Document 4BmJKg62YGwxrEDDK7o8J2MV

7-12 Bnrzsh Journd df Indusrnal .Utdicinc 1990;47:742-746 Cancer risks in painters: study based on the New Zealand Cancer Registry Peter B Bcthwaire, Neil PearccO, James Fraser Abstract to occur for lung cancer,'-' a m t e leukaemia,2b multi- Painters a r e exposed to a range of complex ple n y e l o m a , ~ "bladder cancer,' and cancers of chemical mixtures which include organic the oesophagus, larynx, biliary system and liver,'' !' solvents and dye products with known carcin- skin, large bowel, and Also, increased rates ogenic and mutagenic potential. Trade pain- of non-malignant liver disease?.-heart disease,' and ters or those manufacturing paints and coat- respiratory and upper gastrointestinal tract disease' ings have increased rates of noa-malignant have been described, together with a form of chronic diseases and cancers; including lung cancer, toxic encephalopathy known as "painters syn- acute leukaemia, bladder cancer, and cancers drome.'" of the oesophagus, larynx, biliary system, liver, We report here on the findings of a case-control skin, and large bowel. A series of case-control study, based on the New Ze3land Cancer Registry, studies of painters, based on the New Zealand undertaken LO assess the risk of cancer for New Cancer Registry, a r e presented. These concer- Zealand men employed as painters and to generate ned 19904 male patients registered for the hyporheses for funher :esting. period 1980-4 who were aged 20 or older at the time of registration. For each cancer site studied, the registrants for all other cancer Methods sites formed the control group. Three cancer Our cohorr comprised male ?atients who were regis- sites were associated with work as a painte- tered with the New Zealand Cancer Registry for the namely, bladder cumours (odds ratio (OR) period 1 9 8 0 4 u ho were aged 20 or older at iegistra- 1.52, 9 5 O 6 confidence interval (9S0!, CI) 1.00- tion. ?he New Zealand Cancer Registry is popula3.31), kidney and other urothelial tumours (OR tion based and collects data from public and private 1.45,95O!, CI OSS-Z~50), and multiple myeloma hospitals rhroughouc the counrry. with further infor(OR 1.95, 9S0!,, CI 145-3.65). Risks for multiple macion obtained from death cerrificates and inciden- myeloma were greater among car or spray tal necropsy findings For tFe studv period under painters and signwrirers (OR 3.51) compared ionside:arion reg:srrarion was .,-irruaily complete. with construction and general painters (OR Occupation is defined as t l e patients' current or 1.80). N o increased risk was found for leuk- most recent occ- pari ion At The -:me of registration, aemia or for respiratory. biliary, skin, or gas- 2nd :s a d e d :n :he New Zeiiand Cancer Registry trointestinal cancers. using the New Zcaland Standard Classification of OccuFariuns NZSCO - , chi5 .s a nodification of ihe International Standard Classification of Persons engaged :n the manuBc:ure and Application of paint .ire potenriall:; zuposed :o a k e r o g e n e o u s group of substances. :n aarricuiar 3 ;vide range of organic solvencs and 2:.c L*ruuiucrs. Se-:eral studies have !nvesrigated - h e .xcucar:onal risks of work in the paincing [rads or ;n indusrr:cs .manufacturing pain: md coatings. Increased nsks have Seen shown dccupatiuns: Of :he :oca1 24 742 male regiscranrs, occupational informarion was coded for 19 904 SOo o and dara analysis was confined to these. Table I - h o w rhe disrr:bution of cancers in 'his group. The NZSCO :dencines cwo groups of painters. T h e firsr group consisrs of general painters. painter decorators. a n d tree1 and other construction painters. ?he second gnu? cJmpr:ses car painters. sprav painters. >ignur:tex. and other mclabsined painters. Department of Community Health. Wellington School of ilrdicine. Wrlline=ton. New Zealand P B Bem.rars. X