Document qmge90M8V2X31LJN9kjo2jq1x

13 April 1987 KmCTS or IXPOSUU to virtl CHLOKIDt: AN AMKSSMMT OF 1U CTIDDICZ tlehard Do11, FIS loaorar? Coaaultaoe, loporUl dactr Uoooreh rad Coaeor hi^aiaia.. and Clinical Trial. Balt, iadcllff. laiSaSTSLlK?"1010*7 ucc 000251 oowrans sumukt umtooucno* OCCUPATIONAL HAZAIDS Soureaa of Information US Study UE Study Canadian Study Italian Study Otter Soureaa lasarda of tueir * Aatloaareana Lung cancer Brain eanear Caneara of teooatoyoletle and lymphatic tlaauaa Nalaaoma Thyroid camcar Cancan of dlgaatlva tract Otter cancan Insacda of aoa-oallgnaat dlaoaaa Ctntoala of Ilror Ian1 malignant raoflratory dlaoaaa Cacdloanaenlar dlaoaaa Macoaaloa IAZABBS TO m CmiAL POWIAIION COKUMION uraxncxs Pat* No. 1 7 8 10 1l 13 13 17 18 24 23 27 28 29 30 30 30 31 31 32 32 34 34 38 41 42 ucc 000252 I TCJtUlY The results are reported of a review of: (t) Cho possible offacts of vinyl chloride (VC) oa tho personel health of ecu exposed to It by virtu* of thair oceupetion, othar thaa tba eerly effects of the vary high eoacaatratlooa to which was wart exposed whoa tha ladus try was first developed (unconsciousness, eardlac arrhythalas, aad tha characteristic 'vinyl chloride lllaasa') aad (11) tha carelaogeaie affects that night conceivably be observed in tha gaaaral population, aa a result of tha widespread distribution of VC as a pollutant. Tha possibility that VC eight act aa a teratogen or eight cause nutations la tha gate calls has not been exaalnad, as the little evidence that has been adduced relating to such possible affaeta has bane reviewed elsewhere and tha conclusion reached that no such affects have been danonatrated. Many groups of workers exposed to VC In the nsaufacture of vinyl chloride oooousr (VCM) or polyvinyl chloride (PVC) have been studied sloes the carcinogenic potential of VC was first recognised. The results have shown that occupational exposure can cause saglacareona of the liver and acne have suggested that It any cause several othar types of cancer as well. Whether this is so or set coo he decided oely by exoelnlng ell the evideace sad. la pertleolar, tha eenhlaad melts of those stadias that losluds e sohatsetlsl proportion of ohoorvstlooo oa non aero thee 23 years after their first exposera sad cover c long oaaogh period for aore then 10 par cant of tho npU^u to hevo boon cxpoctod to die. The malts nf foar studios can ha aaafally ecahlaod in this any: eaanly, two actional etodlcc la tha Old end the OK end too atodloc of caployaoo la woo plant la Canada aad two pleats la Italy. The results sf other stadias la the federal lspohllc of Camay, lorvey, Sweden, Italy, Prance, sad Japan esc ha aasd only to provide sapplsasatsty lafomtloa. The assy earlier reports sf exposed vertera la tha Old aad tha OK relate to ucc 000253 2 an cowered more completely ia the two recent actional ctudlee and their raaolta nerve sal? aa aourcee of hypothaaaa. Minor crlelelaaa eaa be made of throe of tha four meet uaaful acodlaa. They do aot eerlouely offact cha value of tha raaolta, except that allowance haa to ha aada for the fallen to determine tha eauaa of 6.3X of tha deatha neordad la cha 01 atedy. Thraa of tha atudlaa uaa oatloaal rataa to aatlmaca the uiabere of deatha that might have bean axpaetad to occur la tha abeeoee of may apodal occupational hatard and tha fourth (Canadian) needy uaaa rataa for tha province 1m which tha plant wma atcoated. It meat, therefore, be kept la mind that the rataa oaad may not hawe beam wholly appropriate for tha localltlea la which the plento ware altoatad. Thla la potamclally Important for tha Ot atedy which cowered workara in 37 plenta, 22 of which warm altoatod la tha eoeth of the country. The other loco Information atudlaa are, for tha aoet part, open to more aarloua crltldaa and the walaa of each amt of retulta aaada to be aaaaaaod aeperately la ralatloa to each dlaaaaa. Tha combined raaolta of tha four principal acodlaa ahow that tha atandardlaad mortality ratio (M), raflactlad cha ratio botween the ntaibera of deatha ahaarwad and tbaaa expected la tha aboeaca of aa occupational hasard aeltlpllad by 100, haa boons (1) 77 for accidence and other wtolaaca, (11) M far dlaaaaea other than cancer, and (111) 102 for caaeera other than aameer of tha Hear. All thaaa rowelta arm what night be entldpetod far aa Induetry deweld of any apadflc occupational baserd. Tha lorn ratio far dlaecaea other than cancer raflacta tha 'healthy worker affect', which raeelta from the aelection ytecaaa that laawltably excludaa acme of the lace healthy maabara of tha pepalatloa from iaduotrlal employment cad ia conpatlbla with a higher ratio for cancer aa tha mortality from concur ia aot normally oubjeet to amah aa affect, apart from cha firat two or thraa yearn laaedlately follawlag the aurt of employment. ucc 000254 3 _ Tha mortality from cancer of cha liver m nearly seven tlaaa that, enpected. hoot of tha 31 excess deaths sore knouo to ha due to angio sarcoma, eves though this diagnosis u not recorded on tha daath cartlflcata. Tha axcaaa corresponds alnost exactly with tha *9 daatha duo to angiosarcoma reported to tha International Kaglatar of Angiosarcoma Caaaa aa occurring la soployssa of tha plants concerned daring tha parloda they vara hope uadar observation. All tha non ahe developed tha dlaaaaa ara likely to have haaa exposed to eaacoatratiaoa of VC of several hundred ppo or oora. Throa othar typaa of eaocar, which, it aaa aaggaatad aooa aftar tha discovery of tho carcinogenicity of VC, might ho eaaood hy VC, ara eaaeara of tha long, hrala, and lymphatic aod haaoatopolatlc ayotaaa. Tho eoablaad data for tha mortality from respiratory eaocar fall to oopport tha hypothaala regarding lg cancer (SM 97). Ughar ration for loag eaocar harm, however, haaa obaarood eomolatamtly la tha ombgroopa of non la ohoo tho offoot of am occupational haaard mould ha moat likely to ha aooa (that la, man employed for more than 10 yuan, anpoaad to higher thau average concentrations, or oboeread nore than 20 yoara after flrat oapoaura). la too .of tho eupplancataty otudlca it uaa aloe noted that tha mortality from lmg cancer ana specifically increased la tha moat heavily anpoaad. Tho ronblaad data aha* small accesses lm tha mortality from cancers of tha hrala and of tho lymphatic and haematopoietic cystoma; hut tho accesses ara not statistically elgaiflcaat and thorn la nothing to suggest that they ara occupational la origin. As omcaptlam la tho observation of aa locreaaod mortality from cantors of tho lymphatic and haematopoietic eyotw la tha eupplmacatary study from tha federal tapuhlle of Oaimauy. Too 'typos of cancer maro roportad to ha is oscoao la tha Rorvoglan studyt namely, thyroid eaocar and me1asms. Tha slgulflcaaca of this finding la difficult to assess because vary little Information about those ucc 000255 4 cancer* i provided by other atudloa. Suggeotloaa that VC Bight cauae cancera of ch dlg*dv tract have failed to toko account of tho contribution of aagloaerceaa of tho lieor. Vhoa thlo dlaaaao lo excluded, tho aortallty free dlgoatlea troet eoncor la found to hoot boon beloa average (S* for tho four principal atudloa, 82). * Moll axceaa aortallty from tha bacorogoaaouo group of 'otbar caacora' la tho combined roaulta of tho four principal atudloa la atatlatlcally aorglnally algalflcaat (S3 daatha agalaat 63.25 expected, p<0.03). tone of tha exceaa la llfcoly to have boaa duo to tha ulaelaaelfleatloa of angloaercoaaa aa aocoadary caacora of tho Hour or carelaooatoala, alto unknown* Throo aou-aallgaaot cauaoa of daath fcevo rogulrod apodal anamin ation: elrrhoala of tha. liver bocauaa of tho daaago to tho Hour that occurred aa port of tho 'rlayl chlorlda lllaeso', aoa aallgaaat roaplratory dlaeaac bocauoo of tho chaagoa ta laag fraction aad ^roy appaaraaco oboorvad la aoa expoaed to TfC duct, aad ayoeardlal lafaretloa from analogy with tho affect of other holottaatod hydrocarhoaa aad bocauoo of aoae obaarvatlone ou twedlah PVC fahrlcatora. far froa holag reload, tha aortallty froa elrrhoala of tha liver uaa loaa than avpacted la tha throo prlaclpal atudloa aad la oaa of tha two oupplaaaatary etadloe uhleh gave aaparata flguroa far tha dlaaaao (SMta of 69 baaed aa 66 daatha aad af 62 baaed aa IS d\aatha) uhlla la tha other oupplaaaatary otudy tho lacroaoo aoa trivial* Bata far ayoeardlal lafarctlaa here aat been reportad aaparatoly, but ayoeardlal lafaretloa Mcuati far aaat af tha daatha attributed to lachaeale heart dlaaaao aad thora lo aa arldaaea that either tachaaale heart dlaaaao or cardleeaaeular dlaaaao aa a uholo uaa unduly twon (SMU af 90 aad 92) or ralatad to occupational oapaaura. ucc 000256 s The data (or the third escegory of aon-maiignant disease (nonmalignant respiratory disease) it< confusing, because the m largo national stadias giro conflicting results. The combined data for tho four principal studios show tho low mortality that Is commonly found in healthy Industrial populations (SMI SO). This, however, hides an Increased mortality from chronic obstructive lung disease (SMI 120) which Includes emphysema and Is due to e grossly Increased mortality attributed to emphysema In the 0S study (SMI 193). The corresponding mortality in the UK study, which was preferentially described as bronchitis. Is less than expected (SMI 82) as la the mortality from pneumonia (SMI 30) and eeher respiratory diseases (SMI 48) In the OS study. There Is no consistent evidence that the mortality from amphysou or bromchltlo was specifically occupational and it seams possible that the reported excess in the 1)S study la am artefact due to nosological difficulties In the use of the 7eh revision of the International Classification of Diseases. Ievlev of the massive data now available relating to the long term health of men occuaptlonally exposed to VC leads to tun clear conclusions. Plrat the mam have experienced a specific hasard of the normally extremely rare angiosarcoma of the liver. Approximately 1 in 333 of che men exposed in the 49 plants has died of Che disease and approximately 2X of observed deaths have haem attributed to it. In the emnrse of time the numbers of eases of angiosarcoma must be expected to be Increased 2 to 3 tines. Secondly, the mortality from all ether censes has beam typical of that la normally healthy industrial uothara* If nay ocher hasard has existed Its effect has boom small. The data provide mo reason to think that any hasard other than one of cancer hea boom overlooked. It la, however, still difficult to decide whether VC dees produce risks of cancer in sites ethor then the liver chat are email compared to those due to eon occupational causes, but vhleh ucc 000257 6 light, la total, cottM alioat as naay deaths as angiosarcoma of eha liver. Than is too little avidities sichar to eonfln or refute eha suggestion that VC light eausa aalanoM or eaacars of tha thyroid, brain, and lyuphatle aid haematopoietic systems. Rone of tha nail excesses that have boon recorded point specifically to an occupational hasard, apart from that attributable to eaacars of tha lyiphatle and haematopoietic system in the Carman study, and lost are likely to be the sort of ehanco offset that la certain to be observed when many types of cancer are examined in nany different stadias. The lack of any increased inrtaliey from lung cancer in the combined results of the four principal studies does not eielade the possibility that there may hare been a small occupational hasard of developing the disease, as geographical variations in the incidence of the disease throw doubt on the validity of using national rates for estimating the expected embers of deaths. The greater mortality in groups of workers who would he more likely to show as ocewpational hasard than in other groups, suggests that a small hasard nay have existed. The evidence is, however, weak and the existence of a hasard Is not