Document 5kEy2nbNR1p0wadLeXrkYwn68

DownloadSearchable PDFRandom document
FILE NAME: Phenolic Resins (PHR) DATE: 2004 DOC#: PH ROM DOCUMENT DESCRIPTION: Journal Article - Lung Cancer and Mesothelioma among Male Automobile Mechanics-A Review V i y i kvo1\t V *yjtj KUUW 'V REVIEWS ON ENVIRONMENTAL HEALTH VOLUME 19, NO. 1,2004 Lung Cancer and Mesothelioma among Male Automobile Mechanics: A Review Francine Laden,51 Meir J. Stampferu -4and Alexander M. Walker3 'Charning Laboratory. Department o f Medicine. Brigham and Womens Hospital and Harvard Medical School. Boston, Massachusetts; departm ent o f Environmental Health, Department o f Epidemiology a n d 4Department o f Nutrition, Harvard School o f Public Heath, Boston, Massachusetts abstract BACKGROUND: Extensive, overt asbestos exposure dearly causes lung cancer and mesothelioma. Elevated cancer risk has also been documented for short term or low levels of asbestos exposure. Automobile mechanics are potentially exposed to asbestos through brake repair work. Few studies have specifically examined the risk of lung cancer and mesothelioma associated with exposure to asbestos in brakes, but automobile mechanics are included in many studies of occupation and cancer. METHODS: We critically reviewed ail epidemiologic studies of lung cancer and mesothelioma risk among male automobile mechanics identified through a broadly based Medline search and scrutiny of references in primary and review articles. We discuss the studies grouped by study design and control for smoking, the major risk factor for lung cancer. RESULTS/CONCLUSIONS: We reviewed all individual analytical studies meeting our criteria, plus all available case series and case reports. When examined in aggregate, the evidence did not support an increase in risk of either lung cancer or mesothelioma among male automobile mechanics occupationally exposed to asbestos from brake repair. ReoriM requests to: Dr. Francine Laden. Chanmng laboratory, 181 Longwood Avenue, Boston, Ma^chuseCs. USA: e-mail: franeine.laden@ctennirigJi^anjx! O 2004 Freund Publishing House Ltd. keyw ords epidemiologic studies, occupational risk, friction material, brake repair, asbestos INTRODUCTION Extensive, overt exposure to asbestos fibers clearly causes lung cancer and mesothelioma. Elevated cancer risk has also been documented for short term or lovi levels o f asbestos exposure M. Automobile mechanics are potentially exposed to low levels o f asbestos through repair of friction materials. Few studies, however, have specifically examined the risks o f lung cancer and mesothelioma associated with work on automobile brakes. In this paper, we summarize and critically review the available epidemiologic studies on the risk of lung cancer and mesothelioma associated with work as an automobile mechanic. About 50% o f the composition o f brake linings and brake shoes consists o f chrysotile asbestos encapsulated in a phenolic resin. The use of such products in automobiles, buses, and other heavy equipment Gauses them to wear, releasing a fine dust o f amorphous material together with very small fibers (V. Automobile mechanics are potentially exposed to asbestos fibers when they grind used linings, bevel new finings, or remove the loose dust from the brake drum and back plates using compressed air, brushes, or wet cloths /3/. 39 e ;jyyd sqi wyitisa now t^oos-e-noN 40 F. LADEN, MJ. STAMPFER, A.M. WALKER Although total dust counts in the ambient air can reach high levels for short periods of time during these processes, control methods, such as enclosed vacuum systems or exhaust ventilation systems, effectively lower the exposure /4-5/, Furthermore, the availability of asbestos fibers in used brakes is markedly reduced because high temperatures during braking degrade the vast majority o f the asbestos into nonfibrous magnesium silicate /6L Less than one percent o f the dust released in work on brakes contains chrysotile fibers that are longer than five microns, the size known to be relevant to respiratory health 171. Grinding and drilling of friction pads also releases fibers, which have been of concern. Detailed study o f this material has revealed, however, that the fibers consistently remained bound to the resin matrix /8/. Thus, grinding and drilling also appear to pose no important exposure to asbestos fibers. The absolute exposure level js influenced by the distance to the source, the type and duration of activity, as well as by the type of brake. For example, drum brakes used for trucks, buses, and certain passenger cars, require more cleaning than disc brakes used only in passenger cars 19!. Industrial hygiene reports taking measurements o f dust and fiber generated during clutch and brake repair work demonstrate that the exposure to asbestos from this work is low, intermittent and brief 19!. In the 1970s, before the current permissible exposure limit o f 0.1 fi'mL (fibers per mitliliter of air) was proposed by the United States (U.S.) Occupational Safety and Health Administration, episodic peak concentrations of up to 16 f/mL were reported during blowing off o f loose dust. Practically all measured levels published after 1987, however, were less than 0.2 FmL, and time-weighted average exposures during passenger vehicle repair in the 1980s were generally less than 0.05 frmL /10-12/. Nonetheless, the question remains, were these low levels of asbestos exposure associated with carcinogenic risk /3/. Concerns about this possibility led to warnings, from both government agencies and companies, but with little or no epidemiologic evidence. We therefore reviewed the epidemiologic studies that provided data on risk of lung cancer and mesothelioma among automobile mechanics. METHODS We identified studies o f lung cancer and mesothelioma risk among motor vehicle mechanics through computer searches and scrutiny of references in primary and review articles. In an attempt to capture all published data on potential exposure to asbestos from work with brake linings, we considered general surveys of occupation and cancer risk, as well as studies focusing specifically on cancer among automobile mechanics. Because of the potential for publication bias by both investi gators and'joumals to highlight positive findings, we performed broadly based Medline searches (1966 through August 2003), using as key words cancer, lung cancer, occupation, and mesothelioma. We identified many useful studies that might have been overlooked because they did not specifically mention automobile mechanics in the title or abstract. We also contacted authors for supplemental data and tables if it appeared that the relevant results were available but did not meet the criteria described for inclusion in the published paper. We focused specifically on automobile mechanics, as opposed to broad occupational or industrial categories. For completeness, however, we also included papers that discuss the occupation mechanics, unspecified. Separately, we considered papers that do not mention automobile mechanics but from their design and results we concluded that automobile mechanics would have been included if the results had been positive. Therefore, we included all studies in which specific occupations were considered. The review is limited to studies o f men. i. 3z>ua !QI WH3T :S0 NOW t>003-8-AON LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS Not all studies in this discussion controlled for cigarette smoking. Because smoking is such a strong risk factor for lung cancer, however, and there is evidence that smoking rates may vary by occupation, we have placed more emphasis on the studies that included smoking information in their analyses. Available occupational survey data on smoking is limited. A survey in the U.S., however, found that automobile mechanics ranked third in smoking prevalence out o f 61 occupations surveyed, with twice the average cigarette consumption of many other similarly skilled workers l\H. In The National Center for Health Statistics 2000 National Health Interview Survey, the prevalence of current smoking among repairers and mechanics was 31.5%, among the highest o f those occupations presented /! 