Document zowOrw1VgLEw92G04MpRX0XnB

I ! Multiple myeloma amobg @anis 9women: employment histoh and workplace exposures i .i ii Linda M.Pottern, Ellen F,Heineman,Jarger)H.,Ol/en, Edith Raffn, and Aaron Blair 4I i I i I i(Received3J March 1992;accepted in tct&dfo'onn 8 / & 1992) I i iI To investigate the rok oCemploymenthistory and wotkplafc &posures as ri k factorsfbr multiplemydoma histories from 1%4 and the most butwvha had an occupational titltc o d 4 as elevatedrisksof2 . h greater were observed for employment in: production of spinniqlwav- ing; other textile and plastics manufaptwrjng; and so@ rtrvices industries, E k v d , but nonsignificant risks were observed for possible and probable exposure$eix~bouat fumes, formaldehyde, wood dust, animals or animal products, andpesticides, The subngest asrocidtionwithmyelomaw a employ- Tment in the agriculturalindustry (OP = 15, CI r 0.8-2.8), h ' ever, the nubbtr of women who worked on family farms was unknown and could not be included in ihis,risk estimate; iI Key words: Agriculture. casecontrol study, Denmark, females, kdusrry, muldplemydoma, occupaD.on,c e d e r . Introduction iI 8 1 The ctiology of multiple myeloma W M )is largely unknown. A number of epidemiologic studies has shown an increased risk of myeloma among mcn employed in occupationsor industriesassociated wich agriculture, rubber manufacturing, and perroleurn refining,'Data on occupationalrisk factars for women with myeloma, howcver, have been limited. The Dm- ish computerized-linkage system2that combines data in rhe national CancerRegistry with population census Drs fortern,Ifnneman, andB L z t USA. Dr Ohrn IE wuh tln Danish ffrgrrrryat rbettmc uf rbr stnd). IS ' Ih Potfern, Omputronal StudiesSkzian, Envnonmcnut IISA. 1992 Rapid Cummunicatronr of Ulriorc)Lid , data bod indusvy $sr~ries from a national Pension PFun4 afforded us t3e oppomlit~to conduct a large otcu rdonal use-+ntrok swdy of mycbmo among woqn. I I Ma$&& and PT'dtwn 1 W d e n diagnosed/with M M between 1970 and 1984 I 418, Bc~besduM. D 2 W 2 , i anccr CawgraddContid.V d 3. looT 427 1 bI 1211Q/97 L. M.Porrem et at 14: BE: DUKE IJNIU. MED CTR LlBRFlRY $1 i !i i 001 (n = 1,010) were identified from the banish Cancer Registry (DCR),'a national computerized registry of all malignancies diagnosed in Denmark. It has been the hygienists estimated that 95-98 percent of all cancersdiagnosed in Denmark since 1943are included in the DCR.' Cases diagnosed with another cancer morethan three months prior to the diagnosis date of myeloma or born priorto April 1897 were excluded from the study. Four population-bad controlsper case {n = 4,940) . wcre selected from the Central Population Registry : (CPR): a computerized roster contdning personal Stlztisticai analysis , information (e.g.. vital status, date of death, occu- Maxibum lik,lihoc# procedures were u r d w com- pational title) on all Danish residents. The controls pute &mates of agt-adjustedoddsratios(OR)and 95 were matched to each case on year of bitch and gender, pqcebt confidcnccifitcrvals(Ct)asdtscribed by Gart.5 and had to be alive in the diagnosis y m of the case For &ch occupatiobal title, ORs were CalcuIad by (index date). Potential controls were excluded if they