Pssrce National Health Statistics Centre, Wellington, Yew Zealand J Frassr T h e cla5k:iicarion itoes nor d l o w for che !dentificarion -3f itorkers manufxruring painr and coarings. The design conssted of 3 series of case-control studies using the over311 group of i o 004 registrants, to examine t h e association between diverse cancer sires and work as 3 painter. F o r each cancer sire (see C m c i r risks In p m n r m : sru& based on :he .Ye% Z d a n d Cmcer Rqrsrry 743 table 1 ' the regisrrants for che ocher cancer sites Results formed the control group. For example. in tit Table 3 shows h e age adjusted ORs for the various analysis for cancers of the buczal cavity and p h a F n x cancer sites in S e w Zealand painters. T h e principal , International Ciassificacion of Di>csse ICD , Sch findings are :he rnereased relative risk for u r i n a p revision. :odes :-IO-I-W , che :ase group ionsisred cancsrs and m u l n p k myeloma. There was little of the 698 registrants wich L!is rype of cancer, and the evidence 6f an increased risk for other cancer sites; ocher 19 306 cancer registrants formed the control notably no excess nsk occurred for lung, gastrointes- group table 1:. T h e identical procedure was then tinal, liver and bilary tumours, or leukaemia. applied to the other diverse sites. T h e analysis was Table 3 shows char che excess risks pertain to both conducted using the 5:;;i;ticai .-Lrz$:;s bladder ;ICD?9th re-.%x, rnde !%) and to kidney Age adiusted odds ratios ORs were estimated by che and other urothelial tumours 'ICD. 9th revision, Mantel-Haenszel method" using 10 year age groups. code 189 I although che number of iases is small in Confidence intervals 95" C! were calculated using the 189 group. Risks for bladder and urinary tumours .Mierrinen's approximate method." were slightly higher in the car. spray. and signwriter Several previous case-control studies based on rhe paixtsr group OR = 1 32. S j " , CI 9.85-2-5' than S e w Zssland Cancer Rsgistry nave used "other among ionsrFxriun and general painters ,OR = c a r e r s " to form *e control group. T h e relevant 1.47. 9joQCi 1-02-2.13 . A larger difference was issues in this design are described '-T h e found for multiple myeloma, with t h e car, spray. and principle advantage of using other cancers as controls signwriter group having a greater risk :OR = 2.81, is che minimisation of information bias, rogether with W o o C1 0.7310.7't,han construcrion and general a reduc:ion in selection bias su-ing to incomplete painters ,OR = 1-80.9 j o 0CI 0.89-3.61). cancer registration. .A potential disadvantage is the Table 4 shows that che ORs for both the urological possibility of selection bias if the exposure under cancers and for multiple myeloma were particularly consideration is associated with an increased risk for increased in younger persons. other cancer sites. Such bias is likely to be minimal in occupational studies. however. unless exposure is assoc:ated with a common cancer '' Even in this Discussion situation the direccion of bias is towards h e null T h e finding of an increased risk of bladder cancer value Generally this bias cw. be minimised by among painters is consistent with previous studies. selecave exclusion of txposure related cancers from .inexcess of bladder rumours was noted in a propor- the control group. although this may in certain tionate mor:alin ,rud\ of artistic painters. .A his- siruacions acrually lead to an increase in n e t bias.-- torical cohorr stud\ ionducred on 19 l o Geneva trade painters. followed up from I070 until 1984, showed a standardised mortality ratio SMR' for painters of Table I Disrriburron oj;ancer cases. 