proven* Clearer answers to some of the gnsstiono that have been posed in this review might he obtained if the various groups of investigators eeuld present their results is nore appropriate and comparable soys. AO VC is a dotages in laboratory experiments and a proven human carcinogen, the nlnete donee that have escaped into the general environment as pollutants mast ha preswod to have censed comparably nlnete risks to tha general public, he ouch risk could possibly ha detected, other than one of angiosarcoma of the liver which is normally an estmely rare disease* Several surveys have sought evidence of the existence of such as offset, end suggestive evidence that such an effect may have occurred ct c time when environmental pollution vno mwah greater than it is now has been found in one. inBOOOCTXOR The inhalation of large Mount* of vinyl chloride (VC) hot boon recognised to bo potentially hacardous for many yoan. Concentrations of tho order of 10,000 ppa in air Induce uaconociousneee and cardiac arrhythnia, while prolonged exposure to eoaeontratloea on order of aagaltude lower have been liable to cause a specific pathological syudrone. This "vinyl chloride illness" has been characterised by four cardinal eigne: nasely, enlargowent of the liver and spleen with a specific histologieal appearance, patchy Infiltration of the akin reseabling telerodeme, bony ehaagee la tho tips of the fingers described as aero-ostoolysis, end peripheral circulatory ehaagee identical with the classical picture of gaynaud's dleease. Those pathological reactions nay occur singly or together and nay possibly ho aceonpoaied by other less characteristic effects They eaa, however, bo csnplotely avoided by ensuring that esposara never exceeds a few hundred ppu, to which level exposures warn gonerally redacad in the aid 19d0s. One other serious effect has, however, boon observed that nay not be avoidable la the sane relatively easy any: sanely, tho production of aagloaarceoa of tho liver* St neat, indeed, ho pmimif that sene risk of developing tho dlseaao will parsiet fvan exposure to doses that are even lower than tho current industrial levels of 3 ppa or lose, as VC has been sheen to ost as a outages (International Agency for laoearch on Cancer, 1979) and it cannot he asstaed that a threshold exists below which no carcinogenic, risk persists. Msreaver, the possibility has to he considered that TC any canoe sene cancers other than aagloearesna of the liver, pertly bacawao laboratory stadias hsvo shewn that it dees this In aninsl experlneata and partly because the Initial studies, ihlch denoostrated the production of angloearcons of the liver in Man, were inadequate la else to exelode a notarial increase in the risk of cancer la cannon sites, such os IJCC 000259 A 8 the lung ad large bowel. Slaea no threshold does can ho postulated It also follows that sows cancer* osy have beea produced In the general public by the enall Mounts that have escaped Into the general esvlreuMnt. Consideration also nseds to be given to the possibility that exposure to vinyl chloride acting over a long period My have noxious effects on Han that could not easily be Men In anlMl experlaents (by, for exsnple, producing chronic respiratory disease) and, as it is a outages, there Is also a possibility that it My act as a teratogen and cause congenital MlfotMtions in the offspring. In this review I have not examined the possibility chat VC acts as a teratogen Mr that it eauses untstloM in germ calls, as there is too little serious evidence to require it. Reviews carried cot for sectloM of the Industry by Downs at al. (1*77) and by MecMahoo (1977) concluded that the few reports of positive effects could not be substantiated and no addltloMl evidence was found in a alsilar later review by Barr (1981), apart from a report that enbryoa wore abeorbed and skeletal ossification produced when pregnant rats were exposed to doses appreciably less thsn those that had boon used by other workers without any such effects being observed. The report of absorbed Mbryuo (Mtrksvs at al.. 1*71) could n t bo evalMted thoroughly, however, m the experiment waa inadequately described. X have, therefore, examined only the possible effects on the personal health of sen occupstlOMlly expoend to VC other chan on their reproductive capacity, and the carcinogenic effects that night conceivably be observed in the gsMral population as a result of the widespread distribution of VC M a pollutant. OCCUVAlKMdL IdZOM Ussy studios of uorkers exposed to VC in the uenufaeture of vinyl chloride nononsr (VO0 and polyvinyl chloride (WC) hove boon undertaken ucc 000260 9 lac* lc mi first found that VC could causa cancer la animals (Viola at al., 1971; Nsltoai at al., 1974) and la Man (Creech aad Johnson, 1974). These have confirmed thft exposure causes a haaard of angiosarcoma of the liver aad, la several Instances, have shown excess incidence or aortality rates from other diseases, which were conventionally statistically significant. Conventional toots of statistical significance are, howevar, designed to help answer single questions defined beforehand and several findings that night be expected to occur by chance alone once in (say) 20 tines oust be expected to occur if donnas of rates are examined in each of several sets of Independent data. He have, therefore, been faced with the problem of deciding whether the excess rates that have been observed In Individual studies are due to occupational haaards or to the vagaries of chance. This problem can be solved la port by examining the overall results, when the date from comparable studies are smedi that is, by adding together the numbers of deaths observed aad expected la each study, sad comparing the totals* This doom not require the eemaptioe that the exposures have been the same in each study, any more them it does for each individual whom the results of each study are considered alone. What It dean require in that each exposed population should hove been observed over a period when its members were at rlah of developing disease (if a genuine haaard existed) sad that la oach study the control population from which the expected sobers of deaths mere derived won appropriate (that la, at the same risk of developing disease an the exposed population would have boom la the absence of exposure), this dess not introduce any new complexity, am hath these conditions era, of course, also required if correct conclusions are to be drawn from the results of tha individual studies *em they era examined on their sun. ucc 000261 i 10 Sources Of Infatuation Pour studies oust these regalresents: naaely, the two large national surveys reported by Jooe* (1986) for tha OK and by Eavlronentsl Haaleh Aaaoelataa (1986) for the USA, and studies of Individual plant* la Canada raportad by Theriault sad Allard (1981) aad la Italy raportad (a* a part of a aatloaal study) by Balll at al. (1986). All four laeluda observations on aan aora than 29 yaars aftar thalr first exposure and tha expected nuabar of daatba la, la aseh east, grantor than 10 par eaat of tha total auabar of oBployaos ladlestlag a long average period at risk. Earlier observations oa UK and USA aeployees raportad by Moasoa at al. (1979), Taborahaw and Caffay (1974), Dock at al. (1979), llcholaoa e al. (1979), Ott at al. (1979), Uexwaller at al. (1976), Vox aad Oolllar (1977), Bgultable Envlroneantal loalth (1978), lafflar at al. (1979), aad Cooper (1981) are subsnad la tha aatloaal surveys aad new serve only as soureeo of hypotheses aad to provide seas detailed laforaatloa aot Included la tha aatloaal reports. Studios of Sanaa workers (Cruiser at al.. 1982), Xorwoglan workers (geldaas at al.. 1984), Swedish workers (lyrsa at al., 1976), Vreach workars (Pierre jt al.. 1979 aad taplaaehs at al., 1987), Japanese workers (Masada, 1979 aad Kakaaura, 1981) sad son other Italian workars (Belli e aj. 1988) provide seas sopplnsatary laforaatloa but, in general, the perleds of sbsorvatlso have set bees leas eaoagb far useful epldeadeleglaal data to bo sbtslaad about diseases that are unlikely to eecwr vlthla 20 years of first exposure, or they report only selected results which ate difficult to interpret as saly escess rates read to have beau selected. Studies of ffC fabricators are act lacloied as the workars will have had each lees exposure to VC thaa those saplsysd la the asoufactors of VOf or PVC aad aay occupational hasard ts which they asp have been exposed Is ears likely to have bees produced by PVC dost. ucc 000262 n as -Study Thu study esrrlsd out by Cnvlrounantsl Health Associate* (1986) on bohalf of thu OS Chuoical Manufacturers Aaaoclutlou la tha largest and uoat Informative. It eovers 10.173 aan who had workad In 37 plants owned by 17 coapaales - 1,214 in 11 plants that produced only VCM, 6,48 In IS plants that producad only FVC, 933 In 3 plants that producad both, and 1,176 In 5 plants that producad hoaopolyaers and eopolyaars, with or without VCM or PVC. Tvonty-two of tha plants wars In tha Southam part of tha country, 14 la tha North aastarn or North control parts, and oaa In tha vast. Nan warn lacludad if thay had baan exposed to VC for at laaat a yaar bafora 31 Dacsabar 1972 and had baan employed la 1942 ar subsequently (tha first yaar depending on tha data tha plant bagan making or using VC and tha aarllast data that parnoanal rocorda wars demand to ba complete, whichever was tbs latar). Individuals aha ant thasa crltarla aura Identified fro* company racordn by conpany parnoanal. laelal charnctarlstlcs warn known only for ltd aaa, 97 par cant of. whoa wara whita and It wan prastand, far tha pwrpoaa of estimating tha mtfbsr of expected daatha, that all 10,173 aan aaro whlto. Pollaw up aaa obtalnad froa plant aad Social. Saeurlty adalnlstratlon Xacorda and (for aaa who dlod aftor 1979) froa tha National South Indan. Plow plants did not collaborata In tha aztansloa of Coopar'a (1911) aarllar atody aad tha 939 oaployaaa la thasa plants who wara known to ba all** on 31 Dacaabar 1972 wars not follooad any fnrthar* Par tha root, follow-up wan attaaptod to daath or 31 Docaabar 19S2, whichever was tha aarllar. On this basis 92.7Z of tha aan warn swccassfully traaod. Thoaa who wara untracad wara axaludod froa tha last data of contact, which was usually tha data whoa aaployaant coasad. Almost half tha aaa (461) wara first Mplsymd bafora 1933. A largo ucc 000263 12 proportion ma, therefor*, obearvad aora than 25 yaara aftar flrat tzpoaura (and la eaay caaaa for aora than 50 yaara) ahon dlaaaaaa with a long latant parlod night ha expected to ha aaaa. Short tarn workers had haan excluded fron tha cohort and aoat of cha ana had coatlnuad la aaployaaat for uany years: tao-thlrda vara aaployad for 10 yaara or aora and tha average duration of aaployaaat me 16 yaara. Fifteen hundred and thirty ala aan vara found to have dlad. In 1,439 eaaaa, tha cauoa of death uaa obtained froa tha death certificate, but no eauaa ma obtained for 97 (6.3Z). The nrnbara of daatha aspaeted froa each of M cauaaa or groups of cauoaa mra obtained hr aultlplylng tha person-years at rlak by cha dlaeaaa-apeclfle national rataa for nhlto aaloa, for tha corroapoodlag age groupo and guloguannla. Thin laportant atudy la open to throe alaor eritlcleao, which are unlikely to have had any notorial offacta on tha rooalta. flrat, tha llata of aaploynea were coapllod by eoapaay peraooael fron eaapany roeorda without any ladepondont chock. Secondly, the coemption that all the aaployaaa were white will have oaaand the expected dontho te have been very ellghtly undemetlaatad, ao tha few black anpleynee cm likely te have had higher nartallty rataa and there la an ratoon for aappoelng that tha aaall aauplo, froa which the proportion of black eapleyeoe wan ontineted, ma nacnaaarlly raptnonet atlvo. Thirdly, ae nlmnat af uncertainty la Introduced by cha failure