Document kvJod6b1Q9yY7Kp2wXr6vjzJ

FILE NAME: Shipbuilders Council of America (SHIP) DATE: 1977 Nov DOC#: SHIP047 DOCUMENT DESCRIPTION: Journal Article - Mortality of a Cohort Exposed to Chrysotile Asbestos aa PLAINTIFF'S EXHIBIT )] CAP-1390 Mortality of a Cohort Exposed To Chrysotile Asbestos William Weiss, M.D. A 30-year historical cohort mortality study was made of 264 men hired during 1935-45 w ho worked in a chrysotile asbestos products tactory fo r one year or more and were alive (anuary 1,194S. Follow -up was 94% complete. The overall standardized mortality ra tio (SMB) was only 0.61 w hile the SMR fo r all cartvets was 0.75, lo r lung cancer 0.93, and fo r gastrointestinal cancer 1.05 Two men died of asbestosls. The overall SMR was higher fo r men w ho w orked five years o r more than fo r men who worked one fo fo u r years but the 30-year m ortality rates were (he same a fte r age adjustment. For asbestos-related diseases the differences in w ork duration had no effect on mortality. The data show a favorable m ortality experience for men exposed to chrysotile alone. t asbestos is a serious health hazard has been well estab- vned hut the imluence o f fiber type and kind of exposure has not iw n Hlccjudtelv delineated.' On a human level the study of these Meti .rj h i . neen difficult because exposures to a Single fiber type .r.- rare usually limited to mining. Most workers manufactur- is JintM os products have been exposed to a mixture o f fiber types, however, the information tfiat is available suggests that .litvsuidi, is not as dangerous as crocidolite.-' While It is difficult to ilfiis conclusions from comparisons between studies o f different jwi'uiat.ons in which exposure conditions are not exactly the ime something may be learned by recording observations of oc- --p4tion.il cohorts exposed to pure (orms of asbestos w ith the nope that comounding variables may be kept to a minimum. Sehkoti et at* * have reported observations of mortality in a eo- <wrt or men employed in an amoslce asbestos factory. The oppor- .u.iity to do a similar study of mortality in a cohort o f men em ployed ;n a factory using only chrysotile asbestos provided data 'or this report M ethod An asbestos manufacturing plant opened in 1896 has used only . irrysot-le asbestos throughout its history. The employee records *v*n me O .Jit*r a . ' i SVfc-UfcM* nt 0<cupit*yul Medicine Department oi Medicine. 6401 New Hahnemann Medical College and HmpiuL 15th and Vine Street*. H102 ...Mi m Occupational Medicine/Yol. 19. No. 11/Novembet 1977 for the period 1935-45 were reviewed. There were 332 male pro duction workers who worked one year or more in this period: 267 were first hired during the period 1935-45 and three of these men died before fanuary 1,1945. Thus, there were 264 men who were first hired in 1935-45. worked at least one year, and were alive on Ianuary 1.1945: this was the cohort for the study. Its age distribu tion in 1945 was: under 20, 6%; 20-29, 50%; 30-39, 20%: 40-49. 9%; 50-59.10%; 60 and over. 6%. The distribution o f work dura tions from hire was: ,1-4 years. 56%: 5-9,16% : 10-19,11%' 20 or more. 17%. Deaths were determined through the Social Security Ad ministration as of December 31.1974. Vital status at that time was known for 94% of the total group. There were 16 men whose social security records revealed no information for variable periods prior to December 31. 1974; these men were assumed to be living throughout the follow-up period. There were 66 deaths and death certificates were obtained.for all but four. No smoking information was available Causes of death were coded by the author. The number of man-years of observation was computed for each five-year period from 1945 through 1974 for each o f two w ork duration groups: 1-4 years and 5 years or more. The number of observed deaths was compared w ith the number expected and the standardized mortality ratio (SMR) calculated. The expected number o f deaths was obtained by applying the mortality rates of U.S. w hite males (almost all o f the workers were white) for the middle year o f each five-year period by five-year age groups to the man-years in each cell and cumulating the results. The statistical significance of differences between observed and expected deaths was estimated from the Poisson distribution.' From 1935 to 1960 the plant produced chrysotile asbestos papers and millboard, bound w ith starch or occasionally Portland cement. The fiber was received from the mine in loosely packed jute bags. Rail cars were unloaded by hand and bags stored in a warehouse. Stock was prepared in beaters, each bag being emp tied into the tub. the bag shaken to recover as much fiber as possi ble, and stacked for later return to the mine o r sale. Stock was formed on multi-cylinder wet machines. Finished paper was con verted to rolls, corrugated paper, corrugated sheets o r blocks, or aircelt pipe .