6.v sire in .VIC Zsuidnd men aged 20 or older, 1 9 8 0 4 ?Ob 95'0 C I 97-38;' and a standardised incidence ratio of 171 9 j 0 0C I 101-172; for bladder m o u r s . s No increased risk x a s noted for kidney tumours in Tire ut idme* L o 01 mih ICD xcupuriun . J K ~ O ~ recorded ' 0 TOW iarzs this study Jensen er a1 in a case-control s n d y underraken in Copenhagen, found an increased nsk for bladder cancer m trades undertakInnp uamting Buccal cavity Oesophagus Stomach Colon Rectum Liver Gallbladder Pancreas Nasal ;avit\, s u m s Lamns Lung Soir :Issue sarcoma .Malignan1 melanoma Prostate T~SIJS Bladdrr Kidnei l*140 I50 151 153 I54 155 156 15- I ou IO1 I h2 1TI 172 I85 i YO IS8 I 80 583 82 385 83 1014 81 2043 A 2 :370 82 184 33 I20 87 571 82 4b 79' 303 SO 4224 84 --142 A 3 !I16 2135 73 100 92 012 i 2 542 '81 708 , O R = 1.j9,95OOC I 1.04-1.86)for 10yearsdurauon 463 1258 of employment, with a significant trend for duration 25w of employment." .A German case-control smdy 1670 2'3 138 examining lifestyle and occupational risk faaors for lower urinary rrac: cancer found an increased risk in 00-5 spray painters ' O R = 4.67, 959'0CI 2.10-10.39)." 58 179 Olsen and Jensen. in an analysis of a linked data set 5031 from r h e Cancrr Regsrry and the Danish Sup- I-! 1 HR plementsr). Pensions Fund found a significant 3347 increase in cancers of the lower urinary tract among 435 1259 mo persons employed in paint and glue manufacrure and among persons working in paint shops." An Brain 101 Non-HodgLin'5 lvmphoma 200.20: Hodgwin's lvmphoma 201 Muirlplc rnveluma 20 3 Lrukaemia Other -204-20Q Tiiral I 4Q-200 431 90 535 82 ! 7 0 01 205 83 534 82 liM 78 !QOO4 80 18 I 652 I80 355 b4Q 1088 247b2 increased risk of bladder cancer has also been des- cribed in painters in the United S t a ~ e s . ! ~ Increased risk of multiple myeloma has been associated with 3 number of occupational groups including wood workers," printers," workers in chemical manufacturing plants," and in 7.14 Bdwuatie, Paarcr, Frarcr Tdblz 3 .-Ip c l d p t z d O R A 95 '1 C I /or cancer in painters; .Vrw Z r a l m d Cancer Rrptsrr; 1 9 8 0 4 S i r z of .rlnccr ICD ; ~ r z r o ~ Exposed smcr Exposed aniroir Sucul d v l n Oesophagus Stomach Cdon Remm L.rer Gdlbiaddrr Pmcrcas Larvnx Ldng Sotr cissuc sarcoma Maliqnant meianoma Prosrare Testis Lrinarv, bladder Brain, nervous svstem Yon-Hodgkin s lvmphoma Hodgkin s lvrnphoma Multiple rnveloma Leukaemia Other 143-149 I50 151 I53 1% I55 I56 I57 141 I42 171 I72 185 IY6 188. 189 191,192 200.102 20 I 203 -10&-?09 10 7 IO 3 25 3 3 6 6 98 1 I4 43 5 38 10 9 ! IO i 37 343 380 005 201 5 1351 353 I17 565 397 265 141 1102 2392 394 315 345 526 169 185 348 314 OR 95% CI r d b k 3 Age ddjusted ORs 95'0 C I for painters by , u a s i e , .Vr.- Zruiand C m c e r R.e:str;., f 9 8 0 - 4 workers exposed to petroleum product^.'^ Several case-control studies have reponed small non-significant excess nsks of multiple myeloma among pamters.' lo ' I A cohort srudy based on Swedish death cerrificates of 416 paint indusrrv workers was undertaken by Lundberg and colleagues. .imong 96 workers who died, three deaths cccurred from multiple myeloma against O i de3~hs-xprcted.' Our study found no excess risk for ieuKaemia m o n q painrers, although chis has been :e;.orr:d by other workers, 3 study based on New Zsalmd de3ch certificates found an increased risk for all kuKaemias :n male painters SLUR2 3 4 9 4 I CI ! O 1- 40i . 3s did a Swedish case- ..control study where painters with exposure io ,".,-,-,.-_,..*-",.I_ _____p.a_nic_s_olvents exhibrced an OR of 13 195O, CI 2-551 ' Tine cxposure o f painters io chemical compounds IS complex and the composition of paints and glues IS likely to vary between countries. No specific study has been undertaken in New Zealand to document the various exposures m o n q different groups of painters. Painters are exposed co various dye and pigment products, with one of the most common exposures being to organic solvents.' I b Painters are exposed to organic solvents either 3s vapours or by direct skin application *rough the regular use of cleaning solvents. .A common jolvcnt used by house painters is white spirit Sroddard's solvent), which is a mixture of around 20'- aromatic and goo, aliphatic hydrocarbons. Paints and lacquers also contam a mixture of aromatic solvents ',G-Ct0olu)e,ne, and xylenes. Spray painters are also exposed to aerosols of paints, fillers. and catalysts.'