to trace aa away aa 7.31 of mplnyoaa. Tun ether crltlclaaa are note laportant. flrat, the oxpactad auahara of daatha have haan calculated oa tha coemption that the am would have experienced tha aaaa nartallty rataa aa tha 'Alta aula population of the whole country at tbo cerroepoadlag datae. Tha am of national rataa la eoawa practice la etudlea of laduatrlal populatloea and tmdo to roault la m ovar-aatlaatlm of cha expected ambore of deatho, ae that the uaptofoea ucc 000264 13 appbar to be unusually haalthy. Thin "healthy worker affaet" 1* mil known and will ba takan Into account la the dlaenaalon of tha raaulta. A aora serious objection to tha uaa of national rataa la tha my aortallty varies froa ona part of tha country to anothar, duo to dlffarancaa In tha prevalence of envlronaentsl and aoclal factor* unralatad to tha occupation of Interact. It la, therefore, generally preferable to uae State (if not county) rate* In place of national rate*. With 37 plant*, however, it eight be thought chat chair geographical dlatrlbutlon would be efficiently wide to aaka the uae of national rate* appropriate. Unfortunately 22 of the plant* are located in the South and a check would have been dealrabl* to tee whether thla could haw* earned any eaterlel dlatoratlon of eh* reeulta. Secondly, eauaea were not obtained for 97 of tha 1,336 death*. Thla ana allowed for la the calculation of tha owerall aortallty by including death* da* to unknown eauaa*. It ma not allowed for, however, in the calculation of tha dlaaaae specific aortallty rataa and will have earned tha standardised aortallty ratio* (SMta) to he uaderoatlaated by, on average, 6.3Z. Per the present purpose, therefore, the ambera of deaths attributed to specific diseases have each been aultlplled by I.0674 (100/(1-97/1336)) aad rewaded off to the nearest integer. OK Study The study roportod by Jeees (1906) sa behalf of the OK Sealth and Safety taacutlve sewers 3,ASS ana who ears eaployed for at least oa* year la jobs that involved potential enpooure to VOI for at least 23Z of the working week aad who were first eaployed la ths period 19AO-74. Details of tha aae wore eoapllad frna the perasasal rocerda of 9 chaaleal plants aaaufacturlng or polyaerising TC aad the vital status of the sen wots determined at the sad of 19S4 froa the rocerda of ths Rational loalth Service Csstral Register. Five thousand four hesdred aad ninety sight asn ucc 000265 were traced (9g.9t). Savon hundred and eighty deaths war* identified mad capita at eha death certificates wars seat to the investigators ceded to the 9th ICD revision of the list of causes of death if they occurred before 1979 sad to the 9th revision if they occurred later. Several specific points about the study need to be noted. First, national mortality rates for England and Wales vers calculated by five year age groups for quinquennial periods for 96 causes of death and these vara used to estimate the numbers of deaths that eight have been expected in the cohort by aultlplylng thee by the corresponding numbers of person years under observation. Sane difficulty ves, however, experienced in obtaining suitable rates for ell causes of death, which could be related to the causes of death coded according to the Sth end 9th revlaloua of the XCD list, and rates for a relatively late period had to he wood for estimating the ambers of deaths from nany diseases that might have boos expected to occur in earlier periods. For 2 categories the earHost available rates were 1990-94, for 1 they wore 1945-49, for 24 they were 1970*74, end for 1 they were 1975-79. Secondly, an attaxpt was node to classify ana according to whether they had high, latemodlate, or law exposure to VGM or FfC dust end each nan's msployaeat history was recorded with advice froa the plants concerned, wader 12 job titles. The ms wars than grouped according to whether exposure to PCM was lihely to haws been high (group 4), exposure to V FTC dust woe likely to have been high with exposure to 9CM lew (group I), or exposure to VCR and FVC dust was intermediate and interuietouc (group O* 911 other non, who would goaorally have had low exposure to both fQt sad PTC duct, wore classed as group B. Within all groups, exposure to TCM was likely to hove boon higher if it hod begun before 1959. The study makes an Important contribution to the mm at knowledge eoucoruiag the long torn effects of VC. Thu use of national rates to ucc 000266 calculate eh* expected abera of deatha ty be juatlfled on tha ground* that tha aea aara aaployad la 9 plant*, which ware preeuaably distributed about tha country, but no datalla of thalr loeatloa ara given. In general, mortality rataa toad to ba higher la tha parta of Irltala whore heavy laduetry la locacad than la other parta, ao that the expected nuabare of deatha are wore likely to ba blaaed downwarde thaa upward*; but thla need* to ba ahowa. The uae of recent ratea to calculate expected auabera of death* Iron any epeclflc cauaaa of death wee praauaably oeceaaary. If the dlaeaae* wore to ba atudled Individually and will have dona ao harm If tha Incidence aad fatality of tha dlaeaaee la qoeatlon have remained atable. It would have baea deelrable, however, for tha dlaaaae* to have bean specified a that tha reader would know which were liable to ba dletorced. The ayatao oaad to claeelfy wan lata four exposure group* la the aort of ayatao that la caooooly oaod If precise oaaaurea of expoaure are not available. It ereatee name dlffleultlea la tha otatlotlcal aaalyala If wen wove froo one job to another aad ara elaooad <aa la thla lnetaace) aa having had high expoaure if they have over had a particular type of eoployooat <a.g. ever boon anployad aa aa autoclave worker), ho evidence la provided to ahow that tha poreoa yearo at rich before a ana entered the category hove been aebtracted aad added to another onpooaro group before the where of expected deatha are calculated. Novenoata fron one job to another ara eald to have tended to bo dot of grange A aad late D, but even ao it wt ba praawaod that the oxpected wbera of doatho la the flrtt tea onpooaro categoric* ara likely ta bo ever-eatlasted. Caaadlaa ftudy The Caaadlaa atady (Theriault aad Allard, 1M1) eao Halted to eapleyeao of a alagle plant la ghaelalgaa, Quebec. The plant, which woe altoatod la aa ladaatrlal conplax, eao opaaod la 1943, VOt aad PfC were ucc 000267 16 both nada until tho lot* 1960a, when th production of PCM eaaaod, while tho production of FVC contlttuad. An attempt uoa aed* to tract all production worker* whoa* senes appeared on the union*' liata or th* eenpaniea' payrolls of th* whole lnduatrlal complex including the VC plant between 1 January I960 and 31 Dacanber 1972, and contact waa node with th* worker or hla next of kin In 1,611 out of 1,639 lnatancaa (97.IX). Detailed occupational and making hlatorlea wet* obtained by questionnaire and 136 non who had been anployed by the coepenle* for lea* than 3 year* were excluded. The remaining nan war* categorised a* (1) exposed to VCM If they had worked on the production of VCM or FVC for at leant 5 years (451 nan), (11) unaxposed to VCM If they had worked slallarly for leas than 6 nonths (170 wen), and (111) other nan (136 In total). The lest were excluded froa the study. Follow-up waa eloaed on 31 Dacanber 1977. Copies of the death certIfleatea wura obtained and the Cannes coded according to the 6th revision of the 1CD list for *11 who had died (39 exposed and 233 unexposed). Information was also soothe for histological or cytologlcal confirmation of all the diagnoses for all th* exposed ana who had died of cancer. The reaulta warn swsnlnod la two nays. First, the mortalities of th* exposed and enoxposod ware compered after standardisation for qulaqueonliai and 3 year age group. Secondly, the mortality of tho exposed nan was compared with that expected if th* non had experienced the sax and age specific mortality rates recorded In Quebec for the year 1971* In both comparisons the censes of death used warn those specifled on th* death certificate sad the additional pathological information was used later only for laterpretatloe of th* results. Mast of tho expoood nan war* exposed for non than 10 yoaro (731), tho average loagth of expoeur* was approximately 17 yoars, sad 442 were observed note than 23 yoars after first exposure. ucc 000268 _ Although nail, tho study aakaa a useful contribution to the overall raaulca. Tha hiatologlcal review of tho eaocor eaaaa la particularly helpful. Thla showed that all S eaacora diagnosed as liver caaear (Including 2 specified as hepetooa aod ooo spodflod aa anglosareoaa) were angloearcouas of tha liver as nil as on that had baas diagnosed as anglosareou of the perltooen. Tn other eases of sngiotsrcous of the liver vara found to have bass certified as hepatic cirrhosis, ie la also helpful to have a cnparlsoa betnea exposed and "unexposed" eaployees of the sane companies as this shows that ehe lev oortallcy observed for all oau-usligunt diseases could be aetrlbutad to a healthy nrker effect sad ns not dn to bias In tho recording of exposure (relative risk for all ooo-nlignant causes coopered to "uaexpoeed" au 0.99) One crltlcln has to he nde> newly, the use of provlaelsl races for on year (1971) to calculate expected aortal!ty spread ever a 30 year period (1949 to 1977 lncluelve). Snaths will have tended to bunch up tonrds the end of the period of observation no that tho rates for thla particular year asy have bees fairly representative, but It oust have caused sew distortion of the expected embers of dnths, the else (and eves che dlrwtlm) of which is laposslble to astints. For not disease gmpe the distortion Is unlikely to have been large. Italian Study A seedy of all ass euployed is the productloo of TC and PVC In 9 Italian plants ns bogus la 1919* All an nre Included she nrs asployed for at lent d souths at say tlw few tbs start up of the plant to the sad of 1991. The study Is still laceaplete, but results are nee available for sas la three plants (lelll e al.. 1994). fee plants (la Ferrers sad Iselgaaaa) hegsa operation la 1953* Four hundred and thirty seven sen acre asployed la tbs sw plaat sad 141 la the ether* All bet three (frou the Ferrara plaat) were followed to the.end of 1994. Expected deaths wars ucc 000269 18 itlHttd by nultlplylag the person-years at risk by the corresponding national eortellty nets for ooeh 5 year *g troop sod galnquenalA. Total expected deaths la ooch case anounted to norc than 102 of tha eaployees U2.4Z and 12.8Z). Clinical laforaatloa was sought about tha causa of daath of all tha 35 euployees of tha Ferrara plant who bad dlad. lawlsad diagnoses, which wars not uaad for couparlaon with tha expected deaths, rawaalad four daaths froa caoctrs of tha llvar la placa of oaa. Tha lawaaaa plant did not bagla oparatloas until 1959. tlx hundred and thirty night nan wars anployad. All but 4 warn traced to tha and of 1913 and 17 warn found to hawo dlad. Ho oaa could hawa baoa followed for wore than 24 yaam and only 23.1 deaths (3.9Z of the workforce) wars expected. Tha data for chin plant hawa not, therefora, boon used la tha principal analyses. It nay be noted, howawar, that one daath woo attributed to Hear cancer whoa 0.1 was axpactad. Other gourcaa The Norwegian study (leldaas e si.. 1904) presided oboeructions on 454 nan who had boon anployad la a plant la Talaasrk where TGM had been aaaufacturod froa 1930 to 1971 sad PTC from 1930 to the cad of the study period, dll warn Included whose omos wore recorded on tha Coupsay's perseauel register and health depertaeat records who hod ewer baoa anployad frou tha start of production to tha and of 1909 and had worked for at least oaa year. Man ware followed freu 1933 to 1979 luclnaise. Booths and cases of cancer wars Identified frou tha records of tha Central Bureau of gtatlstlcs aad the national cancer registry. Bo reference