4/. Also, some evidence indicates a higher prevalence o f smoking-related causes of death, such as chronic obstructive pulmonary disease (COPD), among mechanics. For example, in the 14 year, 28-state National Occupational Mortality Surveillance (NOMS) study, covering 66,150 deaths among male automobile mechanics, the proportional mortality from COPD was statistically significantly elevated at 1.15 /15/. Studies' that do not control for smoking are therefore likely to be confounded and to overestimate any association between work as a vehicle mechanic and risk o f lung cancer. As an illustration, Carstensen and colleagues 111! found that accounting for smoking eliminated an apparent excess lung cancer risk in mechanics and repairmen in Sweden. We discuss separately studies of lung cancer and mesothelioma. The details on study design, including study population, assessment of exposure and outcome, comparison population, and factors adjusted for; and the main results are presented in the tables. We present the risk estimates (standardized mortality ratios [SMR], standardized incidence ratios [SIR], proportional mortality ratios [PMR], odds ratios [OR], relative risk [RR], among others) and their statistical significance as the authors did, including confidence intervals or p values, whichever are available. We made no attempt to calculate additional estimates from the information included in the papers. The tables are organized by study design, and, for lung cancer, by whether the authors controlled for smoking in their analysis. LUNG CANCER Studies of Populations of Automobile and Bus Mechanics We identified five studies that focused specifically on populations of automobile and bus mechanics (Table 1). Jarvholm and Brisman /17/ and Hansen /18/ identified automobile mechanics through population census dala, and Schwartz /! 9! extracted this information from the `usual occupation' recorded on death certificates. Rushton et al. /20! and Gustavsson et ai. !21/ constructed their cohorts and determined exposure from employee work history records at bus garages in London and Stockholm, respectively, in all but the Hansen study the cancer risk o f the `exposed' population (automobile mechanics) was compared with the risk in the general population. Hansen instead constructed a comparison group of skilled workers including carpenters, electricians, instru ment makers, dairymen, upholsterers, and glaziers. This group was thought to be similar to automobile mechanics in terms o f physical demands, life-style habits-- including smoking, geographic distribution and socioeconomic status. With the exception of the Jarvholm and Brisman study (SMR: 1.70, p < 0.01) /17/, no significant association o f lung cancer risk with occupation as an automobile mechanic emerged from these studies. Gustavsson and colleagues 1211 calculated cumulative exposure to asbestos, as well as diesel 3=dy0d :QI WbT80 NOW fr003-8-ON LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS 43 exhaust, over a 25-year period, extrapolating.from work histories and a current industrial hygiene assessment. The authors observed a statistically nonsignificant elevation o f lung cancer incidence overall {SIR: 1.61; 95% Cl: 0.94-2.57), but no dose-response association with asbestos exposure. Schwartz reported a PMR o f 1.12 (95% Ci: 0.82 !.53) and the SMRs reported by Rushton et a). 1201 and Hansen /18/ were not elevated. None o f these five studies adjusted for smoking. Survey and Cohort Studies of Occupation and Cancer These studies were undertaken to understand the relation between a variety o f occupations and the risk of lung cancer specifically, or cancer in general in a defined population (Table 2). Most relied on cancer registries or death certificates for outcome and exposure assessment, usually defining exposure as last known or `usual' occupation. Lung cancer rates among automobile mechanics are compared either to rates in the general popu lation or to rates in the entire cohort under study. Two large cohort studies controlled for smoking. In the most comprehensive survey of cancer and occupation, Blair and colleagues 22! evaluated the relation between occupation and specific cancer sites over a 16-year period in 293,958 t).S. veterans 22!. Information on usual occupation was obtained through two interviews administered near the beginning o f follow-up in 1954, and outcome information was extracted from death certificates through 1970. The study was updated through 1980 by Hrubec !2V. At the interview, information on smoking habits was also obtained. Risks for each occupation and each cancer site were calculated, within categories defined by smoking habit. No excess o f lung cancer was found for automobile mechanics in the original Blair report (SMR=1.03, p>0.05) or of cancer of the respiratory system in the update (SMR=1.10, 90%C1: 0.89-1.36). In contrast, and as expected, risks were elevated for lung cancer in occupations with known asbestos exposure, such as shipyard workers and plumbers. Carstensen and colleagues /16/ took advantage of Sweden's extensive population records system to evaluate the incidence of lung cancer. The authors focused on 1.6 million men aged 30 to 64 who were employed in 1960. The researchers reviewed national cancer registry files for any occurrence o f lung cancer from 1961 through 1979. Occupation was determined from the 1960 census. Carstensen's group did not attempt to evaluate the smoking habits o f each worker, but they did estimate the amount o f smoking in each occupational group based on surveys, and used those data to adjust for smoking indirectly. After such adjustment, the SIR for mechanics and repairmen taken together was 1.07 (95% CI: 0.92 1-22). No separate, more specific estimate was made for automobile mechanics alone. The remainder o f the cohort studies relied solely on registries or death certificates fo r information on occupation and disease surveillance in a given geographic area. Therefore, information on potential confounders, including smoking, was not available. The results from these studies were mixed. Enterline and McKiever /24/ observed a statistically significant association o f work as an automobile mechanic with lung cancer mortality in the U.S. population in 1950 (SMR = 1.28). Milham and colleagues 25/ surveyed deaths in Washington State, first in 1976, with updates every 10 years through 1999126!. In the 1999 publication, the PMR for lung cancer deaths among automobile mechanics was statistically significantly elevated at 1.15. This result is similar to the findings reported in the NOMS study, described above, which included residents in 28 states. In the latter study, the PMR was 1.16 for white and 1.08 for black automobile mechanics /ISA Although the excess mortality for each group was modest, the sample size was large 1 QL n u -ta o ' iM t F. LADEN, MJ. STAMPFER, A.M. WALKER TABLE 2 General surveys of cancer and occupation, organized by control for smoking: lung cancer results Author Study population Contrai for smoking Outcomes included Sample size Number of auto mechanics Outcome/ Exposure assessment Comparison population Results for automobile mechanics' N=no. Cases Bar et al. 1885(22) US Veterans age 31-84 hr Al cancer 1953 deaths n=293,958 195469 Tdtat deaths NA -107,563 Hrubec et al 1992(23) updated from Blair 1985 Carstensen et at 1988*{16) US Veterans who served Alt deaths fr 1917-40 and held health 195460 Insurance potldes, age 31-84 in 1953 n=248,046 Swedish mates employed tilingcancer 1960, age 30-64 1961-79 =1,622,547 Didnot controlforsmoking Total deaths NA - (64.785 HA NA DC/usual occupato fhomquestonaires Entire cohort DC/usual occupato from questionnmes Entire cohort Caner Beglsiiy/1960 . General pcpitetit. CrtfemhQ rnfa fr. Hesiripie ^ - Neversmokets; n=2; Ever smote: rt=26, SMR=0.91. p6,6,`Alt'dbmbmed: n=33, crude SMR*tbip><)5 Nonsmokers: n=2, RR=0.7: Current cigarettas: n=50. RR=1.2 (90%CL- 0.93-1.49): Total smoking adjusted: n=63, RR=1.1 (90%Ct: 0.89-1.36) ^ ^ % r . !22>. 1 ~ Enterte & United Slates, mates, age Lungcancer NA McKrever 1963 2064 deaths 1950 (24) 681,485 DC MilhamA Ossiandw 2001(28) Washington Stale, US whites, age >20 Aideaths Maf9s:195099 Total . 