- * Also, some painters use a range of adhesive m d degreasing agents that are heterogeneous solvent producrs commonly contain- ing rrichloroethl;lene. n-bexane, and methylethyl krone." '' T h e use loibenzene :s prohibited as a pure soivenr in most countries. but mal; Sr present in small quantities as m impurit:; :n orher scivents.' A range of organic solvencs has bern shown io have mutagenic ;.rOpCKleS but detailed :knowledge on the specific risks posed by indi,;idual solvenrs is .+lso, exposure to GipZGi; j < ~ ~ ~ hCasZ bZeen linked IO 3 number hacnaropoimc malignancies, in particular. 1l;mpnornas. ieukaemia. and multiple myeioma.. ii Cancer risks in painrers: srudj bused on the S e w Zealand Cuncer Rzgisrrv 745 T h e interpretation of nnding a greater risk for urological cancers and multiple myeloma in younger male painters is uncertain. This rnay reflect a change over time in the composition of paints and glue products. together with a change in the work practiccs of painters or in other environmental confounding factors. .\Iternatively mortality among younger painters may reduce those at risk of developing cancer at an older age. Further. the difference may be a function of the use of a relative measure of effect radier than an absolute measure.* T w o main limitations of a registry based study of this type are the lack ofvalid information on exposure and the lack of information on potential confounding factors such as cigarette smoking. Knowledge of only the most recent occupation of registrants provides a crude occupational histor!.. wi:h nc; information on spec:fic esposures. Secondary information on smoking behaviour is available from the 1981 census of population and dwellings, which showed that 364'0 of New Zealand male painters were regular cigarette smokers, compared with 3 8 O b of the total male workforce.'' T h e higher proportion of smoking behaviour among male painters rnay partially explain xcess of bladder cancer seen in this group. It is ri,.;worrhy, however, that no other cancer sites srrcmgly associated with cigarette smoking showed an s of cases among painters. Siemiatycki er al have as,cssed the degree of confounding bias related to smoking and socioeconomic state in estimates of occupational associations with bladder cancer. They concluded that relative risks in escess of 1.20 were unlikely to be attributable to confounding bias by these variables.'' Whether ir is prudent to generalise from these results is uncertain, although similar results were reported from Finnish"'and American'a studies. In summary, this cancer registry based case control study has examined the cancer experience of male painters, who are known from previous work to have an excess of certain cancer types. Painters are exposed to a wide range of complex mixtures that include a number of known carcinogenic and mutagenic chemicals including organic solvents and dye products. The findings are consistent with previous work showing an increased risk for urological and haematic cancers among painters. The study found an excess of cases of multiple myeloma, however. among painters whereas other workers found an excess of leukaemia. These associations point to priorities for further research into the occupational exposures and cancer experience of S e w Zealand painters. P Bethwaite is supported by the Medical Research Council of New Zealand under a Training Fe" .vship in epidemiology. James Fraser thanks the Director General of Health for permission to publish the data. Requests for reprints to. Peter B Bethwaite MB, Department of Community Health, Wellington School of.