la node to any ana being loot te fellow* up. Bet It can be Miwod that the maker (if not aero) was Mali as all eltlseas base aa Idaatlty --bar which la used by both caplayers aad the central agencies. Jlfty ms were found to base died against 39.34 expected If the sex- ago- and gwlniooulu^speelfle national ucc 000270 19 aottallty rata* had operated. Twenty-on* mb nn found to hav* developed 23 eaaeara againat 20.16 caacara expected froa tha eouparablc oatIona1 tneidanca rata*, th# oa* of which waa juatlfled by eh* fladlni that tha Incldanca la tha county la which tha plane waa altuatod waa batwoaa 90 and 95 par coat of tha rat* la th* country a* a whole. Oa* aaa who had boon aaployad on FVC production developed aagioeareoaa of eh* liver. Obaarvad and aapaetad niabata of caaaa are (Ivan for caacara of the lung, colon, and thyroid, for aalanonaa, and for all caacara, but no aapaetad auaber* are given for other type* of caacar. It la evident that aevaral other eypaa of cancer auat have bean la deficit, a* chare war* caaaa la all agalaat 14.93 aapactod, and it la difficult to know what weight to give the exeaeaae obaarvad for the reported type* of cancer no they aaa* likely to have bean reported (pacifically bacauaa tha maker* were la eaceaa of ehoa* expected. Tha author* note that on* further eaae of aalanoaa had occurred after the eloaura of tha otudy and that oaa "Incipient caaa" waa alto known to the*. Th* Ceruan atudy (Crelaer at el., 1992) included three groupai (1) 7,021 aen who had been expoaad ta TC la the cauraa of their eaployaane In any of tha 11 fiance la which TC and PTC had baea produced la eh* Federal Republic ef Oetaoay, (11) 4,920 *aa who had been aaployad la 7 cbraleal pleat* witheat havlag had nay expoeure t* TC, and (ill) 4,007 aaa aaployad in 2 ether pleat* where PTC waa proceaaed. tepleyea* were lacludad only If they war* ef Oanaaa or iaetrlaa nationality and ware regarded aa expoaad to TC if they war* production wurkara or other akillad wurkara or labourara aaalgaad regularly tw th* plant*, but aot if they warn aaployad in tha* only oeaaaleaally. dll aaa war* lacludad froa tha tlae of opaaiag of th* plants to tha end of 1974 and ware fallowed ta the and of 1974. Many of tha aaa warn, la coaaegueaca, obaarvad for only e few peer* after flrat wployent, end only 142, 342, and 192 reepoetlvely of tho three groopo ucc 000271 20 were first eaployed before 1934 end eo capable of contributing persetryears at risk sore than 20 year* after first oeployaeat, when so occupational hatard of cancer could bo expected' to be observed. 93.22 of tha exposed group uere successfully followed up sad causes of death were discovered for 92.tX of the 414 non discovered eo have died. The proportions for the other two (roups were respectively 89.41 and 88.7T for the unexposed and 92.12 sad 88.92 for the PTC process workers. The failure to obtain causes of death for all tha non who had died was allowed for In the subsequent analysis by weighting the makers attributed to each cause by a systaa which took account of tha age (roup and calendar period In which death with an unknown cause occurred. The expeetod numbers of deaths free each cause was calculated by multiplying the person-years at risk by the sen. age, and cause specific aortslity rates for tha PBS. Rational data bofore 1988 used sa Idiosyncratic classification systaa and the 1988 rates had to bo used to uultlply all the person-years at risk up to the end of 1988. for suboequeat years (1989 to 1974) the parson years st risk wore aultlplled by the correspeadlag rates for the saae calendar year. Epldealologlcal studies are aura difficult to carry eat la Germany than la l, ieerlca, the OK. or (caadlnavla because tha nodical cause of death is net recorded publlely sad chare la as aaatral systaa which can be used for chsthlag whether an ladrlldusl Is alive sc dead. Xa the clrcwnstaaces. the Geraaa aethers have aade valiant efforts to obtain reliable data sad tha prepsrtiaas of non la tha axpsssi groups aha were not successfully followed up (8.82) sad tha preportlees of deaths for which the cause was set obtained (7.22) were slailar to these la the lavlroaneatal Health Associates (1988) study. The defects, however, wake tha data lass easfel. rirst. as national aortslity rates wars avellabla before 1988 sad the see of the 1988 rates tn ucc 000272 21 estlaete the sorters of deaths la and before 1964 will have overastlaated the mabers attributable to diseases that ware basoning wore prevalent or diagnosed aora oftaa and will have uadareatlaatad thoaa dua to dlaaaaaa that vara batoning laaa prevalent. Secondly, aad aora laportaatly, a largo proportion of tha aan had baaa flrac anployed laaa than 10 years bafora eha follow p waa aodad. It follow# that tha observations on eha faw aan who had baaa flrat exposed long aaough bafora to have had aueh chance of developing aa occupational dlaaaaa with a long latent period auac have bean swasped by obaarvatlooa that had little to contribute. Tha expected deatha enounced to only 6.2Z of the axpoaad aan and there la, therefore, little to be gained, and aonathlng to be loot, by Including tha Canon data In eha overview, it nay be noted, honour, that 12 deatha warn attributed to cancer of tha liver aaoag tha ?C axpoaad workara agalaat 0.9 expected and that awaller exeeaaea ware alae abaarved la tha unaxpoaed ehealcal workers (4 obaarvod agalaat 1.1 axpactad) and la tha FTC procaaa workers (3 deaths against 0.S axpactad). Two Swedish plants have produced TOl aad PTC since 1943 and 1971 respectively aad aaployoes of tha first have been studied by Syren at cl. (1974). g11 persona who had ewer boon anployod Aan exposure to TCH could occur ware listed Iran tha personnel files of tha factory. Twuuty-eae wars excluded because they ware foreigners eha left the cannery after a short period of anplopweat. the renaming 730 were fallowed to October 1974. txpectad carters of deaths were ootlnated by eultlplylng tha person-years at risk by tha eorraapendlug age-specific nortallcy rates far tha whole cannery sad tha axpactad wabota of cancer cases Iren 1931 to 1971 laclwslws (daring which period all cancer cases bsd been registered nationally) were catlantad by anltlplylag by the natlannl age-specific enneer Incidence rstaa. Is bath Instances, tha rates used were these recorded la 1949. Fifty-eight deaths warn found bat aa figure la given for ucc 000273 the expected amber. Detailed figurea are given only for the nuaberu of daatha end uni obeerved tad expected for cancer at eho long ami tor camera of tho liver and pancreaa ceablned and for the nmbera of daaths froa brain eanear and thraa catagorlaa of cardlovaaeular dlaaaaa. Two nan known to have aagiouareoea of tba liver vara cartiflad aa dying of liver eanear or pancreatic eanear racpactlvely, and a third aan died of tha dlaaaaa 17 aeneha after tha felloe up eloaad. Tan French atndlaa provide tha reunite of a long tarn follow up of aan eaployed In ana plant (Plerra at al., 1979) and of a abort tarn follow op of aan aaployad in 12 planta (Laplaaeha at al., 1997). Tha flrat provided obaarvatlona on 1,111 nan expound to TC la tha prodnetloa of VCM and FTC and in aalaetad ancillary oporatiaaa fraa tha opening of tha Tavaux plant in 1933 to tha and of 1979 (Florae ft al., 1979). Six other aeployoea- wore excluded fraa tba atody bacanaa of Inch of oeenpatioaal hletorlea and 190 nan bacanaa tbnir vital atatna at tha end of tha atudy period wan nadeteralned. Twenty-five aan ware found to have died againat 49.73 expaetad fraa coateaporaaaooa aax aad aga-apacifle national aortallty rataa (3.71 of tha am at rink). One danth ana attrlbwead to angioaareoaa of tha liver in a am who had bam expound far aero than 13 yeara. Tho reported data era aa iaeeapleta and caver aaah a relatively abort period free tha opening of tha plant that they add nethlag ef apidaaiologieal valea to the reoalte ef the ether atndlaa, apart frm the addition ef a farther cam ef aagiaaarema. The amend previdaa obaarvntiaea m 1100 am aged 40 te 33 yearn me, in 1990. warn mpaaad er had bom axpoaad te TC la 12 planta, mieh cmtntltntod "mot ef the Fraach KM pelyaarlaatian planta" (Laplaaeha at al.. 1997). Many ef tha am were, or had bom mplayad at Tavawx aad vara praamably aarvlvora ef the cohort atndled by Fiarrn at j. (1979). The am ware followed far 3 yearn cad their aerbldlty cad aortallty coopered ucc 000274 with that observed la 1100 mb of the sate egos (+2 years) esployad la the bom pleats, bat who had never beea exposed to VCM. Men la hath groups ooro Interviewed personally aad lafotMtloa nbtalaod about their aanhlag aad drinking habits which were foaad to bh alallar te the two groups. Men la the exposed group had baea first exposed as avenge about 14 years and had first beea oxployod la the plant about II poors previously. Morbidity and aortallty data wars recorded annually by the pleat physician who auecaosfully traced MX of the expooed aaa aad 96X the controls. Oae of the exposed mu but bom of the uasxpooed developed as anglosareeea of the liver, bate are net given separately for different periods after {lrst anploywent aad It la lapoeslble to aasesa the significance of the finding that I of the exposed aaa developed laag eeacer *g--* 2 of the controls, which say well reflect a chance occurrence of unusually few cases la the control groups as the proportion of all cancers la the control group that arose fran the lung (2 out of IS) Is unusually Inn* . Sue exposed nan developed a cancer of the Xyuphvhaanatopeletle spate* against sons of the controls, but m ms In either group were bacon to base developed nelsnoaas or cancer of the brala or thyroid. A Japanese study reports the uortsilty export uss of 4S24 nan employed for st least one year before IMS la one or other of 2S Japanese pleats which began producing vinyl chloride asM*ar (VCM) or polyvlayl chloride (FTC) between 1M9 and lMd Inclusive (IshMsra, IMS). Tbs ms were followed to S1.10.7S. Twenty sight par asst of the non ware observed aero than 20 years after first o*ploy*set, but none use observed aero then 2d yours. Only O.dX of tha nan worn uatrscsd aad espies of the death certificates Mrs obtained for all the 209 nan who had died (4.iZ of tha initial cohort) Individuals were classified according to the job oa Mich they had boss longest oaplayed at the terwlaatlee of their follow op aad data are given separately for 2944 nan classed as flayed on PTC ucc 000275 24 production sad 197$ othora (Including 900 clnnood fCM production workers). If this study Is continued (or s further 10 peers It should provide useful oddltlonel InfonetIon, but the present date Include too few observations on non oore then 20 peers after first exposure to be of enp aatarlal use. They confirm the evidence of a hazard of liver cancer with 6 deaths la PTC production workers against 2.54 expected free national rates while only 1 such death was observsd la the other workers against 1.92 expected. One of the 4 deaths free liver ceueer la |VC production workers was certified as doe to aaglosarceea of the liver and at least one of the other liver cancer deaths was due to the sens cense. Lung cancer deaths were not in excess (2 observed in PTC workers against 2.33 expected). Ho data are given for caaeers of the lynpkatlc end haenetopoletlc aystees, for cancer of the brain or thyroid, or for aslanouae. The nortallty of 305 Japanese VC workers reported by Nseuda (1979) la preeuaably subsumed in Hskenure*s (1993) later and larger study. fasardp of epe^er The results ef ths four asst useful studies are listed individually lx Tables I and 2. Ths overall results for all senses, liver cancer, and three bread groups ef conditions ere shown la Tahls 3, and those for 10 types, or classes, ef cancer la Table 4, hats hews not been reported for sash type of sanest la each study and the sources of the data