cornbring each in+iduaI occupational title with all were diagnosed with a malignancy more than three other! occupational: tides combined. ORs for each months prior to the index date. lated by comparing individuals Linkuge witb employment kforrndwn Industry informarjon since 1964 wasobtained by link- ing the civil restration number of each case and con- trol to the Supplementary Pension Fund Thc SPF, established in 1964, is compulsory for Danish with those ncvercmpioycdin that jjarticularin-ry. Only womenw i $ ~amindustry histoh were inclu#ed in the analysis d workphcc expo+res. ORsweFulcuLted for probabkad p ~ s - ible +posure to a qorkplace exposure using subjects unexposed to that sbbotanceas tbe comparison group. wagccarners, aged 18 to 66,who arc employed at l~rst 10hours pcr wcck by any one employq. Employment information such as name and address of the company, I I ! I i type ofindustry, and dace of employment at each com- pany is retained in the SPF files on employed individ- I I .- uals (other than self-employed), pensioners, and Ocngbntion * deceased workers. Since occupation wps not available Sixry/percent(607) the cases and 64 percent (2$%) from the SIT, the Central Population Registry was of th+ controls repdrted an occupational title on their used to obtain the occupational title that was r e p o r d tax rccords, Of thbe individuah, occupations of on the subjects` most recent annual income-tax form. pcrctjnt of the casesbnd 49pcrccnt of chccontrols were 'Thedate associated with this tax form was not avail- c o d 4 as cMrs/honjmakcr/ An additional 19percent able. Occupation and industry werecodedusinga five- of rhj cases and 22 percent of the controls were coded digit numberbased on the 1968 InternationalStandardd as !unskilfed' v$orkers, not-otherwise-spccificd Industrial Chssificdon of All Economic Ankrties.` (flC$, laving a tbtd of 166 cases and 765 controls Women who listed Mrs, Miss,or homemaker as heir withia specific occ pational title. Table I prcsents the occupational title were consided to be unskilled age-ddjusted ORs 29 individual or grouped occu- workers if they had an industrial history recorded in pati(nal titles. Onl9.womcnwith the occupational ti& thePension Fund. If they were not listed in the Pension `Mrsyhomemaker'jh`ho were not in the SPF had a sta- Fund, they were classified as `Mrshomemaker' and tistic& significant elevated risk of myeloma Of tbc were excluded from the industry and exposure analyses. Expuswe assessment For each subject in the SPF, the mast recent OCCUpacional title available from the Central Population Registry was combined witheach induetry code. These industrylaccupationa-codecombinations were distributed among b u t Danish industrial hygienists who asscssed exposureto 20 broadsubstancecategories and 27 specifk substances for each combination. Like- 428 C u ~ Gaurcc and b n s o l `dol>.!`?