^isulaticjn Paper products and millboard were cut by 737 # * 4 j -r ! ASARCO ALV 0 0 0 6 7 2 1 saws w ith limited ventilation. Loose fiber was cleared by airhose. Fiber could be seen in the air throughout the plant except in the finished goods warehouse, boiler house, and maintenance shop. Fiber and d u st were visible on the floor, work surfaces, and in terior structural surfaces. Cleaning was done by dry sweeping. At the end of each shift employees used an airhose to remove fiber from hair and clothing. W ork clothes were usually worn home. After 1960 the mine began to compress fiber into blocks shipped in paper bags, and dust in fiber handling was much re duced. The principal stock preparation unit was changed from a beater to a pulper. The paper bags were slit at the edge of the tub and the block broken into chunks which were dropped into the pulper. For certain products the bag was pushed unbroken into the pulper. More and more products were bound w ith synthetic latices and resins, producing less dusty items. Converting opera tions declined. Production o f aircell pipe insulation ceased in 1968 and millboard production was discontinued. All saws were re moved. As a result, no dust is visible in the air. on surfaces, or on clothing. Plant floors are cleaned by vacuum or w et sweeping. Most employees change clothing and shower at the end o f the w ork shift. Fiber counts were not made until 1972. so that no exposure in formation of a quantitative nature is available for the years when dust concentrations were obviously high Fiber counts in recent years have been generally under 2/cm'. Use of an exposure scale was not feasible from the available records. Results The duration of follow -up from the date of hire was 30-39 years. In the last decade o f observation 45% of the men had attained the age o f 50 or over. Twenty-five percent of the men died (Table 1). The group that worked 5+ years had a crude rate which was about 50% higher than that of the men who worked less than five, years but this was due to a difference in age distribution. Using the total group as the standard population, the age-adjusted rates were the same. Table 2 gives the number o l man-years and Table 3 observed and expected deaths as well as SMRs. The overall SMR was only 0.61. During the first 10-vcar period it was 0.56 and rose to 0.62 for the remaining tw o 10-year periods. The SMR for the men who worked five or more years was higher than that for men w ho worked less than five years but both were statistically significantly lower than 1.00. The difference between the tw o groups was large during the lust 20 years but almost disappeared m the last ten years. Cause-specific SMRs for diseases of interest ate given in Table 4 for the 30-year period The number o f deaths for each cause were to o small to provide reliable ratios in each 10-year period. The SMR for all cancers was slightly higher than the all causes SMR but the SMRs for lung cancer and gastrointestinal cancer were close to T Ttbla I. - - 30-Vaar Mortality Dalai. r (a * CoKarl f 264 Mon from 1945 thfOBjh 1974 by Nembor of Vim WuM it a Chrjwtili Aibntoi KiiuifKIutial Plan!. No. ef YotffVforVtd Ns if fMA Oeathi Crvda Ko. K % M 148 31 116 IS Tdd 244 - 66 20 9 745 302 2S4 250 1.00. relatively higher than the all causes SMR There u*e risn deaths attributed to severe pulmonary fibrosis on the de.it* uvlilicates; for the purpose of this analysis both wore categorized as due to asbestoses. Since asbestosis does not occur in the ihsenci' o f exposure to asbestos. SMRs could not be calcualted Table 5 shows asbestos exposure periods, date of death, igc ai death, and cause o f death in ten men who died of diseases which might have