10,473 dea.tes: ; t, MtS! 811,526 . oc ; JP NIOSH2002 Nations) Occupational Alldeaths NA NA DC (15) Mortality Surveillance (NOMS), 28 stBles: age 15-90:1984-1998 General population with n=119, SMR=1.28, PMR=1.27 work experience p<0.05 Genera! population r j S S f . '' ' . Pfyfe=il)V p<0.01 NOMS population Whites: n=5,743; PMR=1.16, psO.01 Blacks: n=76S; PMR-1.08. p0.05 R i c n a r a i_emen n o v u un u a ; i u a A 1 _ A ----------- -------------------------------------------------- > TABLE2 General surveys of cancer and occupation, organized by control for smoking: lung cancer results 3 cs Outcome Sample Number of Outcome! Comparison Results for automobile <x Q_ Author Study population included size auto Exposure assessment population mechanics' mediantes N=no. Cases I- 3 Andersenetai. Detwiafcl971-1987, ABiawer CSoer 353,757 CaricerRegiSliy/1970 GeneraipopuiaBOn Mechanics, IfOrtandmetalwara 1999(27) Sweden 1971-1989, cases cases i:'/e a te :/M I S '.! - womens: ' 3 Finland 1971-1990. trail ' \^ \ t' neM14; 887*1.202-1.4) 3 Norway 1971-1991, age r 26 " 3 N*10mit : , . :r Leigh 1996 Catifomra Occupational Umg cancer Altdeaflis - NA DC General poputaSon n=NA; SMRM.31 (1.08-1.54) I B ________ MortalityStudy: age 16-64 197961 173,438 MenckSt LosAngelas County, Lungcancer. Deaths- 121050 "j DrfetitrCcirKiiemt General : - OocuOatlbn- ftdmefflltfeibd.. Henderson whitemales, age 64 . SB'1968- 2,161; population Industry- n=28, SMl?=f.46, 1976(28) rmt,6&018fo CesesIrtpIdSraiteses \7i>h' Cliikrfd&ejmcd i*'-":'! *? 5?t,, yl;--,{'',l.~ 1' P<0,01 Petersen & California, US. white all deaths Total deaths 1,833 DC Genera! population n=69, PMR=1.23, p>0.05 Wilham1980 males, age >20 195961 --200,000 t {29} t Olsen & Cenk lieffkser: ' AilceWs-i:: Whdfecchfirt 0*34; SPIR=1.17 (0|41,63) J Jaren 1987 etedi 1970-79 90,651 ' 3 <30) . D 3 Mormon 1957 Scotlandmales age 2:15 AHdeaths NA NA DC General population Not mentioned PI) N=1,795,273 1949-53 Pearce& NdwZeNand males, age caiiodfdeaths Cancer M ' Death Registry General population NdtntenRorted . . Q Howard 1986 1564 ' 1974-78 ' deaths- 18%sample 978N.7 0 (32) 6,356 Census a T 3 co 3 G3 r 3 oZ 3 3 Q > QCU 3 03 i Z> o TABLE 2 (continued) General surveys of cancer and occupation, organized by control for smoking: lung cancer results Author Study population Minier and Beer-Porizek 1992(33) Howe& Lindsay 1983 (89) Working population Switzerland, males age 10% Sample, Canadian Labor Fn, males N*415,201 Outcomes included Cancer deaths 1979-82 cancer deadis 1965-73 Sample size Cancer deaths 4,008 Tote! deaths -19,374 Cancer deaths- Number of auto mechanics NA NA Outcome/ Exposure assessment OC Death Registry/ Unemployment lnstfirSm:Canmlsslon Comparison population General population Whole cohort Results for automobile mechanics' Nsno. Cases Not mentioned Nat elevated 4,203 Ltingca- 1,370 Lindsayet at. 10%sample Canadian Cancerdeaths Cancers- NA 1993(34) Labor Force, males 1965-79 9,739 0=415.309 Firthet at NewZealand males, age AHcancer AHcases- NA 1996(35) 15-64 cases 1972-84 26,207 Death Registry/Census Whole cohort 1961 Not mentioned Cancer Registry Sensrei employed population; total employed populaSn of cancer registry Not mentioned Abbreviations: DC=deathcerSficafe. frsftom.n-numb. NA=notaaraSabte. PMR'propotSonaimortalityratio, SiRsstandardizedincidenceratio. SPiR=standerdizedproportional bcldenceratio, Shfft=$tandafd)zsd mortality ratio ' ConfidenceIntervals are 95%unless ofiierwise kujfcsted - 'rosufts are for automobie mechanics unless otherwise specified mechanics end repairmen, nonspecific mechanics,IronandmetalwareworkersgroupedtogeJier, notspecificallyautomobilemechanics 00 J o <x L m Z 3m P > > K 3 Qi--r ma r-l r Q Z o ST (3 Q CU O X I W : U i W tf H b 'M NUW M d k t -b -n U N LUNG CANCER AND MESOTHELIOMA !N AUTO MECHANICS and the PMRs were statistically significant. In another large study, in the Nordic countries, Andersen and colleagues t i l l observed an elevated SIR (1.2; 95% Cl: I.2-1.4) for lung cancer among mechanics and iron and metal workers grouped together 1211. The results for automobile mechanics were not presented separately. The Leigh study, mentioned earlier, observed an SMR of 1.31 (95% Cl: 1.08-1.54) in the California Occupational Mortality Study. The authors did not control for smoking, however, although in the same publication they discuss the high smoking rates in this occupational group 1X51. Menck and Henderson /29/ surveyed incidence and mortality of lung cancer in Los Angeles County and reported a statistically significantly elevated SMR for lung cancer in the automobile repair industry. The authors examined a wide range o f occupations, including automobile mechanics, but listed only the occupations with elevated rates. Automobile mechanics were not included in that list. Petersen and Milbam /29/ and Olsen and Jensen /30/ did not observe a statistically significant excess o f lung cancer among automobile mechanics in California and Denmark, respectively. For comparison, we examined the risks of lung cancer reported for other occupations included in these studies. Consistently, occupations with known asbestos exposure, such as pipe fitters, shipyard workers, and steel workers, had elevated risks of lung cancer. Our search identified five additional survey/ cohort studies that defined exposure to the occupational level but did not mention automobile mechanics /31-35/. For these studies, we surmised from the level of detail o f the occupation and exposure assessment that `automobile mechanic' would have been reported if an excess risk had been observed. Typically, such broad surveys examine dozens or even hundreds o f categories, but only present the positive or interesting associations. That occupations with asbestos exposure are often identified as linked with lung cancer risk, coupled with silence regarding automobile mechanics, provides further, although indirect, confirmatory evidence against an association. Case-Control Studies We categorized the case-control studies by whether th c study controlled for smoki ng, * the source o f the control population, and whether they specifically include information on automobile mechanics as opposed to a broader category of mechanics. Control for smoking and other risk factors is closely related to the selection of the control population, as most population and hospital-based studies can obtain information on potential confounders, whereas death-certificate studies are not. The details are presented in Table 3. The studies that rely on controls selected from the general population and include complete job and smoking histories collected by interview are the most informative. In a U.S. study in areas of New Jersey with high lung cancer rates, Schoenberg and colleagues /36/ identified 763 lung cancer cases from the state cancer registry and 900 population-based controls. The authors defined `automobile mechanics' as anyone working for at least 3 months after age 12 in that profession. The smoking-adjusted OR was modestly elevated but not statistically significant (OR =1.4; 95% Cl; 0.84-2.30). Vineis 1511 and others incorporated these data with similar information from cancer registries and cooperating hospitals in Louisiana, Fforfda, Pennsylvania, and Virginia, but also found no statistically significant differences between the case and control groups in previous employment as an automobile mechanic. Lerchen and coworkers /38/ in New Mexico compared 333 men with lung cancer to 499 men selected from the state's c TABLE 3 L<< c Case-control studies, organized by population type: Lung cancer results Author Study population Outcomes inducted Exposure assessment Controls Adjustment for Mechanic definition; number lung cancer Results Studies that control (or smoking Schoenberg et al. 1967(36) Inchdedln Vmmsot a ttm NJ, US; 6 geo. Areas At high lung cancer rates fr 1967-76; white mates; 1980-81 Lung cancer n=763 Interview, Bfetlro# occu pational history, ob23 mbafter age 12 General poptfaffon n=900 " Race. age. residence, type Interview, Auto mechanics: ; _cafe44, 7 cohfroMO smoking : F. LADEN. MJ. STAMTFER, AM. WALKER Vrneis et ai. 1988' (37) Data from 5 US case-control studies (hospital and population based) residents OfLA, FI, PA, VA, NJ Lung cancer n=2.973 Interview lifetime occupational history, job (8 mo n=3,210 Age, birth cohort smoking Auto brake worker: case=98, ccntrolc90. OR=1.2 (0.9-1.7) OR= 1.2 (0.9-1.7) Lerchenetaf. 1957 (38) New Mexico, Hfepahtewtitteand otfWfwhite mates age 25-84; 1980 82 ' Lung cancer mates n=333 Interview. HbUme occupteUdna) history, Job2 1 yr.afiar 612 ' General population Age, smoking , Automechanics: cassis, , cbhtrc!=25 . 0R4.9 (0.5:1^) r / \ * levin et ai. 1988 Shanghai, China, Lung cancer Interview; ' Gnerai population Ags, smoking, Auto mechanics NA (39) males ege 35-65; n=733 Lifetime occ n=760 not mentioned 1984-85 history, job 2 t yr after age 16 a U3 Abbreviations:DC=deathcertificates. n=number, ORoddsraSo, sMOR-standardizadmortattyratio. RR=refetiverisk Conidence bounds=96%Ct unless otherwiseIndicated oo (S3 a s: 1includes data from Schoenberg et a! 1987; otherstudies did not make our cut, so not includedIndividually sf (3 (3 CU 00 t "i > TABLE 3 Case-control studies, organized by population type: Lung cancer results Author Shirtypopulation Outcomes Exposure Controls Adjustment for Mechanic Lung cancer Li included assessment definition; number Results aa Wiliams at a!. 1977 (* Third National Cancer Sutvey Interview Study, US. Aftcaher rt*3.