% assistants;and housefidldhelp. observed for farmefs, cutter/ 12110/97 14:03 DUKE UNIV. E D CTR LIWfiRY hdM5try According to c h c SPF wmpurer file$, 61 perceni of both the cascs and cGntrols werc not kmpluyed ia an iiidtutry sinco 1964, These subjects, ill addition to 28 eases and 71 controlkjwlio wcre in thc SPF bur had an crnplaycr for which &e bdusuy could no longer bc idtatificd, wcrc excluded from the analysis.T h e find pcpktion on which hindustry anafysiswas based was 363cases and 1,Si7 controls.The a-crage length of mplnymcnt since 1964 (the date of SPPstart-up) was yearsfor c m s aid 5.8 years lor conirols.I'hc m a n number of industrki ii, which cases and contrds warkcd was 2,t. Table 2 presents the age-adjusted ORs of M M associated with cmpkyncnc since 1464 in Rejected Tabk 1. nge-adjuJtedal~sntioJ~OR)oEm~ltmiyykdomr -for mmr rccent accupationd titk~among llanish w m e n .. Ocwprtioiirl title QSCS CMntrolr :OR (CIY (ti = 607) (n=2,596) -~ IS 0.9 (0.3-3.0) 7 0.0 - 8 2.2 (0.5-7.9) 49 1.0 (0.5-1.9) ' -8 I.b (OJ-6S) 4 a0 I 0.5 (0.Q-4.2) 35 B,Y tnil.7 7\ 020 1211om7 14:93 DUKE IJNIU. PED CTR LIBRFIRY 019 i Only one casc employ4 in thc plastics industry had an Let&, silica Juri, othcr inorganic dust, and ionizing occupational titlcand it wascodcdae `unskillcd,&Xi.' Ikadinion. f Of rhc 12 controls crnploycd in thc plastics indysrry, ! rho only non-%lucwllar' occupational codc w& f t w onc office assiatanL- EX~QEI(WS IDiscussion1 `rhhpopulati 11-bascd caw-controlptudy is onc d the first to evalui$ cmploymenr hirtory md workplacc Tablc 3 prcsenu Ihe age-adjustedORs associaredwith possibleand probable mpnsurc to 20 broad catcgbrics of wclrkphce srlbstances lor the 1,880 subjeccs ,wrrb iexposuresas t$sk factors for woincn with MM,Access IV the UaxisF record-hkagc sysicni yrovidcd OWI' 1,WO ncwly diagnuscd FWCS 4 .n,ycltirna aiw11g hdustry information. ORs for the 27 sprcific substances arc not prcscntd in thc table sirace expos+rc IO on+ a fLw ofthcsc subst~ncesincreased M M risk (dis- wonicli,froniitslo-84.Although only 36 p c e n t of 3tlicsccascs an M pcrccnt of the cnnunlr worked in an industry sin.. 1964, thee cmpfoynicnr figures arc 1cussed Ixlow). Among the suiubstaccswithin r h t broad comprabk t `thosein the UnittdStam b a d on 1970 caccpofics, & a d , but mnsignificiznr, ricks wcrc tabor +tativtiqs.b Thiv undcrscwa die difficulty in obsesrcd for b d i p s i b l c and probable cxposu~cto Nudying w&+Iac.cc cxyosurw among wymm, :exhaust fumes, for~naldrhydcw, ood dust, anirnSs or especjally fwjdipcascs that affcct thc cldcdy such as animal products, and pcsticidcs. Exposurc to psdinc, MM. a spccifjc exlmsure{nm shown), had elrvatcd but hon- Nu qrccili4 iiidur~ryor worhplacc cxyosurc w a c si~nificantrisksforpovsihlc(OK 1.8; 4 cascs, Iocon- bund to be st+tjcally associated with myeloma risk trois) a i d prdwblc c x p u r r (OH~ 2 . 1 ; 1 c q c , 2 rnrang l h i i i y womcii. Howcvcr, nonrignificantrivks controls). Expusurcto domestic anirnrla arid raw biiilk A i .3 or great+ were observedforsevcrr!mcupatinns, contributed rhc inoit m rhc incrcasd risk of r n y e l w t a a w x i a t c d with the C X ~ S L IcI c a t c g r y `mioialr or mimol products.'Hisks of one or les~were seen for p h - jtndusirier, an workplacc cxpoauirs which have bccn iug)ic$tcJ a9 r rk factors in prcvious studics. Employme:Iin agricultural activities such as farm- able and possible exposuic to ctjal and oil pmducrs, ng is a wrlijdocumcntcd risk f c r w for myeloma n q m i c solvcnis, pilitits, pl,x,tiL anatrrials, w l d c t i i ~ ~ rinong malcal( and has been suggested as a risk factor ! p :Table 3. Agc-idjUsrcd oddc raiios (OR)of inulriplc myeloma irocistcd with . o&blc and pmbrtrlc. tvorkplacc wcposurcs rrncmg h n k h wnmm4 (<I1`' (0.7-1.5) (hd-3.2) (0.7-1.3) -(0.4-1.6) (as-2.0) w-1.q @7-1.9} (0.l.i.S) (O.l.l.5) (W-I.O} tM4.4) (0.1.2.B) (0.6-&,2} (0.7-2.0) (oa.2.i) (hj-2-31 (U4-3.3) I ! I j (:asci + 2 \4 j. 6 i7 j1 i8 4 3 ;3 t0 :! $1 i .o ;0 1 17 26 js j ; 6 0 I'mbrblc mposxirr Cnnirols , O R , .,. -.