been related to the work exposure at the plant. Al though tha duration of exposure was short (one year' in one rare of lung cancer and the elapsed time from onset o l exposure to death was short (17 years) in another, asbestos could have been a factor In the causation of all four cases o f lung cancer. In one ol the tw o cases o f gastric cancer, the elapsed time from onset of ex. posute to death [four years) makes asbestos unlikely as a raure and the same might be true o f one of the tw o cases of co'on can cer in which there was only 12 years between the onset of ex posure and death. However, information is lacking on whether there was asbestos exposure elsewhere before or alter the time worked at this particular plant, making the total duration of ex posure in these cases uncertain. Discussion Methodologically, it should be emphasized that this 1925-45 co hort was established w ith men who began work in 1933-45 to avoid bias. If men who started work before 1935 and were work ing in this period had been included, this would not have taken in to account those men w ho might have left the company fot reasons of illness prior to the period in which the cohort wax de fined. Although Wagner et al' are of the opinion that chrvmtile the least dangerous form of asbestos w ith respect to adverse health i>t. focts. the evidence is very scanty and there are .a numliw of methodologic problems in developing a well-founded judgment among the cohort studies in the literature. A minor problem in almost all investigations of this nature is exemplified by the current study 6% incomplete follow-up and 6% failure to obtain death certificates to allocate cause of death However, these deficiencies are so small that even if all the in dividuals whose vital status was unknown at the end of 1974 had died and all the deaths for which certificates were unobtainnd among the known deaths were due to asbestos-related disease the conclusions would not change. More serious problems are inherent m differences between and deficiencies in studies reported in the literature. These make comiwrisons difficult. Ideally one would like to have data on large groups'of workers exposed to different forms o f asbestos under conditions in which all other factors related to the host the en vironment. and the agent are held constant. Such utopian condi tions do not exist, however. Peihaps the most important problem lies in confounding the (actor o f asbestos type by the factor o f w ork conditions. It is Table 2. -- Number of Mon imj Man-Yom of Observation. In 1 Cohort ef 264 Men from 194$ thfoofh 1974 by Number of Yam Woriurf it * ChiywtlH Attain Miwrtoetarini PUrt fe lO-ytor PinoJt. No. of Yam No. ol Men No. of Mtfl.Yom tfortsd Jin. 1. 1945 1945-54 1955-64 2965-74 Tetri H 148 1.456 IJ99 1216 4.131 $+ 116 1127 1027 872 3020 W 2 3583 2.421 2147 7 151 - 738 | Mc-*'li,"-nf-a Cohort Exposed To Chrysotile Asbestos/Weiss ASARCO A LV 0 0 0 6 7 2 2  Table 3. -- Standardized Mortality Ratios (All Causes). If! Cohort Of 3 f t M from 1945 throvfti 1974 by Humber of Year* Worked at o Chrywtile AsbMtOf UiAutaeturini Plant in 10-Yav Periods. No. < Pitted Tun 1945*51 1955.64 1965-74 Worted Obi. JP* SMft Obt. Em.4 sm r Obt tip.* SM U s 1220 041 9 1972 046 17 2B.75 059 5+ 8 1119 on 13 1571 083 14 2111 066 ToUf n 2339 055 22 3546 062 31 4991 062 * f ipreted nutnoer of deilhs derived bom U S fP < 0001 t P 005 white ma*e mo^tafitjr itilis ic t by Jge and peuod Toni Obi. Era.- S ifl 31 6066 o sir 35 4804 cm 66 10B76 061 Tabta 4. -- Standardized M ortality Ratios for Aibestos-Retatod D iiaanr. In > Cohan ol 2 (4 u i tram 1945 thiauih 1974 h i H u n t* al T u n iro n s ) i t a C hiym im a ia a to i auiiiiuetitm n r u n . Cause of Death and No. et Years Worked 411 tJftCM < rr tuns cancel M 5- u l unco H 5+ auw tous 4-4 5* 1945-54 Obt. p. 3 323 2 155 X 167 0 0 52 0 022 0 0 30 2 035 2 046 0 0 49 0 -- 0 -- 0 Period 195564 Obt. Eip. 5 53 2 287 266 123 058 0 65 127 0 067 2 060 -- -- 0 ** 1965-74 Obt. Eip Obt. 666 13 3 469 - 397 6 257 4 0 US 2 12! 2 0 161 4 J). 