$39 \ Interview: life Other centers ime,racent ... . . ' Age, raw, education, tntoktog, atoohot, rMfanoB Auto mechanics: 6ases?45; , Net reported as efevafed: ; : LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS Bums &Swanson 1991 (90) (See Swanson et el. 1993} Detroit Occupational Cancer Incidence Surveillance System; age 40 85; 1984-87 Lung cancer n*5,935 Interview; lifetime occupational history, duties, usual occupation Cancers of colon or rectum n=3,956; compared to occupation and industry with least potential for exp to carcinogens Age, smoking, race Motof vehicle mechanio (occupation) v; case*118, controM8, blacks only:case=31, controMj ALL: OR=1.72 (1.15-2.59); BLACKS: OR=4.15 (1.27-13.64) Swanson et ai. 1993 (seeBdns& (41) Swanson 1991) Benhamouetal. 1988(42) France; 1976-80 liing cancer m>3,792 Lung cancer n=1,260 Interview; duration of arhpkiyrriatt . CanCers of colon or rectum; h=1,968 i" iV:. W' , Age, smoking Interview lifetime occ history, a t year Tf}/]'' v ; v:5 ,:;Li 'r:'A iv j ' r - Hospital based, non tobacco related disease cases . n=2,084 Gender, age, hospital admission, interviewer, smoking Aufemechantes, 20+years enjoyed Wlffidsfcase=43, Sritr&i=16; Bfecfe:c8ses18, -3tsSrt&rt=4 Motor vehicle mechanics: case=65, control =96 Whites. OR==1.5 (0.7-3.0) ' Sacks: R*6,(1.435.0)'' RR=1.06 (0.73-1.54) AfibmviaSons:DC=deathcertificates, n:mtnbet, OR'Odd rato, sMOR=standafdizedmortsJByreta, RR=TMiaSvrisk Confidenceboon* 9$%ClunlessotheoriseIndicated a a; cc G z c sr G Ga i ac 1 TABLE 3 (continued) Case-control studies, organized by population type: Lung cancer results WOl U <a a Atrtttof Study population Outcomes included Exposure assessment Controls Adjustment for Mechanic definifion; number Lung cancer Results Spineill at a). 1990 () Diagnoses at Cancer Control Agency of British Columbia, Canada; males >20; 1950-75 All cancer n=6,389 Medfcal records: Hospital based, other usual oocupation cancers Age, yf (did not Auto mechanics: NA control for not fisted as smoking) ' elevated ' Morabiaet al. 1992 24 hospitals in 9 Lung cancer Interview; Hospital based, other Age, race, Mechanics, OR=0.7 (44) metropolitan n=1,793 usual occupation cancer and non-cancer hospital, dale of repairmen, auto; areas, US; males admittances, n=3,228 admission, control$=39 r 1980-89 smoking > Z(4a5f)tmet a. 1989' wMhisitseoumria,lUesS;;1980 nL*un4g,4c3atncer oRcecguisptrayt;ional OtotRheergciasntrcye-resxrdep. otrripte,d Adigaeg,nsomsiosking at . .Mdpbanks and OR=1.3(t:0-1.7) m2a: 85 diagnosis oralcevlty, esophagus, 1=108, bledder, Si-defined arid </> unknown t h*1f,326 Bresiow et al. Caiifomia, US; 11 Lung cancer interview Hospital based: Non Age, race, Mechanic and Not calculated m 1954* (4?) hospitals n=518 Employed in cancer or chest disease gender, smoking repairmen, not F 1949-52 occupation at cases. n=518 available elsewhere > least 5 years classified: k cases=15, eonfrois=16 > Wang et al. 1995* Tianjin, China Lohgeanoer Registry, most Othercancers; n=14,685 Used sutveydata Mechanics,repair Not mentioned as rn rr> (46) males > 20; 1981 n=4,806 recent to ink . industries elevated 73 87 " occupation ocuptfttoniby smoklngststus Oecoufle& Roswell Perk All cancer Interview Hospital based: Non Age. smoking Mechanics and Ever employed: S(4t8a)nislawczyk 1977* MBuefmfaoloriaNlYin,sUtiStute, n=6,434 lhiifsettoimryeocc cuannexceprocsaedse=sclerical rluepngaicrma:ecna:se=84 OR=1.12p>0.05 Q whfla males, 1956-65 workers employed >5 yrs: Employed >5 yrs: OR=1.10p>0.05 OE lung ca: case-63 r- Abbreviations: DC- death certificates, renumber, OR*oft&ratio, sMOR=standariU:i mortalityraSo. RR=relatwerisk Confidencebounds =95%Ctunless otherwisendeated OCD Oe: o* 0 a1i CtD O3 c-y'ju'vg Ui WUi =80 NUW t?00d-8-flON LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS 51 population. The authors found proportionately fewer automobile mechanics (identified as in the Schoenberg study) among the cases than the controls (OR = 0.9; 95% Cl: 0.5-1.9). A similarly designed study in Shanghai China did not mention automobile mechanics in their results /39/. Next we review case-control studies that compare lung cancer cases to men with other kinds of cancer. This approach is not as reliable as selecting controls from the general population because the authors must assume that automobile mechanics are at the same risk as the genera! population for those other cancers used for comparison. As in the general population studies, work and smoking histories were obtained by interview. Williams and colleagues MOl compared lung cancer cases with other cancers in the Third National Cancer Survey population. No excess of lifetime or most recent work as an automobile mechanic was found among cases compared with controls /40/. Bums and Swanson M l/, working in the Detroit Metropolitan Ansa, found a statistically significant OR for automobile mechanics when comparing Jung cancer cases with colorectal cancer cases (OR = 1.72; 95% Cl: U 5-2.59). This result more likely represents a deficit o f automobile mechanics among patients with colorectal cancer than a direct link with lung cancer. In a follow-up of this study, the authors evaluated the risk o f lung cancer by number of years employed in specific occupations separately for black and white men. The authors observed a significant excess of automobile mechanics who had been working for at least 20 years among black lung cancer cases (OR = 6.9; 95% Cl: 1.4-35.0) but not among white cases (OR = 1.5; 95% Cl: 0.70-3.00). The number o f biack cases and controls was small, however, and risk did not increase linearly with increasing years of employment in either white or black subjects. Three case-control studies compared men hospitalized for lung cancer with men hospitalized for other causes, controlled for smoking, and specifically assessed automobile mechanics. None o f the three studies found an excess of automobile mechanics among the cases compared with the controls /42-44Z. We also identified six additional case-control studies using cancer controls that included results for a more general grouping of mechanics, presumably including automobife mechanics. Zahm et al. /45/ observed a modest excess of mechanics and repairers among lung cancer controls in Missouri (OR * 1.3; 95% Cl: 1.0-1.7). The other five studies found no material association between work as a mechanic and lung cancer risk /46-50/. In three death-certificate based case-control studies, the controis were drawn from among the non-lung cancer deaths in the general population. These studies relied on the information provided on the death certificate to determine usual or last occupation and did not control for smoking. Dubrow and Wegman observed 29% /51/ and Milne and colleagues 20% more automobile mechanics /52/ among their cases than among their controls. Neither result was statistically significant. In a similarly designed study by Finkelstein and colleagues /53/, 12% fewer automobile mechanics were found among lung cancer deaths than among other deaths. MESOTHELIOMA Case-Control and Cohort Studies Exposure to asbestos is strongly associated with risk o f malignant mesothelioma o f the pleural and peritoneal lining. A baseline incidence of the | disease is not attributable to asbestos exposure, but in the majority of cases some contact with asbestos can be documented /54/. Although only amphibole fibers, as opposed to chrysotile (the asbestos form 2 used in friction products), have been consistently r' I?vceoto :QI WBT 80 NOW i7003-8-nON 52 F. LADEN, MJ. STAMPFER, A.M. WALKER linked with the disease /54/, the association of mesothelioma with work as a brake mechanic has been explored. We reviewed this literature here (Table 4). Several o f the studies discussed above included mesothelioma as well as lung cancer. Four mesothelioma cases were reported in the cohort studies focusing on automobile and bus mechanics (one each by Jarvholm and Brisman /17/ and by Hansen /187 and two by Gustavsson et al. 721/). In all cases the authors mentioned possible other types of exposure to asbestos from previous employment and concluded