- (Cl)' - 11 0.R (0.1.3.9) 11 1.6 (44-5.5) 46 3.6 (0.2-1.4) 43 0.7 (0..3-1.7) 2 2.1 (0.1-3a2) 55 1.0 (a4.n.i) 12 23 36 ,1 7 0 1.6 (0.i-5.3) 0.6 (0.1-23) -3.4 (0.1-I,>) Q.D 1-9 X (0.4--8.1) -19R 1.1 . ` (0.7.1.6) I 0.0 1 0.0, - . 78 1.0 (0.5- I:?) 102 1.2 (0.7-t.9) 7 . (I 0 -1.9 (0.4-8.4) 0-.0 - . . ... . .. 1 2 4@ e 7 13:02 DUKE IJNIIS. E D CTR LIBRARI 918 Cor wonicn in at Icast onc siudy.6 Tlic spccific exposures rcspoasi&lc for this association arc ncjt known, although suspcctcd agcnts include pcsticidcs.' e n ~ ~ ~ e c ~ h ~ ~ s t s , ~ ~ ~ " g r d~u~rndcustsit:k s , ~ ' a ~ d artirnats," In this study, Danish women employcd ib the agrictiituial industry had A 2.5-fold risk 04 myelr risk was olscrwd onu. E k v a d oonstgnificanr risks of mydoma were obrcrved among wvOmen with possihlc a d probable cxposurc to pcsiiuidi?sand animalsor snhd products. No risk was seen for the occupational tide `fanner' at listed on the tax rccord, however, the sclf- or hniiy. employed wornell who wlwkcd on family farms but did not list rhcir occupation IS farmers were mi iiicludcd in this aiiafyiiis. Thus, this rcport m y not accurately reflect Lhc agriculrurally relaced risk oi myeloma amon5 wumcn. Theelevatcd risk uf myelomaamongDanish women chest: women wcy probably hummakcrs. Home- eniploycd in thr Fpinninglwcaving and other tcxtile rpakbrs potentidlyiarc exposed to a numberofc h i - industries rcscmblcs findings from P Swedish study jals, )cxmlringfurneb, and olhcr iridow pcrllutants that !which mtd an increased iisk UF rrlyeioma among cou2 put tlicm a i $creased risk ofcanccs. "hep s i - femdctrxtilcworkersinthe wool indu~try.D'~clzellct bilit. uf a chnncc finding, howcver, cannot he auld 4P` observed 110 excess of myelonm, howcvcr, in a Gut. !Womcrr who ilisted thc occupational title Mrs, cohort study of C`male textile workers. orked in an indugtry at PsWve asewiatiooc also were found Cor women ere `unskilled workers, employed in food,restaurmt,hOtcI, and mtcminmcnr d risk ofmycfom. industries. Specific occupations such as foodtalcohdl e wggesrcd a pw'tive lohcco workers. and cookdkitchen assistants l i d clcvated risks. '13;s i6 k agrement with Severaf other a and tmployment ar: a gC C Q h t GtUdy Of COS- studies which showed increascd Ii s h fdr v a r i m s f d / and an Aust.mlianyopu- bcversgc-relared jnbs.``B Employment ;19 household s d y ' showed I#) help dso was associated with myeloma in our study. have r s p n c d excaw of 0 t h studies have ahown execs myeloma with r dyes,P*ay In our a ~ d y n, o employment as a maid or janitor.