087 2 0 074 2 t -- 2 _ 2 0 0 Total tip. SHH 1741 0 75 9.11 0 77 830 0 72 432 093 2.16 093 216 091 382 105 200 100 182 110 -- - -- -- So'* ipected num&et of deaths cawed bom US white mate natality statistics by age and period proiwbK not jusrifcihle to pquate exposure conditions o f mining with those in manufacturing or those in use o f asbestos products. Ivon in manufacturing, conditions vary. Enterline and Kendrick' -ludied three cohorts in asbestos manufacturing operations of difterent types: building products, friction materials, and textiles. The r.sxa or death Horn asbestosis increased in the order given, in comp jtiio r. w ith US. white male mortality rates in the same period, the sMR tor ail causes was elevated only in textile workers and this cohort experienced the highest SMRs for lung cancer, digestive tract cancer and chronic respiratory disease. Unfortunately the cohorts were composed of people w ith varying durations of ex posure prior to registration and foliow -up was limited to 16.5 veats alter registration. The duration of exposure and of follow-up are birth important risk factors.4 ' Additional mothodologic discrepancies between studies ac count tor some o f the wide variation in mortality risks among asbestos workers. These include differences in the populations used to calculate the expected number o f deaths, a n d limiting cor rection of causes o f death on death certificates by means of ad ditional' medical information to the study population w ithout similar correction of the certificates o f the reference population. Recently inte rline10 adjusted the results of six studies for dif ferences m length o ffo llo w -u p . the reference populations, and correction o f death certificates, and found that much o f the varia tion disappeared in the SMRs. "A very s'errous problem is the lack o f dose information in most studies. There are few investigators (for example. Mewhouse.* Enrertinc et al.4 a n d McDonald et at11} w ho have attempted to develop ciata tor degree o f exposure and some o f the effort is Journal of Occupational Medicine/Vol. 19, No. U/November 1977 based on semiquantltative estimates rather than quantitative measurements. Changing processes over time may have a-pro found effect on dose, making for additional uncertainty in the ab sence of systematic determination o f dust or fiber concentrations. finally, seldom are the results of investigations corrected lo r variations in smoking habits, although smoking is now known to be an important risk factor. 5elikoff et al" and Berry et al1' have shown the importance o l cigarette smoking in the risk of lung can cer among asbestos workers and Weiss14 has reported its in fluence in the prevalence o f asbestosis among textile workers. Despite all these difficulties, there is a little information suggest ing that chrysotile asbestos may no t be as hazardous as other forms of asbestos or mixtures o f chrysotile w ith amosite or Tabla 5 . -- E ipow ra of Man wbosa O aatht flil(ht lav a b u n Dm la A sbastu. Cm Aibeites tw o w re Ha. 84fM Tern CitlM 1 Oct. 1. 1936 2 Sept 14. 1936 3 Oct 57. 1939 4 Od. 2. 1539 5 NOV. 24, 1942 6 O ct 25. 1942 7 J t a IS. 1940 8 Noe. 18. 1943 9 O ct I . 1935 10 Sept- 7. 1937 , f 1 tunf a r e a 4 lung anccr IQ t a g a r e a 17 bng a r e a ' 4 tttirie carter 3 j U c anccr 1 a k a anccr 8 colon anew 1 pulmonary to m b 4 pulmonary fiirMEt Dutk At< 82 48 51 59 62 85 56 68 S3 62 Ym Iro n f l a t Cipofun 24 22 31 12 4 20 12 20 28 3Z 39 ASARCO ALV 0 0 0 6 7 2 3  crocidolite. or both. Oata from the Quebec chrysotile mines and mills have shown minimal influence on mortality and the effect is limited to heavily exposed people. A better comparison w ith the present study is an investigation by Elwood and Cochrane1' w ho studied 1.16S men from 1936 to 1962. These workers were exposed only to chrysotile. making asbestos sheeting for six months or more, in the analysis o f men w ho were followed 15 or more years after first exposure, the overall SMR was 0 96. that for all neoplasms was 1.10. and that for pleural and pulmonary cancer was 2.32. Unfortunately the majoritv of workers were followed for little more than 15 years. Nevertheless, the risk o f respiratory cancer was much less than the SMR o f 15.38 recorded 10-14 years after first exposure by Selikoff, Hammond, and Seidman among 326 insulation product workers who were exposed one year or more to amosite.4 Enteriine and Henderson1 provided valuable comparisons w ith regard to type of asbestos and mortality in 1.348 retirees. The lung cancer SMR for workers