that no suggestion of an increased risk o f mesothelioma could be ascribed to work as an automobile or bus mechanic. Mesothelioma was reported in three o f the general cancer surveys. Olsen and Jensen 730/ stated that no mesothelioma cases were reported among men who repaired automobiles and motorcycles in Denmark. Dubrow and Wegman 155/ reported on mesothelioma but did not observe any cases among automobile mechanics. The NOMS study reported 18 mesotheliomas among white automobile mechanics (PMR = 0.77) and 2 among black mechanics (PM R* 1.54) /15/. The Blair study o f cancer mortality among U.S. veterans did not mention mesothelioma 1221, and the follow-up study by Hrubec et al. 723/ grouped all respiratory cancers together. We identified 11 studies that focused on mesothelioma and occupation /56-67/. None of these studies reported a significant association between mesothelioma and work specifically as an automobile mechanic. Malker and colleagues 7687 observed a statistically significant elevated incidence o f mesothelioma among the broad category o f mechanics and repairmen (not limited to automobile mechanics). This finding, however, was based on only 16 cases dispersed across a variety of industries, including railroad equipment manufacture, sugar factories, and shipbuilding, so whether the cases were exposed to asbestos through work on brakes is unclear. Jarvholm and Brisman's study 7177, cited earlier, extended Malker's work, focusing more closely on auto mobile mechanics, among whom only one case of mesothelioma was found. All these studies reported an increased representation of insulation workers and other jobs with overt asbestos exposure among the cases, lending further credibility to the null findings for automobile mechanics. Woitowitz and Rodelsperger performed a casccontrol study o f 324 cases o f mesothelioma in Germany and presented the results relating to automobile mechanics in a separate publication 769/. The authors identified 16 automobile mechanics among the case patients, and 16 among 315 hospital controls, most o f whom had lung cancer, and 12 among 182 population controls. The comparison between the cases and the hospital controls is uninformative because lung cancer is also hypothesized to be associated with the exposure. No evidence o f an excess risk of mesothelioma was found for automobile mechanics using the population controls. The authors include individual descriptions o f duration of employment, history o f brake service, and other asbestos exposure for each automobile mechanic. All work histories were very similar regardless of their disease status. Because no comparison statistics, including ORs, were calculated, and the nature of this study was primarily descriptive, it may be more appropriate to consider this investigation a case series as opposed to a forma! case-control study. In 2001, Wong 770/ published a meta analysis combining the results from six studies that assessed mesothelioma among automobile mechanics---McDonald and McDonald, 1980 1511, Teta et al., 1983 7587, Spirtas et a!., 1985 760/, 1994 /617, Woitowitz and Rodelsperger, 1994 /697, Teschke et al., 1997 (63) and Agudo et al., 2000 /657. Wong calculated the RRs for the individual studies from the available data when they were not 1 > > ---------------------------------------------------------------------------------------------------------------------------------> TABLE 4 Studies o f mesothelioma among mate automobile mechanics Author Study Population Outcome Assessment Exposure assessment Comparisonpopulation Mesothelioma Results Studies ofallcancer, Including amention ofmesothelioma JatyhotmS Brisfrmn1988 m Mates, age 20-59, employed Death Registry as ntechanic or in car repair 1961-73 iii 1959ln:Sweden: Cancer Registry n=2L905 1961-79 Cersus occupational General population arid industry codes 'Car repair mechanic'; eases=1. (also worked a construction worker) lung cancer and mesothelioma in auto mechanics Hansen 1989(18) Occupationallyactive, skilled Death Registry, DC, mate workers, age 15-74 in 1970-80 1970 InDenmark, n=21,800 Census Carpenter, electrician, instrument maker, dairyman, upholsterer, glazier n=51,747 `Auto mechanic' or 'mechanic, employed at car-repair workshop' cases=1 Gustavsson et. 1990(21) Mstes emptoyed at least 6 rrwnfh$in 5 Stockholm bus garagesfrom 1945-1970, rp695 ... Death Registry&other local records 1952-86 Cancer.Registry 1958-84 Cumulairyaexposure Stockholmoccupaikrnally towbestosfam active popilbitonand Cbnipariyrecordsand generalpopulalion ex^iidtevete) MeChBflics.setvicernsn, hostlers cKdd^;;:' " Oteen 6 Jensen 1987 (30) Denmark, all cancer cases 1970-79 n=90,651 ' Cancer Registry ; Supplementary pension fund; central population registry Supplementary pension fund; central population registry Whofe cohort of cancer cases 'Repair of motor vehicles and motor cycles' cases*0 Dubrowi Massachusetts US; all DC VVegman 1984a deatha4jB7i-73; white (55) males.aga^O ..... .... -ft NIOSH2002(15) National Occupational DC Mortality Surveillance (NOMS), 28 states: aae 15 90; 1984-1998 0C DC based: Olher (auxiliary) diesseS, excejjtcancer and rtoeasesm^tehi^ vetlic!e(9fiNPed) drrtfestecdttwivdr) DC NOMS populalion 'Automobile mechanic": Pleura and peritoneal- Whiles: n=18,PMR=0.77 Blacks: n=2, PMR=1,54; p>0.05 Abbreviations: DC*deathcertificates, renumber, OR=oddsrat, PMR=proportionalraonaityratia,RR=reiatwerisk, StR=standatized irteidencerate, sMOR=sfendardizedmortaSyrat Confidencebounds =95%Clunlessotherwiseindicated CiD aI> I > TABLE 4 (continued) Studies o f mesotheiioma am ong m ale autom obile mechanics Author Study Populate Outcome Assessment Exposure assessment Comparison population Mesothelioma Results . Studies that focus ou mesothelioma McDonald ef ai 1970(50) McDonald and McDonald 1980 (57); (includes cases k m McDonaldet at 1970) Quebec Canaddeads from pfeural end pefifonal mesth!ima1959-67; n=165 : Quebec Canada 1959-72 and USA 1972. deaths from pleural and peritoneal mesothelioma in n=344 man tdentified through registered pathologists Identified through registered pathotogisls Ever employment from next-of-kin questionnaire Next-of-kin questionnaire Pitetafyaid secohdaiytoig cancer deaihsfromsame ^Kttegfsts (Tilcbberdsefrcase) : Secondary king cancer deaths fromsame pathologists (1:1) 'Brakelining installation' dases=2, confrols=0 'Garage workers" cases=11.controts=12 F. LADEN, M.J. STAMPFER. A.M. WALKER Ttaet al. 1983 (58) Connecticut USA deaths Connecticut Tumor frommesothefcma and Registry, confirmed by other pleural tumors, mates pathology review > 30 yrs, 1955-7?,n=136 tnesotheTtomas, 11 other pleural tumom .. . DCs ftr usual employment, City of Division of Health directories ter . Statistics of the Conneeficut employment histories State DepartmentgfHeaith 1,10,20,25.30,40, SOrvices.age 20-98 50yrabefore date of .. diagnosis Jobtitle: 'automobile mechanic' -cases not listed , ` Industry: 'Automobile repairand related 'Splices'- : Cdses=6, RR=0.65 (0.08^5.53) Malkeretal. 1985 Sweden, incident (59) mesotheliomas 1961-79, n=318 Nateal Swedish Cancer 1960 Census Registry General Swedishpopulation 'Mechanics and repairmen' S!R=2.4.p<0.01 Spulasela!. LACounty, CA, NYStete, Cancer Registry Next-of-kinInterviews Notspedfed 1985(60) mdECHtfieffOrras 197 ri2SB,:: 'V Spirtas el al. U County, CA, NYState, Cancer Registries and Next-of-kinInterviews Deaths fromcauses othe: 'Brake finingworkor repair' s 1994(61) (update ol Spirtas et si. m si Wotowteand RodelSpenjer Veterans Admin hospitals, mesotheliomas 1975-1980, n=2Q8 confirmed cases VACentral Office Hdfpitel setting Interview than cancer, respiratory disease, suicide, or violence, n=533 315lui^ cancer casesr.182 cases=33 (15.8%), controls=72 (13.5%) (at least 55%had additional overt asbestos exposure} 'Motorvahlds m ed ian' . ceae8*18, terig.cai6erC6nth?l#-i6, 1994(69) populBtion contro!s*12 Justes^^Unrverstfy cn Gissen.Gefitey^ rnesctheliotnasri=324 oo <s Abbreviations: D C = death cw fiteates, nm um ber, OR=odds ratio, PMR=proportionat mortality ratio, KR -relaft/e risk , SIR^stetiO ardizeJ incMeftce rate, sM OR=starxfame<l mortality rat oz Confidence bounds * 95% C l unless otherwise indicated z tr CQ3 OJ ------------------------------------- > ------------------------ ---------------------------- ------------------------------------ --------------------------------------- > ---------------------- 770-493-1734 TABLE 4 Studies o f mesothelioma among male automobile mechanics i Author Coggonetal 1995 (62) Teschte et at. 1997 (63) Study Population England and Wales, males, age 20-74, cancer of pleura (n=?,848). cancer of peritoneum (n=362), asbestosis (n=281) 1979-80,82-90 British Columbia, pleural .masothefoma, 1990-1892 (n=S1) Outcome Assessment DC British Columbia Cancer Agency Exposure assessment Comparison population DC General population Interviewwithpatient Randomseieoffonftm or next-of-kin provfnclatvotesrs lisi Mesothelioma Results 'Motor mechanics" pleural cancer n=12, PMR-46; peritonea! cancer n=3, PMR=88 "Vehlctemechanics - ever,employed"; CdsesaS, contas*20,OR=0.8 (0.2-23): brateMrig iSteEetfdnorrepair"; oasas=2, `vtefiiefg; .V LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS Cocco & Dosemed 1999 (64) A0udoetat.20OO (65) McDonald at si 2061 (66) Wong2QQ1 (70) 24 US States, peritoneal mesotheliomas 1984-92, n=408 males 2 provipcesln SpBin, Barcetorie, Carfe, pfeurat mesothelioma 1993-96, m=132 , ' United Kingdom: Surveillance of Worit-related mid Occupational Respiratory Disease (SWORD), bom 1943 or later, mesotheliomas 1990 96, n=115 mates Meta-anaiysls ofMcDonald rindMcDonald188Q, Tetaet 8t.l983,Sptrtasela). 