are, therefore, specified separately for sash type, additional Isforaetloo la given is Table 5 for 7 types or classes ef eaaasr that has been obtained frou four ether less Informstlve studies (the guinea study reported by Orelser sl. 1912, ths Horweglan study reported by Haldane at si., 1994, the bwndlah study reported by byres pi.. 197*, and the part ef the Italian study relating to ths Baveaaa pleat reported by belli el., ucc 000276 25 1M6). table 3 iton (hat. apart (tea eaneor of tho liver, Cha overall orcalicy la what would be aotlelpatod for ao laduatry without any najor haaard of accident or dlaaaaa. la particular tho ataadardlsod Mortality ratio (Ml) of 14 for dlaaaooa other than cancer la typical of the ratio* that aro coMoaly obaarvad la group* of aaployad mb* a low SMg of thi* order refloat* the "healthy worker effect", which reaulta frou the aelactloa proceaa that laevltahly exclude* aene of the leaat healthy oubera of the populatloa froa laduatrlal onployneat. Ihla effect dooa aot, however, norually affect the ortallty free cancer beyond that obaerved in the flrat few yeera eftar the atart of onployneat and aa SMR of 102 for cancer* othar than cancer of the liver la coapatlble both with the abooaco of haaard and with Me of M far ether dlaaaaaa and 77 for accidenta, peiaonlaga, and violence. Beath eertlflcatea are an ^reliable aowree of lafornatlon about the hlatology of cancara chat cnaaa deach, bet there la ao reeooa to ouppoae that the enceao ortallty attrlhwted to liver cancer (or, la the OS aerie*, liver and gallbladder center) la ant entirely accounted for by the known haaard of aaglooarcona. fifteen of the 37 death** attributed to cancer* of tho liver and gallbladder la the OS eerlaa are known to have bnen due to aaglooarcona nf tho liver (tavlreanental Bnalth daacetate*, 19M). In cha Srltlah oerlea, 7 of the 11 daatha attributed to liver cancer, aot apeclfled aa oecondary, were known te be angleaareonaa and all oceurrod in autoclave wothera ngnlant 0.31 enpnttad liver caacera of all type* (K>0'S, Increaond to 39 la Table 1, te take aceewat of the 97 extra death* of ucc 000277 26 J(Mi| 1986). In the Canadian iirlu, histological rsvlav shooed that 7 of the ao eallad livor eaaeara vara aaglosarsoaas (1 had boon described at aa aagloaareaaa aa tha daath certificate, 2 aa hepatoses aad S aa uaspecIflad liver eaaeara). Oaa ao eallad liver eaaear daath aaa found to have boon daa eo eaaear of tha algaold colon, ahlla 1 daath attrlbntad to aaglo aareaaa of tha perltoaem aaa found to have baaa doe to aagloeareoaa of the Hoar (Theriault aad Allard, 1981). la tha Italian study, further evidence revealed that 3 farther daatha ehould have boon attributed to eaaear of the liver (aaklag 4 la all), hut only 1 of tha 4 ana deeerlbed as aa aagloaareoaa (lelll at al., 1986). further evidence that tha excess uortallty froa liver eaaear (or liver aad gallbladder eaaear la tha 08 aeries) can bo attributed vbolly to tha kaovn hasard of aagloaareaaa la ohtalaad by cefsriag tba casase deaths with the auabors of deaths froa aaglooareoaa before tha ends of tba follovuf period that are recorded la the haglater of Liver Aagloaareaaa Cases aelatalaad aa behalf of the Assoclatloa of plasties Maaufaeturers la Karapa by Xaperlal Chaalcsl Industry VLC (heaaatt, 1986). fifty oaa axeasa liver eaaeara are recorded la the eonblaad data aad 49 aagloeareosaa are recorded la tha laglstor far tha relevant periods for tbs three countries aad tba cue Italian plants* that arc covered by tbs survey. ana of tha 120 eases yet recorded aa tha loglstsr vara is asa oho vara first axpsssd after 1969 sad none of the 43 non affected la horth Aaerlca ware first axpsssd altar 1964. All nay, therefore, have bees exposed to eeacentratlaca of several hundred ppn and easy nay have base twenty alas arc registered aa occurring la the OIA against as eattaated excess of 33, 9 are registered to the IK against aa excess of 9, 10 la Canada agatoat aa axsass of I, aad 1 to a oaa anploynd to oaa of tba cue Italian plants studied against aa excces of Inna than 1. Two of tha Caaadlaa cases are recorded la tha laglstar aa occurring la non mo had bans anploynd for 3 years and It to passible that the actual duration of axployuaat had baaa slightly loss than 3 years, man tha nan oould have boas excluded froa tha study. ucc 000278 27 Mposed to concentrations appreclably hitter (terr, 1981). bn| Cuetr The 14m that exposure to VC might mom cancer of ete lung was suggested by Homos *1. Is 1974, whan ttey oocod 13 uui in a study of proportional mortality against an expected ambit of 7.9. Tha eooblnad data that aro ahovn Is Tabla 4 do mot provide aay support for ttaa hypottesls oalthar for respiratory eaocar as a stela (INK 97) nor in tha opacified data for lust easesr la tha CIA, UK, and Italy (SMI 99). Thara arc, however, coMlatastly higher risks Is tha oshtrospa of mas is ete US aad ok oarlas la stem oeeapatlamal haaards scold ha more likely to be scan ttea la otter grasps. This is sees from Table I, sblch atesa that ete Mi are ellghtly higher la aaa observed 20 yoara or more after flrat exposure them la mas otearvod aarlior, la aaa employed before 1936 la ete 01 than in mam first employed after 1933 (tees axpoasre levels are believed to have bees laser), la aaa employed for laager ttea for a shorter periods la the USA, aad la aseoclave sorters la ete OK, la team ete aagloMrooms cases tesa meetly occurred, ttea la otter sorters* Tte dlffareacas are all small or wary small, bet ttey are all la tte aaaa direction aad tte probability that tte rates should all be higher la ete grasps la telch aa occupational teserd la nave litely to bo aaaa la each of tha fear pairs of groups is 1 la 16. Additional laformatloa from otter sources la gives la Table 3. Thirty deaths (or aaaas) sera sheened la total aad this la raised to 31.3, teas allosaace la made for tha enter of deaths due to unidentified Muses is tte Games study, glvlag aa as of 103* Xa tte Oarmea study ete M was higher la am she had tees owpaasd tor 10 years or more ttea la am she had boas exposed for lass (111 agalast 79) aad la tte Homaglaa study, 4 of the 3 eases observed eeeurred la aaa shoes secapetiaae wars regarded as lsvolvlag high exposure agalast 1*12 out of the 2*64 expected, loth the ucc 000279 28 Catuaa and the Swedish studies derived the expected numbers of deaths from actional mortality nets for a alafla roar towards the oad of eho study ported. Tho expected auabers of deaths ara likely, therefore, to have boon overestimated and the SKks correspondingly oadaree timeted ea the mortality from lung eaacar had haaa rtaiag throughout the period of observation. train Cancer The Idea that VC mlfht canoe brain cancer was outlasted by Moneon at al. (187*). when they reported 3 eases against 1.2 expected, in the same paper la Aleh they suggested that VC might cause lwg eaacar. The combined data chat are shorn la Table * provide some support for this hypothesis. Maaaoa et jl.'s eases were, however, observed la US workers sad oust be presumed to he Included la the total reported by Environmental Sealth Associates (1916), la which esse they will have contributed a substantial proportion of the total la table 4* To tost the hypotheals we ought, therefore, to subtract Mouaoa at al.'s data. Their study was not a cohort study aad Chair expected daatha do net correspond exactly to those la Table 4; wa cam, however, oaly subtract bath the ehearvmd and the expected cases from the present totals. This leaves 24 observed against approximately 19.3 expected, a difference which might easily occur by chamau (f, one-tailed, - 0.1).* Additional Information from turn other sources la gives la Table 3* Tho small aaaaas reported provides little further evidence of am occu pational hasard, as ams of the two deaths staarved la the Swedish study seeurred la a yowg asm who bed bees auplsysd for leas than a year wham the diagnosis was made, while the excess death vatu for brain eaacar observed If we regard haxweller etal.'s (1976) atudy am the origin of the hypothesis uu should ha Taft with 26 deaths against 19.94 expected (P, one-called, - 0.07). 29 la cba Sanaa atudy i laaa than chat oboerved la ehaaical worker* aot exposed to vinyl ehlorlde (2.9 death* aftar allawaaea far deaths froa unknown eauaaa against 1.6 expected) and la workara la tha rvc fabrication Industry (3.9 deaths aftar allawaaea for deaths froa unknown eauaaa against l.l expected). Caneara of Lyophatle and laaaatapolatle Tlaauaa Tha Idea that VC night eauaa eaaear af tha lynphetle and haaaato- poletlc tlaauaa - and aora specifically tha lyuphatlc tlaaua - waa suggested by Tabarahaw and Caffay (1976) aad by Mexwaller at al. (1976) whan ebay fond, respectively, la two cohort atudlaa (1) 3 daacha fraa lyaphoaaa la tha aoat haavlly exposed workara agalaat 2.34 expected, aad (11) 4 daatha fraa caaeara af tha lyapfcatlc aad haaaatopoletlc tlaauaa agalaat 2.5 expected. Thaaa aaall aacaaaaa sight have boon Ignored If tha laboratory flndlaga had aot boon interpreted ao auggaatlag that lyaphoana wara produced experimentally la aalaalo aapoaad to vinyl ehlorlda by Inhalation (Haltoal and Lafealna, 1175). On thlo baala, tha idaa that alallar axpaawra alght aloe eauaa lyaphoaaa la Mao aoadad aarloaa eonaldaratloa. tha data froa tha four principal atudlaa that arc natrltad la Tablo 4 provide little aapport for tha hypothaala Aon all eaaeara of tha lynphaela aad haaoatopolatle cleaves era eoeeldarod together (37 deaths agalaat 30.t7 anpactad. Mi 112) aad vary little aora la obtalaad fro* tha separata data for eaaeera of tha lyphatlc eystaa (ualag Tabarahaw aad Oaffay* a daflalcloa of ZCD Hat aahora, 6tk rwvialon, 200-203 aad 203) that are thaws la table 1 (33 daatha agalaat 29.40 anpactad). tha position la, aoraovar, hardly altered by avbtraetlag tha data la tabarahaw aad Caffay'a initial report (29 daatha agalaat 23.36 anpactad, M 124). loan little additional lafotMtlaa la provided by tha reeulta of tha Oathaa study, com of which era ahsam la thhla 3. thin obtalaad an M of 214 for anpeaed workara (baaed oa IS observed daatha, lacroaaad to 16.3 ucc 000281 30 whan allowance la ends for the aunber of deaths of unknown cause) cg*in*t SMKx of 7? sod 34 for oa uaexposod croup of choulcal wtktra oad o croup of PTC fabricators and showed that the oxcoao la the exposed worker* waa present only for aon who had been exposed for aore than l pear and we* aost narked for non who had been exposed for 3 pears or aore (10.7 deaths after allowance for the aoaber of deaths of unknown cause against 4.0 expected, SMI 264, P onertailed <0.01). Helanoaa An excess of aslanoaa was reported la Msrvwglaa workers bp leldass at al. (19C4), who raised the posslbllltp that TC night haws producad the dlseaae. Pour cases wars observed whan 0.79 were expected, three of which were la non whose occupation* Involved the highest exposures (against 0.31 expected). At the tins of writing their report one further case had been detected with onset 3 pears after tho closure of the study. Subsequent studios la other countries ham, so far, reported only 2 deaths against 2.0 expected (one Tables 1 and 3). Thyroid Cancer An excess of thyroid cancer was also reported in tho Borweglaa study, la which 2 cases warn observed against 0.1b expected, the Investigators were ant swam of any other stadias Indicating an excess of thin type of cancer and thay draw as conclusion frea their observation. Two of the three aajor studio* that have boon reported since tho goruction observation was node give an data for thyroid cancer; the third reported two deaths against 0*43 expected (ana Table 1). Own death Iron thyroid cancer, it nay bo noted, was reported la the USA by Monaco aj: fl. (1974). Csacors of Digestive Treat uggestlous that VC night cause cancers of the digestive tract in general have sonatinas boon nods but thay have ant taken