exposed to chrysotile alone was 2.4. to amosite alone 3.9. and to chrysotile plus crocidolite 5.6. In the cement industry thev were able to separate those w ho were ex posed to chrysotile from those exposed to chrysotile plus crocido lite'. the SMR was only 1.4 m the former and 6.1 in the latter. w h ile the dearth o f pertinent information makes for an un satisfactory state of affairs, the favorable experience of the cohort reported in the current study suggests that the Hazard of chrysotile in asbestos products manufacturing is minimal. The results are consistent w ith what is known of the relative dangers o f chryso-. tile, amosite. and crocidolite. References 1, Bccklake MR: Asbestos-related diseases of the lung and other organ*: Their epidemiology and implications for clinical praefee .Amr- ^ On 114*187-227. 1976 2. Enterlme PEand Henderson V: Type of asbestos and fC<pKto'v. ar.cr' in the asbestos industry. Arch fm/Von Hlth 27:512-317, 197). 3 Selikoff If, Hammond EC. and Churg I* Carcmogenir'tv .if vtvKifp asbestos, Arch fnv/Von tilth 25.183-186. 1971. 4 Selikoff II. Hammond EC. and Seidman H: Cancer ri*k of in<ul,ii.<y workers in the United States. In Biological Effects of AsK*<to P. Bocovski. V. Timbrell. |. C Olson, and |. C. Wagner fed.', pp. 209*218. intefpatffml Agency for Research on Cancer. Ivon. 1973. 5. Pearson ESand Hartley HO: Biometnka Tables for Statisticians A*.ond Edition, p. 203. Cambridge University Press. Volume 1 Table, 40 19511 6. Wagner |C. Cilson |C. Berry C. and Timbrell V- En'demtoloev of asbestos cancers. But Med Bull 27:71-76, 1974. 7. Enteriine PE and Kendrick MA: Asbestos-dust e xpo se s at \a'*mj< levels and modality. Arch Environ Hflh 15:181*186. 1967 8 Newhouse M l: A studv of the mortality of workers tn an a<be$tcx he lory. Bni i IndiM Med 26:294-301. 1969. 9 Enteriine E DeCoulle P. and Henderson V: Respirators* cancer n rela tion to occupational exposures among retired asbestos worker* 8m l induU Med 30.162-166. 1973. 10 Enteriine PE: Estimating health risks in studies of the health effect; of asbestos. Amer Rev Ro<p 0 113:175-180. 1976. 11. McDonald |C. Becktake MR, Gibbs GW. et al: The health of chrysotile asbestos mine and mill workers of Quebec. Arch Environ HUh 28 61-6B 1974. 12. Selikoff If and Hammond EC: Multiple factor etiology of occupational lung cancer. In Proceedings of the Second international Svmoosium on Cancer Detection and Prevention, C. Maltom led,1, op. 152-160 American Elsevier Publishing C o. Inc.. New York 1974. 13. Berry G. Newhouse MU and Turok M: Combined effect of asbestos exposure and smoking on mortality from lung cancer in factory wo ko's Lancet 2.476-479. 1972. 14. Webs W : Cigarette smoking, asbestos, and pulmonary im - Rev Ro<p Dn 104:221-227. 1971. 15. Elwood PC and Cochrane AL; A follow-up study of w orker lr vn au asbestos factory. Btit I inrhtU Mod 21:304-107. 1964 Unemployment and Health i i . . . A study analyzing the effects of the economy on social problems was recently completed. It is entitled "Estimating the Social Costs of National Economic Policy: Implications for Mental and Physical Health, and Criminal Aggression" by Or. Harvey Brenner o f lohns Hopkins University . . . . According to Or. Brenner's study, a 1.4% increase (about 1 -1 2/ 'million people' in the number of unemployed, sustained over the six-year period between 1970 and 1975. resulted in 20.240 heart and kidney disease deaths. 495 liver cirrhosis deaths. 920 suicides, and 648 homicides. In all. the study cited that there were 36,887 deaths (27% were o f blacksl from various causes. In addition, there were 4.227 state mental hospital first admissions and 3.340 state prison admissions . . . . W e must make it clear that Unemployment rates . . . are not just isolated, meaningless, theoretical abstracts, but that they indicate a reality o f human stress, illness, and death. Full employ ment. therefore, is a healthy concept for our society. More people would live healthier and longer lives, during which they could do more work longer, that in turn would beef up the economy so that more people could work. -- from Erfitoml by Thptman . fvans. M D . m In tim l tt1 1fir amr/ican Vetfir,/ Ai.oculmn Wiv 2. 1977 i i 740 non .1 jMortality of a Cohort Exposed To Chrysotile Asbestos/Weiss ASARCO ALV 0 0 0 6 7 2 4 -