1985, DC Hospital based SWORD records (see individual studies) DC Interview, occupation 6 months Deaths from non-malignant diseases Other hospital patients oCCUpeBors" cases=1, controls=15, OR*0:4 R & a ., ; , 'Not specified mechanics and repairers (occ)'as above OR=1.8(0.5-6.5) "Occupation, mechanics, motor vehicles' cases=3, ewtlrois*14 Occupational history fromMedical records, interview with practitioner Census 1980,1970,1980, 1990 Automobile mechanics not listed (see individual studies) (see iridMdua!studies) *AatIKWemecfiishics* (0.66-1:23) ; HemminMet al. 2003(6?) 19W ^drifeetil99T*r Agudoetal2000 Sweden, all persons bom after 1931 and their parents, peritoneal mesoteeftoma 1961-98, n=96 males Cancer registry Census 1960 Standard European population "Mechanics' - cases=6, SIR=0.86 (0.31-1.87) Abbreviations: D C = death ceiW tates, im g n b e r.' O R=odds ratio. PM R^proportional mortality rat. RR=relalrve risk, SIR *stand sitiiied iid d en ce rate, sM OR-stanrfartfized m ataSfy ratio Confidence bounds = 95% C l unless otherwise indicated ' Richard Lernen <--*k *Jw'VU : U1 W c ^ E l NUW f td W d - - n U N 56 F. LADEN, M.J. STAMPFER, A.M. WALKER r presented. All o f these studies separately were null employment as an automobile mechanic t l 6/. In an and the combined relative risk was 0.90 (95% Cl: additional study, 13 cases o f mesothelioma were 0.66-1.23). reported over a 5-year period among members of the International Association o f Machinists and Case Series and Case Reports Aerospace Workers. One case had been employed as an automobile mechanic P li. Because mesothelioma is rare--The Surveil We also identified two studies where lung- lance, Epidemiology and Risk Program (SEER) fiber analyses were performed on series of meso reports an age-adjusted incidence rate of 2.1 cases thelioma cases identified by occupation. Roggli per 100,000 males /? ! /--diagnoses are often and colleagues /78/ reported asbestos body counts described in case series or case reports. Such studies and levels o f commercial and noncommercial provide little useful information in assessing risk amphibole fibers (a form of asbestos consistently because the lack o f a comparison group precludes associated with mesothelioma) among 1445 cases, any quantitative estimate. Because of the rarity of including 51 with some work as an automobile the disease, however, exposure histories do provide mechanic, brake-repair worker, or brake-tine important insights into the potential sources of worker. The results o f the fiber burden analysis led , exposure. Therefore, for completeness, we include the authors to conclude that the tumors were these studies in our summary. unlikely to be due to exposure to automotive V Case series: We identified six studies in which brakes (chrysotile fibers). The same researchers --------- all mesothelioma cases in a population diagnosed also published a study focusing on 10 cases of during a specific time period were assessed for mesothelioma for which brake dust was the only their potential exposure to asbestos, including known source of asbestos exposure /79/. (Whether r- employment as an automobile mechanic. In a series o f 41 male patients diagnosed at the London these cases were included in the 2002 study is not clear.) Again the lung fiber analysis suggested that Hospital from 1915 to 1965, one case reported work as an automobile mechanic is unlikely to exposure to crocidolite asbestos through the manufacture o f brake linings 1121. One case out of contribute to the development of mesothelioma. Case reports: Three reports o f individual 4)3 identified by the Mesothelioma. Register in mesothelioma cases potentially associated with Great Britain from 1967 to 1968 was employed as exposure to asbestos through brake-repair work an automobile mechanic. Ship workers (n " 75), were published in the 1980s. Two were automobile asbestos factory workers (n = 39), and insulation workers (n = 13) were the most represented mechanics who experienced no other recorded potential exposure to asbestos /80-8I/. The other occupations /73/. Seventy cases o f mesothelioma case was an elevator mechanic who worked with were observed in the region of Nantes-St. Nazairre, elevator brake pads /82/. In a letter in response to France between 1956 and 1978 /74/. Four were this report, Kilpatrick /83/ argued that elevator employed as mechanics. In the Australian mechanics often work in buildings that are under Mesothelioma Surveillance Program, 1980-1985, going renovations and construction. Thus, primary 871 confirmed cases were Identified. Of these exposure might have been to amphibole from cases, eight were employed as automobile asbestos fire retardants sprayed into the building mechanics /75/, In a series o f 148 cases o f mesothelioma identified among peasants residing in rural parts of Turkey, one patient reported by other workmen, as opposed to chrysotile asbestos in brake pads. Without a quantitative comparison, the case reports are merely anecdotal. 81 k iI - E6k-Oii ueuiai p je q o ty ^ 1 : 8 0 bO 8 0 U1 WCHcJ'-ata NUW p w d - t - n u N LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS 57 DISCUSSION AND CONCLUSIONS In this paper, we have reviewed the analytical studies o f risks o f lung cancer or mesothelioma among automobile mechanics, plus all available case series and case reports. These studies represent thousands o f cases o f disease and hundreds o f thousands o f individuals who were followed. When examined in aggregate, the consistent pattern that emerges from this research indicates no increase in risk o f either lung cancer or mesothelioma in this occupational group. Many studies have important limitations, principally in the exposure assessment in most instances. In many studies, men were grouped into occupation based on one-time reports o f usual or last occupation on death certificates, medical records, or the census. Other studies included some assessment o f lifetime work history, but often it was not detailed. One consequence o f such inadequate exposure assessment is that the duration/^') of exposure was largely ignored; men who worked an entire lifetime as an automobile mechanic might ` > be grouped together with those who had worked in this occupation for only a short time. Furthermore, some automobile mechanics who were employed in the past might not have been identified because at the time o f reporting they were working in another occupation. Employment as an `a u to m o b ile /^ mechanic' also does not necessarily capture exposure to brakes or clutch repair work. Few of p g the studies actually included an assessment o f _ exposure to friction products. The length o f time spent repairing brakes,, the actual activities, the condition o f the brakes, and the implementation of :) safety measures all might influence a person's exposure. Grouping together workers with su c h /fT ^ different experiences leads to a misclassification of exposure, which may tend to attenuate relative risk estimates. Another potential source of m isclassification/^ o f exposure is the long latency o f lung cancer and mesothelioma. It is estimated that it takes at least 10 to 20 years after first exposure to a carcinogen for these diseases to develop. When duration and timing o f employment is not considered, essentially unexposed people are included within the