adeguato account ucc 000282 i 31 of ch contribution that cancan of tbo liver aaka Co tho total nuaber of eaocerc of tho digaatlva eyatn, particularly when It la bona lo alnd that cove liver cancan an llknly to bo aiadiagaoaad aa eaneara of othar organa. The contained data froa tha four principal atudlaa ahovn In Tabla 4 uelgh heavily agalnat tha ldaa that any auch offact haa baao produced. Othar Cancera Oaa of tha reeainlng typaa, or claaaea, of caacar llatod In Tabla 4 hove a atatlatleally algnlfleant accent naaaly, tha heteroganeoua group of "othar eaneara" ($3 obaanad deatha agalaat 63.24 azpactad, P, tvo^ aided, <0.0S). Thla axeaaa la only narglnally aigulfleant and aay ba a chaaea obaarvatlon. Tha aoat llkaly explanation la, hovover, that a fan aagloaarcoaaa of tha Hoar nan not recognised aad vara dlagnoaad aa aacaadary liver caacar or carciaoaatoala, alta oakaova, aad to booatad tha maker of doatha la thla category. laaarda of ooo-aallaaaat dlaaaaa Vo previous atody haa aaggaatad that aay aaa aallgaaat caoao of death other than clrrhoala of tha liver veald ha likely to be lacraaaad aa a revolt of aapaaara to VC, aad that la oaly praamad to bo lacraaaad becauae of tho liver cheagea that von obaervod aa part of the 'vlayl chloride lllaaaa* (Mmtaller e jl., 1973 aad 1973s Joeea aad faith, 1912). Too other peaalbllltlaa have, bovevar, boon reloads aaaaly the production of aoo-aallgaaat rooplratory dlaaaaa bocoaoa of the cheagea la laag function aad x-ny appearaacoa that havo beea recorded for aaa aapoaad to PVC dnat (Ullaat al., 1976j laaorat al,, 1995; Sovtar, 1990; Lloyd at al., IMS) aad acata cardiac death, tree analogy with tba affect of other halegaaatad hydrocarboaa (Jones, 1996) aad tho ebeervatloe of aa lacraaaed aortallty froa ayocerdlal lafarctloa la tha fav yean following the conation of oxpooura la tha ivedlab PTC procoaalag ladaatry (Holca at al., 1911). ucc 000283 32 tolnnt fliuu for th* ibi of deecha fro* thoae ond other ooahU|um cauaca choc oro obtainable fro* th four principal atodloa --r* given in Table 2 and are eu--arlted la Table 7. Clrrhoela of Liver Throe of the four atudloa gave aaparaea figure* for drrhoala of the Hoar, non* of uhleh ahooed an lacroaoed wortalicy (Table 2); la combin ation they game aa SMK of 19 baaed on 41 deethe. The fourth atudy, which did not glee aeperata data for drrhoalo of the liver, reported 4 daatha fro* all dlaaaaea of the dlgeatlve eyat-- coeblaod agalnat 3.13 espeeted and noted that the 4 Included 2 that were certified aa do* to drrhoala of eh* liver, but were actually due to aagloaarcoaa (Theriault and Allard, 1981). la Che two eupplaaentary aeudlea la ohich data are given for ehla dlaeaae, the M -- 82 la oua, baaed oa 13.1 daatha after aafclag allowance for the uaker with uakaowa cacao (Gralaer et *1.. 1982) and 13) la the other, baaed a* 7 death* (lakaaura, 1983). Roe-unllgaant Baapiratory Dlaeaae The data far non-uallgaaat reaplratory dlaeaae are coufualng la that the total Stt fro* the ceablaed data for the tour principal atudlea in 80 and la the oort ef figure that la coaaoaly found 1* healthy laduatrlal population*. yet the DS atudy recorded a auhotantlally lacroaoed aortallty fra* eaphyae-- (41 daatha and M ef 180 before aay alio--nee la aad* for deatho of wah-- eauao). So ouch eseeoa no found in the OK where 34 daatha fra* breech1 tin gave -- 88 of 82. inter--tie--1 e--pnrlooaa of chronic -- --llgaaat reaplratory dlaeaae are c--plicated by the uaage of differ--t tat-- -- doocrlbe what It la now agreed, la beat called chronic obetruetlve 1--8 ( pal--nary) dlaeaae, but which la th* pact tended to be called --phya--a la the 084 aad chronic broachltla la th* 08. It --at, tharefera, be proa--ad that the to* catagerlee of '--phya--a* aad ucc 000284 33 'bronchitis' used respectively la th* two lir|i national studies war* aaant to dascrlba the mm thing, On this basis, wa should haw* to aaawa that eh* axparlancas la eh* Cm couatrlaa war* wary different, dasplts th* fact that both ralatad to eahorta chat had wary alallar axparlaacaa of angloarena* of th* llwar and ao, praauusbly, fairly alallar exposure* to VC. Separata figures are shows, where *wallabia, for tha aortallty obaarwad la a*o with diffaraae duratloaa aad lataaeltlaa of aapoaur* la Table B, Qnliko th* data for cancer of eh* tag thee war* ahowa la Table 6 they prowid* ao coaalatent awldaaca of a sraatar rlak la eh* teoups la which aa occapatioaal haaard would be expected to be coaeaatraead. Th* author* of th* tawironoaaeal laaleh Associates (1996) report war* uaabla to giwe any explanation for the lncraaaad ooreallty froa aophyaaoa aad point out that It could hardly ba due to excaaa cigarette sucking *a there waa no owarall excaaa for cancer ef eh* lung. It la atrIking, howawar, that th* axcaaa la Mr* than coepeasatad for by daflclaaclaa la eh* other catagorlaa of aou-wallgnaat raaplratory diaoaaa (pooMoala IS death*,* SMB. 47.0; other reaplratory diaoaaa 14 daatha, SMt 42.4) aad tha conation crlaaa whathar tha euphysana axcaaa could ba a claaalflcatory artefact. Buvlrooweotal Boalth Associates (1919) Hat all tha 41 daatha which ahow that they war* coded wader ICS Uat ao. 327 which, ia the owe ef data 7th rewialoa that wea need for th* coding of all daatha la tha study, waa th* cod* for 'other raaplratary disease not otherwise classified' aad included anphy*aa. Under that rewialoa, howawar, anphyaena that was associated with bronchitis should he classified with bronchitis Mder Uat nsubars 300 to 302 aad th* *Th* daatha attrlhwted to different groups *f respiratory diseases aad tha corresponding SMBa ehat ara deed ia this aectlan far th* OB study era as glwen by uwlronaantal lealeh daaeciatas (19M) and haw* not baa* adjustad to taka account ef th* nwuhars of deaths ef unknown cause. To taka account of the deaths tha obaarwad daatha and BMBa can hath be aultlpliad by 1.0974. ucc 000285 3* possibility asy be considered that sows of tht caphyseM deaths should have boss classified In bom category of rosplrstory disease other than U*t no. 32?. If this were so, It could account for both the excess nortslity froa enphyaena and the grossly deficient nortslity froa other aoa-uaUgnant respiratory diseases* No excess aortellty froa "bronehltls, mphyeoM. and asthma" was observed la the Osman study (Crelser ^t al., 1962; SMt ** with 6.3 deaths observed after eekiag allowance for the enaber of deaths of unknown cause). Cardiovascular Disease Date for lscheaalc (or arteriosclerotic) heart disease (vhieh My be presved to include the vest Mjerlty of ell deaths certified as due to acute cardiac disease) art given only by the tweo big national studies and they provide no evidence of sa Increased aortellty. The SKt of 90 for this group of dlseMea sad of 91 for ell cardiovascular disease recorded In the four principal stadias are typical of the Ms that are recorded la healthy Industrial populatloM and there la no suggestion of say eccupatloMl hasard in the subsidiary enslyees provided by the two national studies. In particular, there la no evidence of an lMreaaed Mrtallty within one aonth of leaving aeployneat la the OK study either for all workers (92 deaths, M 61) or fur the aeet heavily exposed autoclave workers (9 deaths, tm 62). A slight lMreese la fschoMlc heart disease aortellty was recorded for the exposed workers la the Genes study (Crelser el.. 1962), tot it was lass than that recorded for the uMrpnsil ehealcal workers and the PVC fabricators (MU of 127, 111, sad IS* based oa 97.2, 126.7, sad 109.7 deaths after allow!ag for the oMber of dMths of unknown cause). The Infatuation that has now bona obtained about tha long tarn health ucc 000286 33 of aa occupationally exposed eo vinyl chloride la aaesive and coupares favourably with chat available for any ocher occupational group. Two facta are outstanding. First the nan have experienced a specific hatard of a type of cancer that la normally axtraaely rates nanely, angiosarcoma of the liver. The rarity of this disease under other conditions eade the detection of the haaard easy; hut the long latent period before the disease appears after first exposure (almost always aore than 10 years and usually ora than 13 years) neant that a large nunber of non had been exposed before the hstsrd was detected and that It will still be eany years before we can be certain about the extant of the protection provided by the reduction la exposure In the 1940s and the further reduction that followed the recognition of the haaard la 1974. There la. ^fortunately, no effective treatment for the disease and tbs amber of cases Is refleeted in the nunber of deaths. Sene 50 desths have occurred In the 14,740 non who have been followed up la the four principal stadias that have been reviewed In this report, eo that approxlaately 1 la 333 nan have been affected, while It of the deaths have been due to this one cause. Iweatually aeny wore wen oust be expected to develop the disease. One estlaato (lleholson at el., 1994) awggeets that the total nay he Increased 10 tines, but a wore realistic aatlnata la 2 to 3 tines (Femes ml.., 1993). The noaoed outstanding observation la that the nortallty of the exposed nan, ether than ehat doe to anglesarcana of the liver, Is typical of the normally healthy Industrial worker. That Is net to say that no other haaard exists, bat that the offset of any ether haaard Is snail. The nesaive data that ace men available provide no reason for thinking that any haaard ocher then one of cancer has been overlooked. It Is, however, still difficult to decide whether 9C dons produce a risk of developing cancer, ether than aagleaarccna of the liver, *dch night be naell compered to the risks produced by aon-occwpetloaal causae, but yet ucc 000287 36 abeolutcly alaoat aa large aa the rlak of developing eha nomlly very rara engloaarcoea. Om of eha aaay haaarda choc have baas auggaatad can ba dlaaiaaad, aa ehara la no evidence to aupport it: nanly, eha Idaa that VC eauaaa any eaacar of eha dlgeatlve trace other than angloaareoaa of eba llvar. Two haaarda (of nelanone and caacar of eha thyroid) have baaa auggaatad only wary rocanely aad faw of the available etadlee have provided lnforaaclon about then. There la no good theoretical raaaon or laboratory evidence to auggaat that either ahould ha produced by VC aad, on praaaat evldaaea, the alaplaat explanation la that the reported eacaaeea are the chance effecta that auat ba expected when May different typee of cancer are atudled la aeveral different population. to far aa mIobom la coMerned It haa to be reaeabeeed that the dlaaoM haa becooa ouch Mre rneeun la recent yeara la Scandinavia (where the excaaa haa bean reported) due, it la believed, to eha popularity of auabathing aad the lacreaaed opportualtlea for ScaadlMVlaM to travel to the owner part* of Southern lurope and berth Africa, The extent to which thla My have affected the obaervetleu in Purvey aeeda to be exanlaed. Two other haaarda (of lyphan aad of brala cancer) ware auggaatad by the early raewlta of aaee of the leericaa atadlM. That VC eight produce a hacard of lyaphnea u initially aupported by the prellalnary reeulta of aalael etwdlea, hut the ceuplete reeulta of the uaay atudlea that have new been undertaken (aee Melteol end lefeelna, IMA) do not auggaat that lyaphooa or any ether cancer of the haeuetopolatlc eyetee la liable to be produced. There la, however, aoM evidence that brain tauoura can be produced la rate (aee Maltoal end lafoalM, IMA). The hypotheeea that lyuphooaa aad brala caacera eight he produced by VC have been aupportad by the oheervatiM that both thaee typea of cancer have caueed death Mre often thaa eight be expected free Mtloaal natality ratee, but the ucc 000288 37 excaasee observed la th contained data froa the four principal studies arc saall aod aoc statistically significant and tba hypotheses raaaln unproven. The aaall axeaaa of