in this study to screen for tobacco use serve to minimize the misclassification of smokers as nonsmokers. lrt addition to obtaining pathology reports for each case, representative slides were requested from the hospital to confirm and uniformly classify each case. Slides were available for 85% of cases and were reviewed by one pathologist specializing in pulmo- nary pathology. After review, 2% were found to be ineligible and were excluded from the study. One of the specific aims of the study was to evaluate the histo- logic specificity of the ETS to lung cancer association by exain.ining the relationship for each of the main histologic types. A very high proportion of the cases were adenocarcinomas, around 75% consistent across all study centers. Therefore, the report considered the association of ETS with all primary lung cancers and with primary pulmonary adenocarcinoma. RESULTS The estimated risks of lung cancer in nonsmoking women associated with ever having lived with a spouse who smoked are shown in Table 1. The excess risk associated with exposure to any type tobacco smoke from a spouse was approximately 30% regard- less of which control group served as the comparison. An increase in risk of approximately 50% was ob- served for adenocarcinoma of the lung with each con- trol group. Separate analyses were conducted for sub- jects who personally responded and for those for whom information was obtained from surrogate re- spondents. The findings were consistent for self- and proxy-respondents. All odds ratios were adjusted for age, race, geographic region, respondent type, income, and education. An approximate 30% increase in the risk of lung cancer associated with spousal ETS expo- sure persisted after additional adjustment for con- sumption of vegetables (the dietary risk factor confer- ring ring the greatest protective effect), family history of lung cancer, and employment in high risk occupations or industries.s Household radon was determined in a sample of case and control homes under separate funding. Radon levels are quite low in all of the areas included in our study. Less than 1% of all homes tested had levels of 4 picocuries/liter or higher. The observed increased risk of lung cancer associated with ETS expo- sures are unlikely to result from confounding by these factors. A positive dose-response was observed for all lung cancers and pulmonary adenocardnomas with several measures of spousal ETS including dose, dura- tion, and pack-year (Figure). Trends were similar, but tended to be higher, for pulmonary adenocarcinoma than for all lung cancers combined. Exposure to ETS from various sources during adult life was evaluated, in addition to spouse-related exposures. Exposures to cigarette smoking from other household members, on the job, and in other activities of adult life ("social exposure": exposure of 2 or more hours a week from sources other than occupational or household) were associated with overall 40% to 60% 134 JOURNAL VOL 145 APRIL

. TABLE 1 ~ A A,SSOCWT}UN 'OF LUNG CANCER RISK AND SPOUSE'SSAIOKNJG HISTORY : Adjusted Estimated Relative Risk Colon Cancer Contro/s Population Controls v5. vs. All lung carcinomas Adenocarcinoma only All lung carcf~omas Adenocarcinoma only AII types of tobacco 1.28 1.44' 1.29 1.47' Cigarettes 1.17 1.31 1.20 1.36' Cigars 1.14 1.05 1.26 1.15 Pipes 1.17 1.16 1.21 1.20 ' Adjusted for age, race, study area, annual family income, and education p<0,05 ® <15 1rr39 40-79 >79 P€clt YouY 0 AI t.txng Ctrrwr, lrM p-0.07 ®Advccarokwa, tr€rod p<9.01 Figure. Adjusted odds ratios for all lung cancer and for adenocarcinoma of the lung associated with pack- years of exposures from spouse(s). significant elevations in risk of adenocarcinoma of the lung and all lung cancers combined (Table 2). Sibnifi- cant positive trends of increasing risk with increasing years of exposure were found in each exposure setting. No assoc<ation was found between risk of lung cancer and childhood exposures to cigars, pipes, ciga- rettes, or all tobacco types combined. These exposures -----, . $,.C{{. .-,.