&'* was Q ~ S C ~ a&mong hair- Employmefit in tiiepla~ticsmanu~actuiingindustry e and six controls}. . was associated wish ntye101112rkk, dtbough based on tudy need to bc colisidered sndI ntlmlxrs. The plastics industry has not been s k g t d out previously as a workpiace with excess ings. Although the pcrccni- ' rd since 1964 was siruilar nrydoma b - d c n c e ; however. or~anksolvents and counmies, the number of otherliotcntial carcinogensare used coinrnonly.A few spccific indtlsrry wvs8 ftiriy studies have suggcsd a link between myeloma and ignificmt findingswas low occupational ~xposurcto benzene ".%trtourcxposurr fncd industries, however, . analysis rewaled no aswciattinnwith any organic sol- ii w a g /lot low for sbme'ofthE bi-oadrrindustria or for venw including bcnzchv (OK=0.8, for possiWc or mqny bf the exposus+. XnduswiaI Lmpioymeriihiator- probable exposure). Excess deaths due to mydoma ies',sin\c1964 were as& 011 records maintained in a have been reported for womm employed in an aircraft na/ionp.ideccntrai and were highly accurateu ~ d maintenancefacility, whn had exposures to avaricty of cokplitc. Industry i$ormacion prior IO 1964 was no! solvents." Although an association with organically ir\.aiLk$e; thus, indusjrzial workplace rnposurcs which high polymers (piastics and dastowm) has bccn covld not bc erahared. Thc observed in a popdation-bkd casccontrol study," cxposurc to plasric materials or nonforimldehyde or phenol-based adhcsives, in our study, did nut increase .*UthtehcAssmk ddmmy.t_*el~oxmpoas.OurUe tIodafwormdoa-lsdu.egh.gyedset tmhaaty, ui.nnlleilkue- COHar' .obs for asseging .. workplan: cxpos.urcs. The ence myelotmi risk. eXces~cs of mycluma among , exporub afitessmm$ provided by die industrid hy- GYI I CUUH~ Md chmlul. Vet 3. t9w ' 43 1 I L. M.Pottcm et al ! :. i, : 4gienistii were bzwd on their general knowledge chc codedindustries and occupauonsmher than interwiew or monitoring dau. This method d evaluation Yad a nuiirtc~c r i liini~iitms:(i) cxposutt miscl.assifiawon may have occurred if oceuprcion and industry +dcp Cmld not be link4 tpceifically to exposures; (ii) the specific chemicals used by certain industries ma)' not have bccn mident from the industry codes; and (iii) changes in exposuresowt time could notLc rddrascd. Despire these potential Iimi~titmt,he Danish link- age computchzcd s y m m was an ~xecllenrtewurdcfor identifyingfemalemyeloma cases and controlsand for investigating eqdoymtnt hiswy aid workplace uposurcs OS potential rkk factor$. Our findings of an incrcascd myeloma risk among Danish wqmen employed inagriculturaland textileindue&es arcFonsistcnt with yr&ously suggcstcd asso~iationrr,md deserve funher investigation, 2. JenrcnOM, Storm HlI,JcorcnIiH. Gnccr rcgisi+ua in Dcnrnark and the study of multipleprimary caqwrs, 1943-80. N C I Monogr 1985,m245-51. 3. OlsenJt?, JcnbenOM.Occupa6on and risk of caGer in Dcnrnark: An analysis d 93,810 txncer c a w , 19+-79. ScanJJ Work,Emiton H c 4 h 1987; 1311-91. 4. united Nations. IndcrPr to de lsrmurional Scrndurd fIndustrwCC:brss~camno AUE~nmt'cAiiim3iesN. ew Yorhi Statistical Office o rhe Uniud Nations, I%&. 5. Gdrt JJ. Point and interval orthation of thc cemmon odds ratio in thc romhnrtion d 2 X 2 rphlcs with fixed marginaia, Biomerrika 197k 5R 4714. 6. US Bureau of the CAMUSS.1.rirtinrl Abstract oj';he Uruted Srrtsc 198? (fttb e&). Wachingoe, DC:US Govcrnmcnt PrimtineOffice. 1916: 374. 7. a. 9. 10.