exposed group, possibly leading to biased relative risks. The few studies that considered this latency, however, did not find increased risks. Potential confounding factors are inadequately evaluated in the majority o f these studies. Cigarette smoking, the key risk factor for lung cancer, was seldom assessed. Several studies indicate that smoking among brake mechanics was more common than in the comparison group, which would yield observed relative risks that are overestimates. Although nearly all studies reported relative risks near the null value, those that adjusted for smoking tended to yield relative risks closer to 1.0, as one might expect from confounding by smoking. In addition, some automobile mechanics may have been employed in other occupations or involved 'in other activities with overt exposures to asbestos potentially leading to an over-estimated relative risk. All these limitations also apply to evaluations, in many of these same studies, o f other occupations with known overt exposures to asbestos such as pipefitters, construction workers, and insulation workers. The expected elevated relative risks, however, were indeed observed for these other workers, implying that if an excess risk of lung cancer or mesothelioma had been present among automobile mechanics, these studies would have been able to detect it. Furthermore, the exposure to asbestos, in a form likely to be carcinogenic, was found to be quite small in industrial hygiene surveys o f brake repair work /IQ/. In addition, studies of workers tn factories that manufacture asbestos* containing brakes do not appear to be at higher risk for such cancers /U ; 84-88/. Thus, these observations, along with the weight of the evidence from the studies described here, supports the i-)0CALl'-) , fe- si uaujai p je q o ta B 8 I : 8 0 bO 8 0 rtow -->--*Jw'Vl.J : U i WtHcUbtd NIJW 7 0 0 5 -8 -A O N r r F. LADEN, M,J. STAMPFER. A.M. WALKER conclusion that automobile mechanics are not at elevated risk of lung cancer or mesothelioma from their occupation. ACKNOWLEDGMENTS The authors thank Sarah Corey, Rachel Adams, Jeanne Grol, and Jaime Hart for their help with the preparation o f the manuscript. This work was supported in part through a consultation agreement with the law firm o f Barnard, Mezzanotte, and Pinnie, who had no control over the content of this paper and did not see any draft of the text prior to the final proofs. Dr. Laden was also supported by a National Institute of Health National Research Service Award T32HL07427. REFERENCES 1. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological profile for asbestos. ' Atlanta, Georgia: U.S.A, Department of Health and Human Services, Public Health Service; 2001. 2. Hickish DE, Knight KL. Exposure to asbestos during brake maintenance. Ann Occup Hyg 1970; 13:17-2!. 3. Huncharek M. Brake mechanics, asbestos, and disease risk. Am J Forensic Med Pathol 1990; 11: 236-240. 4. Castleman Bl. Asbestos Disease in Brake Repair Workers, tn: Castleman B!, ed. Asbestos: Medical and Legal Aspects. 4,h Edition. Englewood Cliffs, New jersey, USA: Aspen Publishers, Inc.; 1996; 539-580. 5. Weir FW, Tolar G, Meraz LB. Characterization of vehicular brake service personnel exposure to airborne asbestos and particulate. Appl Occup Environ Hyg 2001 ; 16: 1139-46. 6. Williams RL, Muhlbaier JL. Asbestos brake emissions. Environ Res 1982; 29:70-82. 7. Rodesperger K, Jahn H, Brucke! B, Manke J, Paur R, Woitowitz H-J, Asbestos dust exposure during brake repair. Am J Ind Med 1986;10:6372. 8. Weir FW, Meraz LB. Morphological characteristics of asbestos fibers released during grinding and drilling of friction products. Appl Occup Environ H yg2001; 16: (147-1149. 9. Kauppinen T, Korhonen K. Exposure to asbestos during brake maintenance of automotive vehicles by different methods. Am Ind Hyg Assoc J 1987; 48: 499-504. 10. World Health Organization (WHO). Chrysotile Asbestos. Geneva, Switzerland: WHO; 1998. 11. Cheng W-N, Kong J. A retrospective mortality study of chrysotile asbestos products workers in Tianjin 1972-1987. Environ Res 1992: 59: 271-- 278. 12. Cheng VKI, O'Kelly FJ. Asbestos exposure in the motor vehicle repair and servicing industry in Hong Kong. J Soc Occup Med 1986; 36:104-106. 13. Leigh JP Occupations, cigarette smoking, am) lung cancer in the epidemiological follow-up to the NHANES 1 and the California Occupational Mortality Study. Bull NY Acad Med 1996; 75: 370-397. 14. National institute for Occupational Safety and Health (NIOSH), Division of Respiratory Disease Studies. Work-Related Lung Disease Surveillance Report 2002. Section 16: Smoking Prevalence by Industry and Occupation. Cinctnatti, Ohio, USA: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention; May 2003. 15. National Institute for Occupational Safety and Health (NIOSH). Letter to M J. Teta from J.T. Walker regarding inquiry about PMRs for auto mechanics, with attached data sheets. Cincinatti, Ohio, USA: U.S. Department of Health and Human Services, Public Health Service, NIOSH, 2002. 16. Carstensen JM, Pershagen G, Ekiund G. Smokingadjusted incidence of lung cancer among Swedish men in different occupations. Int J Epidemiol 1988; 17: 753-758. 17. Jarvholm B, Brisman J. Asbestos associated tumours in car mechanics. Br J Ind Med 1988; 45: 645-646. 18. Hansen ES. Mortality of auto mechanics. Scand J Work Environ Health 1989; 15; 43-46. 19. Schwartz E. Proportionate mortality ratio analysis of automobile mechanics and gasoline service station workers in New Hampshire, Am J Ind Med 1987; 12; 91-99. 20. Rushton L, Alderson MR, Nagarajah CR. Epidem- 02 V E A I - E B V - OAA U0 U1 0 -) p j e q o t y B 8 X : 8 0 VO 8 0 AOW W '_li^'VU ! * ; U i W S S80 NOW t ? 0 0 5 - 8 - r i O N LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS 59 r ologicai survey of maintenance workers in London Transport Executive bus garages and Chiswick Works. Br J lnd Med 1983; 40: 340-345. 2!. Gustavsson P, Plato N, Lidstrom E-B, Hogstedt C. Lung cancer and exposure to diesel exhaust among bus garage workers. Scand J Work Environ Health 1990; 16: 348-354. 22. Blair A, Walrath J, Rogot E. Mortality patterns among U.S, Veterans by occupation. 1. Cancer. J Nat) Cancer Inst 1985; 75; 1039-1047. 23. Hrubcc Z, Blair AE, Rogot E, Vaught J. Mortality risks by occupation among U.S. veterans of known smoking status NiH No. 92-3407. Bethesda, MD. USA: National Institutes of Health; 1992. 24. Emeriine PE, McKiever MF. Differential mortality from lung cancer by occupation. J Occup Med 1963; 5:283-290. 25. Milham S, Jr. Occupational Mortality in Washington State 1950-1971. Cincinnati, Ohio: USA: U.S. Dept, of Health, Education and Welfare; 1976. 26. Milham S, Ossiander E. Occupational mortality in Washington State 1950-1999; Washington State Department of Labor and Industries; 2001. 27. Andersen A, Barlow L, Engeland A, Kjaerhcim KL, ., Pukkala E. Work-related cancer in the Nordic countries. Scand J Work Environ Health 1999; 25 suppl 2: 1-116. 28. Menck HR, Henderson BE. Occupational differences in rates o f lung cancer. J Occup Med 1976; 18: 797-801. 29. Petersen GR, Milham S, Jr. Occupational Mortality in the State of California 1959-61. Bethesda, Maryland, USA; National Institute of Health; 1980 January 1980. Report No.: NO 1-CP33353. 30. Olsen JH, Jensen OM. Occupation and risk of cancer in Denmark: an analysis of 93,810 cancer cases, 1970-79. Scand J Work. Environ Health 1987; 13 Suppl 1:5-91. 31. Morrison SL. Occupational mortality in Scotland. Br J lnd Med 1957; 14: 130-132. 32. Pearce NE, Howard JK, Occupation, social class and male cancer mortality in New 2ealand, 1974 78. Int J Epidemiol 1986; IS: 456-462. 33. Minder C, Beer-Porizek V. Cancer mortality of Swiss men by occupation, 1979-1982. Scand J Work Environ Health 1992; 18; 1-27. 34. Lindsay JP, Stavraky K.M, Howe GP. The Canadian labour force ten percent sample study: cancer mortality among men, 1965-1979, J Occup Med 1993:35:408-414. 35. Firth HM, Cooke ICR, Herbison GP. Male cancer incidence by occupation: New Zealand, 1972 1984. IntJ Epidemiol 1996; 25: 14-21. 36. Schoenberg JB, Stemhagen A, Mason TJ, Patterson J, Bill J, Altman R. Occupation and lung cancer risk among New Jersey white males. J Natl Cancer Inst 1987; 79:13-21. 37. Vineis P, Thomas T, Hayes RB, Biot WJ, Mason TJ, Pickle LW, at al. Proportion of lung cancers in males, due to occupation, in different areas of the U.S.A. IntJ Cancer 1988; 42: 851-856. 38. Lerchen ML, Wiggins CL, Samet JM. Lung cancer and occupation in New Mexico. J Natl Cancer Inst 1987; 79:639-645. 