~y ~ ~.t A-C40CtA Aderiocarrinoma Exposure Source All Lung Cancer of the Lung Spouse 1.21 (0.96-1.54) 1.38 (1.041.82)' Other household members 1.23 (0.97-1.sfi) 1.39 (1.85-1.82)' Occupational 1.34 (1.03-1.73)' 1.44 (1.06-1.97)' Social' 1.58 (1.22-2.04)' 1.60 (1.19-2.14)' ' Adjusted for age, race, study area, annual income, and educa• lion 'p<0.05 , - were limited to the first 18 years of life. After that age, exposures were attributed to adult 1ife. DISCUSSION This study included methods to evaluate recall bias, minimize classification of smokers as nonsmokers, en- sure accuracy of diagnosis and classification of lung cancer, and adjust risk estimates for potential con- founders. The overall 30% increased risk associate•i with ETS exposure from a smoking spouse is remark- ably close to the 25% to 34% estimates of the evaluation of relevant studies in the 1986 report of the National Research Council. The significant positive dose re- sponse to exposure to tobacco smoke within house- JOURNAL VOL ]45 APRIL 135 ®

SETAMMANY PA R I S H H O S P I TAL Lifetime Fantily Hecdthcare St. Tammany Parish Hospital is seeking a physi- cian to provide services at our urgent care center located in beautiful, expanding Mandeville, Louisiana. The center is open daily and sees ap- proxirnaiely 900 patients per month. Services are shared with another physician. The candidate must have Family Practice, Internal Medicine, or Emergency Medicine experience/residency trained. For more information, please call or send letter and CV to: Charles Shcrw Human Resources St. Tammony Parish Hospital 1202 South Tyler Street Covington, LA 70433 (SII4) 8sa-44s2 I liE..fl441i11ka A PREMIER GROUP PRACI'ICE. A QUALITY OF Lim SECOND To NorrE. At Mtirsh.Ddd Clinic, a 400-physician multi-specialty practice, you'll experience an ideal baLulce txtween professional and clinical excellence and superior quality of life. As e pbysician with MarshSeld Clinic in Minocqua or Rice Lake, Wisronsin, or other regional center locations, you will 6nd a friendly, relaxed but industrious community focused on family. Plus, you'll enjoy the best that mother nature has to offer year 'round, while - prai:ticing with one of the most respected, and highly recognized group practices in the nation. I:f this perfect combination intrigues you, please cone.nct David Draves at 1-800-782-8581, ext. 7-5376. 1P MARSHFIELD CLINIC 1000 North Oak Avenue Marshfteld. W15d449 136 JOURNAL VOL 145 APRIL soElAM4FMN holds, in occupational, and in social settings during adult life strongly supports an etiologic role of ETS in lung cancer in nonsmokers and extends the findings from the home into the workplace and public settings. Louisiana physicians can play a key role in secur- ing passage of legislation to minimize exposure to en- virorunental tobacco smoke within the state. The Or- leans Parish Medical Society and the Louisiana State Medical Society have been leaders in the Louisiana Smoking or Health Coalition which has lobbied for several much-needed bills to regulate exposure to to- bacco smoke and availability of tobacco to minors. Grass roots support from the medical communi.ty is very effective, and additional effort is likely to result in passage this year of legislation to further reduce ETS exposure in public places and work sites.  REFERENCES 1. Fontham ET}i, Correa P, Wu•Williams A. et al. Lung cancer in nonsmoking women: A multicenter case-control study. Cancer Epidnniot Bimnarkrrs Prev 1991;1:35--13. 2. Enuironrntntat Tobacco Smokt: Meaturing Ezpoeures and Assessing Nmlth Ff- frcts. Committee on Passive Smoking. Board on Environme tal Studiee and Toxicology, National Research Council. Washington. DC: National Academy Press, 1986. 3. O'NeiB tK, Brunnemann KD, Dodet B, et al (editors). Envirvnmrnfal CGrctno- gens. Methods of Analysis and Expoeure Meas uernent, Vol 9, Passive. Smoking. Lyon, Frarxr. internationa] Agency for Research on Cancer, 1981. (1ARC Scientific Pubtiotions no 81.) 4. Mattcsen ME, Boyd G, Byu D, et al. Passive smoking on covtmeldal airline flights. fAMA 19H9:2b1:867-872. 5. Fontham ETH, Coales R, Dilley A, et al. Lung cancer in nonsmoking women: Dietary antioxidants. Cancer Epidcmiot BiomarFcrs Prcv 1992;1250. 77u authors are from the Drpt of Pathology af Louisiana State University Medical Center in New Orleans. Reprint requests to: t3r EliznGrfh Fontharn, Dept o/Patiwlogy, LSU Medical Center, 1901 Perdido St, New Orleans, LS 70712. This work was supported by ROl-CA-40095 and a grant from the Stanley Scoft Cancer Center.