brain caacar la particular!? difficult to evaluate aa aortallty rataa froa thla dlaaaao Kara changed rapidly over claa in coapany with laprovod aathoda of dlagaoala aad tha auaplclon of an occupational haaard {which waa rained la 1375) could have Influenced the finding*. What axeaaa haa occurred hae bona Halted to the USA and Cernany and tha Careen finding* carry little weight, aa tha axeaaa waa found In each of tha three occupational group* todlad. Irrespective of che ehaalcals to which they ware exposed. The auppleaeaeary data froa the Caraan *tudy *howlag an Increased aortallty froa lyaphattc aad haaaatopoletle cancers are aora lepresslve, particularly aa the excess was aost asrfced for aen who had been eaployed for at least 5 years* In these elrenastaacss, wa aust still suspend Judgaent until the data for each study are analysed for each specific type of cancer, by Intensity and duration of exposure, and tine since exposure began* There reaelas the suggestion that VC alght cause lung eancer. At flret night, thin la ruled out by tha M of 97 for the coahlaed date for respiratory eaaeer for the fear principal studies, lung cancer la, however, aoraelly eo combos (accountleg for shout IX of the expected deaths) chat aa laareaaa la aortallty that eaa half aa laportant (emarleally) aa tha aortallty froa sagioeareoaa of tha llvar alght easily be overlooked (9SX eoafldoace Halt of the Ml. 15-112). The Incidence of the dleeaae la aoreaver vary variable wltkla a country end thnre anst be doabta uhether the eetlonel axperleaca prowldee e selteblo control for am eaployed la pleats that an esc evealy dlstrlbated abont the country, in these drcmataacee as eaaaet enelude aa eccupetlooel haaard naless it can be aheua chat tha aortallty within tha exposed aaa la Independent of the factors that sight ha expected to taflseace It, if eoae of It mra ucc 000289 occupational In origin: namely, tho intensity and duration of exposure and eho time since exposure bogon. It in not possible to auxin* thoao relationships in detail, aa the roporta do not provide all th* necessary infore*t ion. Such information u they do provide, which waa smaarited in Table 6, support* the idea that VC produce* a email hessrd of lung cancer. Takan in conjunction with th* knowledge that lung tmours have bean produced in aoveral species of anlaala exposed to VC by inhalation (see Maltoni and Lafenin*. 1PI4) it would seem that a email haurd of lung cancer probably did occur. The evidence is not, however, strong enough to conclude that it definitely did. If it did, the haurd we* evident only in am who had hem aaployed for may years at a time when exposures of sever*! hundred ppa or wore were comm and any persisting risk mu be only minute and Incapable of detection* The attention* that have ham left unanswered by this discussion night mil ha answered definitely If (1) all tha expooed am could be followed to (uy) the end of 1M4, (il) tha Investigators could present thalr data in comparable ways, taking account of duration of employment and time sine* first employment, and presenting data separately for sms first employed before (my) 1N3 and hetman 1H5 and 1V74, and (111) estimates would bo made of tha effect of correcting tha results for each group of employees for the locality la which they lived sad worked. IAZASM 10 IB CBODUL POPQUTXM As VC has ham proved to amass cancer la Warn and la a mtages la laboratory experiments. It out ho ptoamod that even tha minute doom that scaped late the general eavlrsmset from production pleats or (la th* early days of aaaufactura) from PVC mtarlals, will ham caused earn risk of earner to tha general public. Them risks meet, havener, ham bom mry am11, aa air coeccetrations of VC, aeon within a kllmetr* of VC handling ucc 000290 39 pines, uesd co bo (la or arouad 1973) of tho ardor of 10 to 40 ppb (Cnvironaancal Froeoecloa Agvncy, 1975; Air Products sad Choalcsls Inc., 1976; Baxter ^1., 1977) sad this Is about oao too thousandth of tho concentration that has cauaod as occupational hatard. It la obvious, thorafora, that It would bo lapoaslblo to doeoce tho risk of an? cancer that sight bo producod bp VC other than a risk of aatlooareoaa of tho llvor, as It has proved so difficult to detect any other risk In non who wore exposed occupationally. The position with regard to aaglossreooa of the liver la different. This disease la normally so rare that, la the absence of specific exposure to one or other of the known causes (VC, thorm dioxide, and arsenic la pesticides and nedlelnes), the annual Incidence Is of the order of 1-2 x 10"7 (tyros and lelaberg, 1973; Brady at el.. 1977).* Is these elrctastancea the discovery of even one cose In e nan living eloaa to a factory Is whleh VC was used In the days before exposure was tightly controlled nay he retarded as praatapclvs evidence of the effect of environmental pollution. Several surveys haws been Mdarcahsn to see If any sash esses hove occurred. Baric e a^. (I97*) sad Bllador and Ferahagea (1971) sought for essae to the vicinity of plants handlist 96 la Yufoelawla sad Sweden sod fowad nans, telderop (197B) fswad B eaaflenad esses Is lollaad not attributable to theretraec or areaale sad eould trace "ns cootact with vinyl chloride", hot they asda ns specific mention of the patient's place nf residence. Baxter (1977) found U confined cases In Great Britain over a 12 year period, nan of which wan in n ana who had lived half a klIsaacre froa a V9C asasfsstarlat plant, sad Brady (1977) found 19 cssso la Saw Turk Bests soar an It year period that could sot ho the fitarn of 1.4 a 10"! citad hy Booth at el. (1915) scene to have been a misprint far 1.4 x 10 . ucc 000291 40 attributed to any known eauoo, 5 of which were la people living within a lit of plants manufacturing or uolag VC. Tho overall laddanca rotas In those lest two studies are not unduly high, but tho occurrence of as many aa 6 caees la people living so close to nenufaeturlng plants Is surprising. Irady and his colleagues..ooreover, coupsred their series of patients with notched controls and found that none of the control subjects lived equally close to a plant. Two of these ft neighbourhood cases (1 In Ingland and S In Ifev York State) cannot be attributed to environmental pollution with VC, as the non who developed the disease bed lived near the plants for 6 and 8 years respectively before developing the disease and this la too short a tine to allow for the necessary latent period. The other 4 eases, however, all occurred after 13 or aore years local residence and the discovery of these cases strongly suggests that pollution of tho environment round plants manufacturing VCM or FTC nay have caused a minute hasard to the general public. Current concentratlona mad VC handling plants are certainly such lower than those reported previously by the IFA. leceut Irltlsh measure ments made within a fan hundred netres of the VCN areas have given average values below tho dally limit of detection (3 ppb) for throe of five plants, the readings at tho two others bolag 20 ppb (100 n outside the boundary fence) and It ppb Just Inside It (Tuner, Fayee, Beshby, 1M4) though substantially higher values wero recorded on two eccaaleea associated with ceanlaslonlag of one plant and aa accident at the ether. On any reasonable criterion the hasard to the general public (If thorn la any at all) oust be negligible (loyal Society, IMS). So other hasard to the general population, ocher then a hasard of cancer, can reasonably be postulated. ucc 000292 41 CONCLUSION Occupational exposure to concentrations of vc of the ordor of 1,000 ppw haa caused the claaalcal ' vinyl chloride lllnaas'. Cxpoaurs to such levels, or perhaps lower levels of the order of several hundred ppw, has caused anglosareoas of the liver after a latent period that la saldoa laas than 13 years. The evidence now available frow following up aany thousands of nan occupationally exposed to vinyl chloride suggests chat a snail risk of lung cancer any also have been produced, which Bust, however, have been substantially lass (In absolute tens) then the risk of anglosareoas of the liver. The belance of evidence does not support the Idea that VC has canoed cancer In any other organ, but It Is lnpooolhle to state categoric ally that It has not. Clearer evidence, ono way or the other, eould be obtained If the various Investigators would prooeat their results In eoaparable ferae that would allow than to be eoabloed and subjected to the standard testa for an occupational hasard. There Is no evidence that exposure to VC haa caused any other dloeaae. Suggestive evidence hss been obtained that past levels of onvltoonantal pollution with vinyl ehlorlda any have caused a few cases of cancer la the general public. ucc 000293 urftnas 42 Aik nODOCTS AID CBKKXCAU, I>C. < 1976) cwwti w tin wwnd Standard for vinyl chloride. Latter to 0.1. Goodwin, CPA, 23 Septeabcr 1976 (cited bp korr, 1976). SAkk JT. (1991) latard evaluation and rlok UMiwiftt for vinyl chloride. kopert to Air Product! and Chaaieals, Ise. (unpublished). kASCk ME, TOaaUH MS, omirr TP. (198S) Pulaonary function and respiratory syaptose Is polyvinyl chlorldo fabrication workers, poor kw kaao Pis. 131. 203*209. IAXm tJ, uraon PP, MmSDEH M, icnm PJ. (1977> Aafloaercooe of the liver is Croat irltala, 1963-73. rit oad j. , 919-921. STOMP 6 NOLMUkC 1. (1973) Two possible cases of aatlesartooa of the liver la a imp of Swedish vinyl chloride workers. San 9T lead Scl. 246. 249-230. ULU S, SSXTAZZI PA, COKSA P, PDA V, NALT0M1 C. MASHA A, PXEASTO k. UCXAMK1 A, V10Q1TX MA. (1996) Iodacloe nolle nortallta del prodottorl 61 PTC la Xtellet dlsefao della stadia a prlal rlsoltotl. Cancer Utters (la press). SSMITT I. <1966) leslster of cases of agglossrcoos ef tbs liver doe to aspssore to vinyl chloride* (oapobllshed) turn J. ununu p, iAsm r, cexemalo p, sum w, dotes t*. ixsmop N, P0US A, VZAMHA S. (1977) A^lecercoaa of the Uvort an aptdwslologle sorvey. J Mat Coocer lent. 39. 1393-1393. HPPtlk PA, 9000 9, IZStn C, SOASIC l, GUAM 0J. (1979) Mortality aspsrlssee of workers la e vinyl chloride aoanasr prodactloa pleat. J Occno Mod, ii, 193-203. STUS S, 1MCM0LM G, EMCLUMD A, wnnou P. (1964) Mortality sod ccacer aorbldlty la e snap of Swedish VQI sad FVC prodactloa workers* Invlroa kith Persnect. 17. 167-170*. uco 000294 teforen&ea eoaed: 43 OOOPKK DC. (1991) Ipldoalologle *tudy of vinyl ehlorldo vorknr*: Mortality through December 31, 1972. Environ 11th Portooct. 4i_, 101-106. CUECH Jl 6 J08MS0N MR. (1974) Aagloeereoae of llvor la tho aaaufActure of polyvinyl chloride. J Occuo Wad. 16. 130-131. OALDUDP Ul, nm SC, MAS C, H01CXHU1ST fl. (1976) Angloearcoma of tho llvor. Lancot. 246. OOHRS TO, STALUWIS U, PIAKOWSKl Rf, U1AATB 01. (1977) Tlavl ehlorldo. birth dofoeto. pad fotal aoataaoi a critical review. toport to tho loelety of plaatle ladootrloo, lac. (unpublished) DOCK W, CAATS1 JT, COOMKS CJ. (1973) Mortality atady of aorkara la a polyvinylchloride production gloat, laacot. , 1197-1199. KLIIMX CC, rniucn C. (197I) Pilot study concernlna tho aortolltv la Duranda Comunity. Svodlah lataro Cooaorvaacy loord. (eltod by Sort, 1991) MVIIOMMBITAI. OUin ASSOC1ATSS (19*6) Aa aadoto of aa ooldoololoolcal atadv of vlavl ehlorldo oorhora 1962-12. Pinal report to tho Choalcal Maaafactarara Aaaoclatloa. Kavlreuaeatel lealth Aaooelatoo, Oohlaad, Callforala. uniwiiai raonenw ACXXCT (1973) gtaadard aoooort doct^oat aad iwiwMii laaoct atataaoati oalaoloa ataadard for vlavl ehlorldo. SPA 690/3-73-009. gavlroMoatal protoctloa Agency, Waohlagtoo. KOTfAKS nvnonntAL ttALtl (1971) loldololoalcal atady of vlavl ehlorldo oorhora1 final rooort to Maaufactarlaa Chemists Aaaoelatloa. Igaltahle Environmental loalth, lackvllla, Maryload. TALK I, mus K, IOAK BB, T10MAS II, POPPgt I. (1979) tpldoalology of thorotraat-Induced hopotlc aagioaareemaa. Kavlroa lap. 11. 