39. Levin LI, Zheng W, Blot WJ, Gao Y-T, Fraumeni JF, Jr. Occupation and lung cancer in Shanghai: a case-control study. Br J lnd Med 1988; 45: 450 458. 40. Williams RR, StegensNL, Goldsmith JR. Associ ations of cancer site and type with occupation and industry from the Third National Cancer Survey ` Interview. J Natl'Cancer Inst 1977; 59: 1147-1156, 4L Swanson GM, Liri C-S, Bums PB. Diversity in the association between occupation and lung cancer among black and white men. Cancer Epidemiol Biomarkers Prev 1993; 2:313-320. 42. Benhamou S, Benhamou E, Flamant R. Occu pational risk factors o f lung cancer in a French case-control study. Br J lnd Med 1988; 45:231-233. 43. Spinel JJ, Gallagher RP, Band PR, Threlfall WJ, Raynor D, Schellekens H. Occupational associ ations among British Columbia cancer patients. Can J Public Health 1990; 81: 254-258. 44. Morabia A, Markowitz S, Garibaldi K, Wynder EL. Lung cancer and occupation: results of a multicentre case-control study. Br J lnd Med 1992; 49: 721-727. 45. Zahm SH, Brownson RC, Chang JC, Davis JR. Study of lung cancer histologic types, occupation, and smoking in Missouri. Am J lnd Med 1989; 15: 565-578. 46. Wang Q-S, Boffetta P, Parkin DM, ICogevinas M. Occupational risk factors for lung cancer in Tianjin, China. Am J lnd Med 1995; 28: 353-362. 47. Breslow L, Hoaglin L, Rasmussen G, Abrams HK. Occupations and cigarette smoking as factors in lung cancer. Am J Public Health 1954; 44: 171-181. 12 freI-E 6 fr-0 ueuia-j pjeyoiy *61 :8 0 60 80 aou C.C- J W Q : a i WUcE ' 8 0 NOW t z 0 0 5 - 8 - r i Q N r- 60 F. la d en , m .j . stam pfer, a m . w alker 48. Decou fie P, Stanislawczyk K. A Retrospective Survey of Cancer in Relation to Occupation. Cincinnati, Ohio, USA: U.S. Department of Health, Education, and Welfare; 1977. 49 De Stfani E, Kogevinas M, Boffetta P, Ronco A, Mendilaharsu M. Occupation and the risk o f lung cancer in Uruguay. Scand J Work Environ Health 1996; 22: 346-352. 50. Pezzono SM, Poletto L. Occupation and bistopathology of lung cancer: a case-control study in Rosario. Argentina. Am J ind Med 1999: 36: 437 443. 51. Dubrow R, Wegman D. Occupational charac teristics o f cancer victims in Massachusetts 1971 1973. Cincinatti, Ohio, USA: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health; 1984. 52. Milne KL, Sandler DP, Everson RB, Brown SM. Lung cancer and occupation in Alameda County: A death certificate case-control study. Am J Ind Med 1983; 4: 565-575. 53. Finkelstein M. Occupational associations with lung cancer in two Ontario cities. Am J Ind Med 1995; 27; 127-136. 54. Blot WJ, Fraumeni JF, Jr. Cancers of the lung and pleura. In: Schottenfeld D, Fraumeni JF, Jr., eds. Cancer Epidemiology And Prevention. 2TM1 Edition. Oxford, UK: Oxford University Press; 1996; 637 665. 55. Dubrow R, Wegman DH. Cancer and occupation in Massachusetts; a death certificate study. Am J Occup Med 1984; 6:207-230. 56. McDonald AD, Harper A, McDonald JC, el-Attar OA. Epidemiology of primary malignant msoth lial tumors in Canada. Cancer 1970; 26*. 914-919. 57. McDonald AD, McDonald JC. Malignant meso thelioma in North America. Cancer (980; 46: 1650-1656, 58. Teta MJ, Lewinsohn HC, Meigs JW, Vidone RA, Mowad LZ, Flannery JT. Mesothelioma in Connecticut, 1955-1977: Occupational and geo graphic associations. J Occup Med 1983; 25: 749 756. 59. Malker HS, McLaughlin JK, Weiner JA, Blot WJ, Ericsson JL. Peritoneal mesothelioma in the con struction industry in Sweden. J Occup Med 1987; 29:979-980. 60. Spinas R, Keehn R, Stark A, Beebe G, Dickson E. Mesothelioma risk related to occupational or other asbestos exposure: preliminary results from a case-control study. Am J Epidemiol 1985; 122: 518. 61. Spirtas R, Heineman EE, Bernstein L, Beebe GW, Keehn RJ, Stark A, et al. Malignant mesotheli oma: attribuiable risk o f asbestos exposure. Occup Environ Med 1994; 51:804-811. 62. Coggon D, Inskip H, Winter P, Pannett B. Differences in occupational mortality from pleural cancer, peritoneal cancer, and asbestosis, Occup Environ Med 1995; 52:775-777. 63. Teschke K, Morgan MS, Checkoway H. Franklin G, Spinetli JJ, van Belle G, et at. Mesothelioma surveillance to locate sources of exposure to asbestos. Can J Public Health 1997; 88: 163-168. 64. Cocco P, Dosemeci M. Peritoneal cancer and occupational exposure to asbestos: results from the application of a job-exposure matrix. Am J Ind Med 1999;35:9-14. 65. Agudo A, Ganzaiez CA, Bleda MJ, Ramirez J, Hernandez S, Lopez F, et al. Occupation and risk of malignant pleural mesothelioma: a case-control study in Spain. Am J Ind Med 2000; 37: 159-168. 66. McDonald JC, Edwards CW, Gibbs AR, Lloyd HM, Pooley FD, Ross DJ, et ai. Case-refereni survey of young adults with mesothelioma: 11. Occupational analyses. Ann Occup Hyg 2001 ; 45: 519-523. j 67. Hemminki K, Li X. Time trends and occupational risk factors for peritoneal mesothelioma in Sweden. J Occup Environ Med 2003; 45:451-455. 68. Malker HSR, McLaughlin JK, Malker BK, Stone BJ, Weiner JA, Erickson JLE, et al. Occupational risks for pleural mesothelioma in Sweden, 1961 79. J Natl Cancer Inst 1985; 74: 61-66. 69. Woitowitz H-J, Rodelsperger K. Mesothelioma among car mechanics. Ann Occup Hyg 1994; 38: 635-638. ' 70. Wong O. Malignant mesothelioma and asbestos exposure among auto mechanics: appraisal of scientific evidence. Regut Toxicol Pharmacol 2001; 34: 170-177. 71. Surveillance Epidemiology and End Results (SEER) Program {www.seer.cancw.gov). SEER* Slat Data base: Incidence--SEER 9 Regs Public-Use, Nov 2002 Sub (1973--2000): National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch; released April 2003 based on the November 2002 submission. 72. Newhouse ML, Thompson H, Mesothelioma of pleura and peritoneum following exposure to & 3 3 ` oi v ezi-eet-o zz uauiei pjegoty e 6 1 : 8 0 * 0 8 0 rtow OI Wd3 :80 NOW freOS-S-ON LUNG CANCER AND MESOTHELIOMA IN AUTO MECHANICS asbestos in the London area. Brit J hid Med 1965; 22:261-269. 73. Greenberg M, Lloyd Davies TA. Mesothelioma Register 1967-68. Br J Ind Med J974; 31: 9) -1 04. 74. de Lajartre M, de Lajartre AY, Mesothelioma on the coast of Brittany, France. Ann NY Acad Sci 1979; 330: 323-332. 75. Ferguson DA, Berry G, Jetihovsky T, Andreas SB, Rogers AJ, Chung Fung S, et ai. The Australian Mesothelioma Surveillance Program 1979-1985. Med J Australia 1987; 147: 166-171. 76. Baris Yl, Artvinli M, Sahin AA. Environmental mesothelioma in Turkey. Ann NY Acad Sci 1979; 330: 423-432. 77. Mancuso TF. Mesothelioma among machinists in railroad and other industries. Am J ind Med 1983; 4:501-513. 78. Roggli VL, Sharma A, Butnor KJ, Spom T, Vollmer RT. Malignant mesothelioma and occu pational exposure to asbestos: a clinicopathological correlation of 1445 cases. Ultrastruct Pathol 2002; 26: 55-65. 79. Butnor KJ, Spom TA, Roggli VL. Exposure to brake dust and malignant mesothelioma: a study of 10 cases with mineral fiber analyses. Ann Occup Hyg 2003; 47: 325-330. 80. Langer AM, McCaughey WT. Mesotheliomas in a brake repair worker (Letter). Lancer 1982; 2: 1101-1103. 81. Huncharek M, Muscat J, Capotorto JV. Pleural mesothelioma in a brake mechanic. Br J Ind Med 1989; 46: 69-71, 82. Huncharek M, Muscat J, Capotorto J. Pleural mesothelioma in a lift mechanic. Br J Ind Med 1989;46:500-501. 83. Kilpatrick DJ. Pleural mesothelioma in a lift mechanic (Letter). Br J Ind Med 1989; 46: 894. 84. Newhouse ML, Berry G, Skidmore JL. A mortality study of workers manufacturing friction materials with chrysotiie asbestos. Ann Occupy Hyg 1982 2 6 :899-909. ' 85. Berry G, Newhouse ML. Mortality of workers manufacturing friction materials using asbestos. Br J Ind Med 1983; 40: 1-7. 86. Newhouse ML, Sullivan KR, A mortality study of workers manufacturing friction materials: 1941 86. Br J Ind Med 1989; 46: 176-179. 87. McDonald AD, Fry JS, Woolley AJ, McDonald JC. Dust exposure and mortality in an American chrysotiie asbestos friction products plant Br J Ind Med 1984; 41:151-157. 88. Finkeistein MM. Mortality rates among employees potentially exposed to chrysotiie asbestos at two automotive parts factories. CMAJ 1989; 141: 125-130. 89. " Howe GR, Lindsay JP. A follow-up study of a ten-percent sample o f the Canadian labor force. 1. Cancer mortality in males, 1965-73. J Natl Cancer Inst 1983; 70: 37-44. ; 90. Bums PB, Swanson GM. The Occupational Cancer Incidence Surveillance Study (OCISS): Risk of lung cancer by usual occupation and industry in the Detroit Metropolitan Area. Am J Ind Med 1991; 19:655-671.