\ldicine. PO Bos 7613, Wellington, New Zealand. 1 Stockwell HG. .\latanoski G.\1. .Icase-iontrol study of lung cancer in painters. J 0c;up .!Id 1985::' 125-6 2 Englund .I.Cancer incidence among painters and some allied tradn. J Torrcol E n n r o n H z d r h 1980;6:1267-i3. .+I3 Dubrow R. Wegman DH. C n c e r and occupation in .Massa- chusetts- a death xrriticate study J Ind .\.fed 1984:6: 207-30 -1 Xlenck HR Henderson BE. Occupational differences in rates of lung cancer 1 O i ~ u p.lid 1976:18:707-80! 5 Gubcran E. Usel .\t. Raymond L. Tissor R. Sweetnam PM. Disaoiiir: . morralit\. and incidenie of ianccr among Geneva painrerr 2nd decrricians: an historical prospective study. Br .I Ina .$ISJ : V 9 . 4 6 l b 2 3 b Lndquisr R. Silsson B. Eklund G. Gahrron G Increased risk of developing acute leukemia aiter employment as a pamter. Cmcer 1087:60:1378-H4. 7 Pcarcc S . Howard JK Occuparion. social class and male cancer mortalit) in New Zealand 1 9 7 1 7 8 . Inr J Eprdemrol 1986; 15:45601 8 Lundberg i Morralitv and cancer incidence among Swedish paint industp workers with long term exposure to organic solvents. Scand J Work Enwrron Healrh 1986;12:108-113. 9 Cuzick J,De Stavola B. .Multiple myeloma- a case-control study. Br J Cmczr 1988:5':516-20. I @ Fnedman GD .Multiple myeloma: relation to propoxyphene and other drugs. radiation and occuoation. Inr J Eprdcmiol I986;1j:Gi-01 I 1 Adelstein A.M. Occupational mortality: cancer. Ann Occup Hyg 1972;15:5>7. 12 Jensen Ohl. Vahrendorf J: Knudsen JB, Sorensen BL. The Copenhagen case-reierent study on bladder cancer: risks among dnven, painters and cenain other occupations. Scand J Wora Emrron Healrh 1987:13.129-34. 1 3 C:aude 1. Kunze E. Frenuel-Beyme R, Paczkowski IC, Sch- nieder J, Scnuberr H. Life-style and occupational risk factors in cancer of the lower unnary tract. A m J Eprdmrol 1986;124:j~aa9. 14 Olsm JH, Jmsen OM. Occupation and cancer risk in Denmark: an analysis o i 93.810 cancer cases 1970-79. Scand J Work E m w a Health 1987;13 ,suppl 1):1-91. 15 Wynder EL, Onderdonk J, .Mantel N. an epidemiological investigation of cancer of the bladder. Cancer 1963;16: !388- I w7. 16 .MorganRW, Kaplan SD, Gaffey WR. A general mortality study of production workers in the paint and coatmgs manufacruring mdusrr)..J OccupMed 198123:13-21. 17 Dossing .M. Arlien-Soborg P, Petersen LAM,Ranek L. Liver damage associated with occupauonal exposure to organic solvents in house painters. Eur J CIrn IIWCSI 1983;13:151-7. 18 Vaidron HA. Solvents and rhe brain. B r J Ind Med 1986;43: 7%. 19 Foster FH. The New Zealand Cancer Regrsu?.. .Vew Zealand .Mtdicoi Journal 1977.86:341-3. 20 Deparunenr of Statistics. Xew Zealand rrandard classifiarion of occupanonr. Wellington: Department of Statistics, 1980. 21 International Labour Office. Intemrronal Standard Classrfica- :ion of Occupariai. Geneva: ILO, 1968. 22 World Health Organisation. .Manual of the tnrernarional srarrs- t ~ c o cl lassinurron ot drieaser, rnjunes and causes 01dearh. 9th revision. Geneva. WHO, 1977 13 Statistical .+nalvsis Svstem. Crer'r Guide. Starrsrrci. Cary NC: SAS Institute. 1985 24 .Mantel S , Haenszel W.Statistical aspects of the analysis of dam from rerrospecrive studies of disease. J .'Jar1 Cancer Imt 1959;:::; 1W 8 . 25 .Mierrinen OS. Estimation and estimability in case-conuol studies. ,din J Eprdzmrol 1976:103:226-35. 26 Smith AH, Pearce SE:Callas PW.Cancer case-control studies with other cancers as controls. Inr J Eptdemrol 1988;17: 29e-306. 