13-71. ucc 000295 htirneu ecntd: 44 rxtamn j, tens c, ierheistek x. t ai. 0976) cPidaioiotic note* nd reporta: antloecreoea of Che liver - Wlaconalc. Morbidity and Mortality Weekly hapert. 23. 37-5*. rODUM 0, IWCT >, RATVOU J, DOLL t. (IMS) Expoaure to vinyl chloride and aagloaareeaa of the Heart a report of the regleter of caaaa. rlt J Induatr tted. *2, 730-733. FOX AJ 4 COLLIE* PF. (1977) Mortality experience of vorkere exposed to vinyl chloride aoaoeer la the acoofacture of polyvinyl chloride la Greet Britain. rlt J Induatr Mod. 34. 1-10. cuxm von E. UHL V, WEE* I* (19*2) Flayl-chlorld expoaltloa ond nortalltit deotocher chaelaarheitor la vergleich car nortclltlt nlchtexponierter choaiearheiter ead FVC-verarbelter. Zbl Arbattened. 32, 44-62. 8EAXE Of, FILE H, CEXECI A. (1975) Gharaeteristica ef caeeo of aagloearcoae of the liver oaoeg vinyl chloride eorkera la the Dalted Statae. 4aa WT Acad Eel. 246. 231-236. WELDAA* IS, UMOiiP A, MBIUW A. (1964) Xacldeeca of eaacar anonf vinyl chloride aad polyvinyl chloride vothere, irit j Xodoetr Med. 41, 23-40. XMTEH4TX<MUL 4GKMCT FO* BEWAACN ON CMCEI (197*) IUC naanaraoha oa the \ ovaloatloo ef the carciaoneaic rich ef chenlcalc to h^ana. Vol. 19, pp.377-436. International Agency for laeeareh an Cancer, Lyon. JONES 89 6 MXTB FM. (1982) frogreeeien of vinyl chloride ladoeed hepatic fihroeia to angioeercena of the liver. Brit J Indoatr Mod. 39, 306-307. JONES U. (1906) A aortal!ty otody of vinyl chloride nonaner oorherc anployed in the Belted Elngdea 1940-1964. (la prose) ucc 000296 Kaferencea contd: 45 LAPUXCHZ A, cum F, COBTASSOT J-C, UKOCZinx C. (1987) txpoaura to vinyl chloride mmmi: report on cohort atudy. Iaotltut Guaceve toueay (to bo published). LIUS I, AMDKKSOM H, KTIXU A, SELIKOFF I. (1976) Fulnonary choncoa auong vinyl chloride polynorleatien workera. Cheat. 2. (Suppl.), 299-305. LLOTD Ml, 6A0L0 S, COFLAHD L, SOCTAA CA. (IMS) Ipldonlologleal atudy of lung function of eorkora at a factory nanufacturlng polyvinyl ehlorldo. Brit J Induatr Hod. 41. 329*333. NaeMABOK, I. (1977) Flnvl chloride and hunan raaroductlon. Keport to Chleal Manufacturer* Aaaoclatlon. (eepubllahed) HALTOtf C, CXLXUtrX A. CIAMI l. CniCO F. (1974) Tlnyl chloride carclnogenealat current reaulta and perapectlvea. Wed Lav. 65, 421-444. NAUOn C. 4 UFKKWS C. (1973) Carcinogenicity bioeeeaye of vinyl ehlorldo current reaulta. Ann It Acad id. 246. 199*219. KALiovz c, unxm c, caimn a. com c, caiutti o. df4) Ixparlnantal reaearch on vinyl chloride carclnogeneala. Archlvea of laaenrch on Induatrlnl Catclnononocia. Tol. 2. Ida. Nnltonl C. 4 Wahlnan HA* Frlncaton Scientific Fubllnhern, Frlnceton. MAUTSUIt KJ, DSLBAfli W, NOUSt K, OOIOC F. (1973) Ononual aplenonegallc liver dloeaae an evidenced by perlteonoocepy and guided liver Mepey anoeg polyvinyl ehlorldo production eorknra. Ann WT Acad Scl. 144. 93-134. NAismus u, onjua w, must t, job s, lams a. aonn m, ulimab 0* (1973) Chronlacb-toslacha leborachldea bol Arboltorn la dor FFC-Froduktlon. Ptoch Mod bechonachr. 99. 2311-2314. ucc 000297 kaforaacaa eontd: 46 HASUDA T. (1979) Long tar* aortallty atody ef vinyl ehlorlda and polyvinyl chloride vorkara la a Japaaaaa plant. Arch Inductr Hv Toxicol. 30 (Suppl), 403*409, MI1K0VA E, MIHAILOVA A, MOSKO N. (1978) Oil Zdravaooaa. 21., 440. (cltad by larT, 1986) MOLAN I, MOUH C, R0LMBC8C 8, CLOPSSOM S, B0LM0HD L, MOOSINC I, WESTEMOLM P. (1981) Mortality and eaocar rataa aaoag vorkara la tba Svodiah PVC procaaalag laduatry. Environ Hlth Paraaact. 4U 143*131. MOMSO* U, PETEKS JM, J08MSQ8 KM. (1974) Proportional aortallty aaoag vlayl-ehlorlda vorkara. Laocat. 2, 397*398. ltAKAMJlA K. (1983) A aortallty atody of vinyl ehlorlda vorkara la Japaa. Saanro ika Maavfca Eaaahi. (Sappl). 49-37. N1CMU0H HJ, HAMDHD EC, gUMUM I. SEUEDPP U. (1973) Mortality asparianea of a eobort of vinyl ehlorlda-polyvinyl ehlarlda vorkara. Ann KT Acad cl. 148. 223-230. Hicaouo* m, mtHElSKCEt PE, TAU S. (1904) Trooda in caaaar aortallty aaoag nrtan la tha apathetic polyaera laduatry. lot Indnatrlal haaarda of olaatlca aod avathatlc alaatlaora. pp. 63-78. Llaai Mav fork. OTT Ml, uron tt, HOLSEE P8. (1973) Vinyl chloride axpooora la a caotrollad ladvatrlal aavirooaaat. Arch Eaviroa Hlth. 30. 333*339. PIXH1E e, TAMICMOM JP. pong I, fPEUm J. (1979) Etoda da U aortallta chaa daa traveillaura aspoaaa aa chlorvra da viayla. Arch aaladlaa orofaaaioaallaa da aadlclaa da travail at da la Eocurlta Eoclala. 40. 1131-1143. OTAL SOCIETY ETTOI CAOOP (1983) Biak Aaoaaaaaatt laoort by a Study Croon, loyal society, London. ucc 00Q298 tefereacea coned: 47 SAIIC N, KULCAI 2, 20EICA M, CCLIC J. (1976) Malignant tori of Che Uw.r cad luaga la aa crcc vlth a PTC lnduatry. Environ gleh Pertenet. 17_, 189-92. SOCTAI CA. (1980) Epidemiological atndy of respiratory dlaacaca In vorkari exposed to polyvinyl chloride duat. Thorax. 33, 644-632. TA1EIS1AH II. 6 Cl/m HI. (1974) Mortality aeudy of oorkara In the aanufacture of vinyl chloride cod lta polymera. J Occuo Mod. 16. 309-318. ranXAITLT 0. i ALLAH) P. (1981) Ccocor mortality of o groep of Canadian workora axpoeed to vlaylehlorldo monomor. J Occuo Wad. 23, 671-676. TUttEI CA. PATH AP 6 10107 St. (1984) Determination of Rblant Lovola of Vinyl Chloride Monomer (TOO around Waanfactnrora in tho OK. Part 7. 1984. Barren Sprint Laboratory, Deportment of Trade and Induetry, Stevenage. VIOLA PL. IICOTTI A. CAPOtO A. (1971) Omenfenle reaponae of rat thin, lease, and bonne to vinyl ehlorldo. Cancer lea. 31, 318-319. WAXHIILIl U, STUMCEK H. HAOORI JK. Jont J, PALI I. CAIXII C. (1976) Deeplaatic rlak ament norkare empoeod to vleyl ehlorldo. Ann IT dead. Sal. 271. 40-48. ucc 000299 48 Table 1 NOMinS or DEATHS FIOH DIFTEUNT CAMCEIS OISOTED AMD EXFECTED: foot STODIES Cause of death USA Obs*/Exp OK Obs/Exp Canada Obs/Exp . Cancer of buccal cavity 4 pharynx oesophagus stoMoeh largo intestine rsctia liver llwor 4 gallbUddar pancreas ochar digestive larynx long other respiratory bone akin (non-nalanona) Ml SIIIMS prostata tastla bladder kldnoy othor 4 oaapacified urinary brain ay** 4 On thyroid lynpho- 4 retleuloaarcona odgkla's disease letibasnla anltlpla nyoloM char lypbatla char censors All Mini 13/11.35 7/8.07 11/14.01 21/28.73 no 33/3.77 17/18.40 118/113.87 3/4.38 2/1.81 4/7.34 14/13.20 3/8.44 12/3.04 23/12.74 12/7.38 3/3.43 14/13.34 > ) 11/8.37 44/417.30 4/3.38 4/14.34 24/23.31 3/13.34 11/10.43 11/1.34 7/3.88 4/2.21 ) 81/32.12 ) ) 2/1.74 12/3.33 , 2/1.38 * 14/8.00 { 3/4.10 J 3/4.23 > 4/4.18 ) ) 2/0.43 4/2.33 ) 3/2.30 ) 7/3.14 ) 2/2.33 ) -) 18/8.12 0/0.44 M M ea 1/0.14 4/5.21 2/5.74 m 1/1.33 0/0.40 ' 1/1.47 2/0.33 383/341.73 233/338.40 30/14.37 Italy Obs/Exp 1/0.8 1/0.6 3/3.0 0/1.2 1/0.6 0/0.7 1/0.9 12/6.1 Oft , 0/0.2 1/0.7 * 1/0.7 0/0.4 0/0.7 9/4.3 30/21.1 *0boor*<a4 doatha Multiplied by 1.0474 / for ducki iltkMt dlMivmd eauo. ff to Morose Integer to allow ucc 000300 49 Table 2 9UKBEKS OF DEATHS FftOH NON-MALXCKAIR AND ALL CAUSES: POOI STUDIES Causa of Seach USA obt*/bp UK Obe/lxp Canada Oba/bp Benign A achat unspecified ttaoura 4/5.OS Cerebrovascular Slaaaaa 75/91.93 laebooole haarc Slsaaoa 321/597.73 Othat elreulacory disease 157/123.55 Bronchitis** 44/22.S3 Pneuoonla 14/31.941 Ocher respiratory Slaaaaa 15/32.44J Clrrhoala of liver 37/5S.0S OChar digestive Slaaaaa 27/39.32 Dlaaoaa of genltO'erlnerp ayatao 11/20.41 Other Slaaoaaa A7/115.SS Suicide 49/32.781 AcelSeata A ocher violence 130/173.19J All noo-oallgnant eoooaa All cauaaa 1153/13S3.34 1334/1705.27 -' 274/2SS 103/141 34/44 40/41 3/3 . - 43/74 40/30 343/443 7S0/S94 25/31.47 4/3.21 *4/3.83 * 2/5.40 2/10.38 39/34.74 39/71.07 Italy Oba/bp 0/0.5 19/27.4 3/5.0 4/3.2 3/2.7 2/7.0 1/0.9 4/7.7 34/54.4 44/77.3 * S ftOtMtt CO flkU 1 MfaphjNM to OS Soto * Includes 2 certified oo elrrbeala of the Uwt oklth proved to bo aagloserco of lleer loeloSoo 1 ucc 000301 50 Tabla 3 moktality non cajtcu or tu ura aid own causis is Vim C8L0UBI UOUXU II 49 PUMT8 II A STUDIBS lo. of daatha Staadardliad Mortality ratio Obaarrod Ixpaccad (SMO Caaear of llrar* * of othar altaa Othar diaaaaaa Accldoata, polaaniega, aad vlolaaca All eaoaaa 59 60V 1547 226 2441 S.4S 599.35 1144.89 295.15 2747.14 698 102 84 77 89 Xnclndlat eitein of tba (lUklaMn la tba OS aariaa UCC 000302 51 Table A MORTALITY FROM 9ARI0CS CANCERS IN VINYL OtLORIOC WORKERS IN 49 FUMTS IH 4 STUDIKS Source of laforaadott Typo of cancer No.' of daacha Observed Expected 3Ha 1.2.3,* 1.2.3,* 1.2,3,4 1.2,4 1,2,3,4 2.4 1.2.3 2 1,2,3,4 1.2.3,* 1.2,3.* I Cascor of aouth and pharynx " " dlioetlee eyatoa (othar than Hear) " " raaplratory oyataa - - luac " " genlto-urlnary ayaeoa Caacar of brain " thyroid " " lyaphatlc and haaeeto- polatle oyataa Othar cancara All caaeora othar than Hoar IS 16.57 109 123 134.39 81 223 229.36 97 211 214.09 99 70 62.SI 111 2 1.94 - 29 19.34 148 2 0.43 at 37 S0.S7 112 S3 43,24 131 409 399.35 102 1 OS etody 2 OK atady 3 airilw atady 4 Italian atady ucc 000303 labia 5 NOBTALIYT* VMM MMOM CMCBBS IN VINTL CNUM1DB NDMCIUt SOTPUEHOCTMT KVIDCNCC Tppa rf caacar PK (11 |Uati) Oba. Bap. a. of Oaatba* la atallaa lat (1 plaat) (I^Uat) Italy <1 plaat) Oba. o*r* Oba. Bar. Oba. Cap. Pour coaatrlaa (14 plaata) Oba. Bap. SMC Caacar of ll|aatlia ajrataa (atbar Cbaa lirar) a* - 1mm Malaaa of brala m tbpralO lpapballc aaJ boaaatopolactc ajwtai Ocbar caarara (arbor tbaa ilvar) 35.0 13.5 2.1 10.3 10.1 31.0 20.0 1.3 1.1 24.3 All cccare atbar Cbaa Hoar 01.0 09.1 3* 1.44* 5 2.04 4 0.10 -2 0.10 -- 0 14.93 22 20.10 -- 1 1.2 39.0 34.44 113 3 1.10 -2 0.33 --- 0 1.5 0 0.1 --0 0.1 31.5 4.0 4.1 2.0 14.5 30.12 0.09 1.03 0.14 0.4 103 - 190 -- 4 1.0 22.1 41.43 53 5~ 2.II** 3 5.1 119.0 111.41 102 ucc 000304 * ImMmci mI caaaa la Namaclaa itlf * Caacar of latoatlaa oaljr Caacar of luaf a*l brala oaljr 53 Table i MORTALITY ntOM LONG CAMCKt I* TO OS AMO OK SC1Z1S IT auiucmzsncs ULEVAMT TO All 0CCUPAIIOKAL RAZAKS Data characteristic Mo. iat deaths la category 1 Obs Exp. 2 Obs. Kxp. SHE category 12 Observed 20 yrs or eore after first eeployaeat (1), others (2) Eaployed 10 jrs or aora la USA (1), ethers la USA (2) Eaployed before 1954 la UK (1), others la UK (2) Sear replayed as autoclave worker la UK (1), others la UK (2) 114 U5.9S S3 93.63 100 91 53 32.43 63 43.44 103 99 52 31.39 29 40.30 101 72 IS 17.OS 45 74.S2 94 87 See footnote Co Table 1 for observed deaths In 08A. ucc 000305 54 Table 7 HMTAL1TT TIM SELECTED HOM-KALICKAIfT CADIES AMD ALL CADIES im root peimcifal studies cdmiimed Source of Inforaatloe Type of caacer 1.2 1.2 1.2.3.4 1.2 1.2 1.2.3,4 1.2.4 1.2.3,4 1.2,3,4 1.2,3,4 1.2,3,4 1.2,3,4 *Broaehltle, aapbyaeaa Other reeplratory dleeaee All reeplratory 41aeaee taeheaalc heart dleeaee 'w0ther circulatory All circulatory dleeaee Clrrhaela of Hoar Other Alaaaae All aoa aeUgeeat dleeaee All eateraal caaoea All aaa-aallgaaat aaaaea All caaeoa Mo. of death* Oheeroad Expected SnX 0 71 140 7t7 232 1103 44 230 44.83 123.78 200.82 883.73 244.33 1208.33 44.24 348.07 120 54 SO 90 93 91 49 43 1347 220 1844.30 283.13 84 77 1773 2441 2138.43 2747.83 S3 89 * 1 US etudy 2 UK eta*y 3 Caaa41aa ataiy + Ireaehltle la UK atody, aykyMi la US etady 4 Itallaa atady Iaclodaa earabrovaacalar dleeaee la UK ead Itallaa etadlae ucc 000306 55 Table 8 WOtTALITT PIOM C8ION1C 08STK0CTIVE LUNC DISEASE* IT CEAtACTEEISTICS RELEVANT TO All OCCUPATIONAL BA2AKD Data characteristic No* of deaths la category 12 Obe. Exp. obs. Exp. SKR category 12 Observed 20 pro or oora after flrat aaployaeat la DSA (1), others la USA (2) Eaployed 10 pears er eore la USA (1), others la USA (2) Eaployed before 1936 la UK (1), others la UK (2) Ever eaployed as autoclave worker la UK (1), others la UK (2) 30 1S.S u 7.0 190 157 It 10.9 23 12.0 147 208 26 30.17 10 13.60 86 74 3 6.33 33 37.22 46 89 *Deacrlbed aa aaphyeaaa la DS itadr aad broachltls la UK itudy. *H*n mr employed aa a baftr or drier, ahlch avoid have eaoaod tta greatest oeeopatloaal exposure to PTC dost, asparlaoead l death froa 'brooehltla' again*t 4.98 expected. ucc 000307