27 Pearcc NE. Checkoway HA. Case-control studies with orher 746 Brrhwaile, Piarce, F r u d r &senses 3s controls: 2 r o b l m s of excluding exposure related diseases. .4m J Epidzmioi 1488;127:851-o. Pcarce NE, Smith AH. Howard JK, Sheppard RA, Giles HJ, Tcaque Ci.Cise-ionrrol studv ut' multiple myeloma and farming. Er J Canrzr 1986;54:493-500. Gallagher RP. Spincili JJ, Elwood J.M.Skippen DH. .Allergies md ~gncultural exposures JS nsk iacrors for mulriple rnveloma. B r J Cancir 1983;18:353-7. G n r o r KP, Blair A. Frnnlng md rnonalitv from multiple mveloma: 3 case-control smdv % w i t huse of death certificares. J .Vu11 C m c e r Insf 19&);72:151-5. Tollerud DJ, Srincon LA. Stone SJ,Tobacman JK, Blatrner W.1. .Mortalitv from multiple myeloma among North Carolina furniture workers. i .%ti Canczr l n r r 1985;74:i9%301. 3-1 .Milham S . Occupuriunal morrdiir:., in Woshmgron Starc 1950- 1971. Washington DC. US Government Pnnting Office, 1976. D H E W Pub1 No 76.. 33 Grecne .MH.Hoover RS. Eck RL, Fraumeni JF. Cancer mortalitv m o n g pnnting plant workers. Emiron Rer 1979; 2o:b6-73. 34 Blattner WA.Blair h,Mason TJ..Multiple rnveloma m che U S , 1950- 1 9 3 . Cuncir 1981;UI:2547-54. 35 Liner .US, Harlow SD, .LicLiughlin JK. .A case-control srudv of rnulriplc rnveloma in whim: ihronic mrigenic stimulation, occupation and drug use. C m c z r Res 1987:47:297Wl. 36 Kumai .\t, Kaizumi A, Sairo K, al. A nationwide survey on organic wivent iomponenr in various solvenr products: Pnn ?-Heterogenous products such as painrs, inks and adhesives. [ndusrnal Hzulrh 1983:21:135-G. 17 Inoue T. TJkeuchi Y. Hisanga N. d..A nationwide survey on organic julvent a m p o n e n t !n various mlvenr products: P a n I-Humugenuus 3roducrz ,ucn JS rhinnrrs. degreasers and rrqents. :ndusrrrili H e a m i98321:175-83. .\iereroin T. Jxvcntans H. Sossu .'A, V?inio H. Chromosome ~berritions:n 1vmpnoc::res o i workers exposed to rtvrcne. S u n d J %rR i?miron Hduirh 19782 suppi ):?59-04. Funes-C;avitu F. Kulmodin-Hedman E. Linrlsten J, Nordcnsk- old .\1. Chromosome Jberrations md sister-chromatid exchanges in workers inchemicallaboratoriesand a rotopnnr. Ing facto? and in children of women laboratov workers. L.inczr 1977:ii:322-5. 4 0 Fomi A. Pacinco F, Limonu A. Chromosomestudies in workers exposed ro benzene or toluene or both. Arch Emwon Heulth 107 I :22:37?-8. 41 Rinskv RA, Young RJ. Smith AB. Leukemia in benzene workers. Am I 1nd.Cfed 1981;2:217-45. 42 Checkowav H. DFilcosky T, Wolf P,Tyroler H. An evaluation of r i x Jssociations of leukemia and rubber industv solvent exposures. .-fm J Ind .Med 1984;3:23-9. 43 Olsson H, Brand1 L.Occupational exposure to organic solvenrs md Hodgkin's discme in man. Scand J Work Envrron Hzalrh +& 1980:6:302-5. Hardeil L. Ericksson .M, Lcrner P, Lundgren E. .Malignant lvmphoma and exposure to chemicals, cspcciallv organic sblvcnts, chlorophenols and phrnoxyaddc A ase-conrrol studv. Er J C m c c r 19RI;42:169-i6. 45 Rinsky RA, Smirh AB, Hornung R, 61 d.Benzeneand Ieulrmia: an cpidcmiological nsk assessment. .V Eryi J .Ued 1987; 3 16:1044-70. 46 Rothman KJ. .Lfodern rpidrmology. Boston: Little Brown and Co, 1986. 47 Deparunent of Statistics. .Vw Zzaland Cenrur of fopularion md Dweilingr / 981 : Bullerin Jn iigurcrrr smohng. Wellington: Deparunent of Statistics, 1983. 48 Siemiatvcki J, Wacholder S. Dewar R, Cardis E, Greenwood C, Richardson L. Degree of confounding bias relared to smoking, ethnic group and SES in estimates of the assoclarion between occupation and cancer. J Occup .Mea 1988;30@17-25. 49 Asp A . Confounding bv variable smoklng habits in differcnr occuparionai groups. Scand J Work Emiron Health 1982; 10:335-h. 50 Blair A. Hoar S, Walrach J. Comparison of c z d e and smokrng ~diustedS.MRs. J Occup .bled 1985;27:981--i. .Asce:ted 1 1 June :990 Destruction of manuscripts From I July 1985 articles -ubmitted for publication I will not be returned. Authors whose papers are , rejected will be advised of the decision and the rnanuscnpts will be kept under secuntv for three months to deal with any inquiries and then destroyed.