Document rpB3paJ1RMogaR9bkdDJg7paq

MAY 15 THE INFLlrENCE OF OCCUPATIONAL AND ENVIRONMENTAL ASBES TOS EXPOSURE ON THE INCIDENCE OF MALIGNANT MESOTHELIOMA IN CT. J.C. Levinsohn, J.W. Meigs, M.J. Teta, Ct. Can cer Epid. Uni*., Yale Univ. New Haven, Ct.06510, U.S.A. Medical, demographic, and occupational data were collected for the 229 cases of malignant mesothelioma ascertained by the Ct. Tumor Registry between 1935 and 1977 and for a random sample of approximately 700 de cedents (1935-1977) from the Division of Health Statis tics of the Ct. State Dept, of Health. Sources of in formation include the files of the CTR, death certifi- . cates, and city directories. A vital statistics tape of Ct, decedents was utilized to match spouses of cases. An asbestos-product listing with its associated indus trial and employer codes has been created from the oc cupational histories of cases and controls. A substan tial increase in registrations of mesothelioma over this 42 yr. period is evident with 852 of the cases enumerated after 1954. Changes in patterns of diagnosis and the possibility of misclassifIcation are suggested. analysis performed on the 56 confirmed cases of pleural mesothelioma and the 20 cases of primary pleural tumors (not mesothelioma) with respect to age, sex, survival, stage, and number of tumors yielded 59% misclassification. Geographical clustering is evident in certain industrial cities and Labor Market Areas. I Final results will include case-control comparisons with re spect to occupational asbestos exposure, an analysis of the health status of spouses of these cases, and the occupational histories of spouses of non-occupationally e: A0045C UCC 004886 A "(jaft ytuMXb ^IM'T? au/-* i^i?y^uM\ fcS) } *-tei ^ua*-t? * 3$> a^C- y^ftuUtP *VfcJf U/4*3u2Ijia^ l*4uK UyS^/J? Vi -kc. &**z t^2CC^i '<{1?. au^ ^ <iL2^LC2W34^ OM$ C?lXu^iX>4Vvi4 kov* b^Q*e ^ -{fou ba/>*S> OJuUS) ^<H1l*ju` **4^3*SxM. 4^tf\l4fi,J ^ &\ OJ UktoJf (jLuCAUO^) to |kLJ^kd^U4 i l\ t (\) Va^) \mjJUm4Uati (Xu) tO^b AC4 tc^ Uou^o cut/ CUiX^^ ""fee--" L+i/si af Ei^ ot--awuifccLdu^--\ <ig lu*e ^ "T^t au>aIxiL *fc* mtuAUd tUauy 4^*-^ `Atts cLca&Xuacm e4 CjCxca,ua* oL4 aJEu- k|JJaauuva4 Autu^lo-' QJtou^a It ^Lcra^t^Jku_4j clu^^&JuAi ^ C6Lf */ Ch Uau* euJlatufe uxft-a^ UmaEa*j *. "tUt ku* \**J*)y cJjjis> b&lee 3cft te4* f^djaJ cMJ *fl^ V$M> fc| dLaut>. Suum fc*e (btaC^ * Uu*> c^bvu^ '&v<'A 3f?4l ^ 2j& -2? ^ W CVH> y^XivitJ[-A. A^Vo-lf??) "W tq^^v JLL4/ VoJluiftd t&> **&*Qu ^ V*xt U^wSSU*\ cloudy ^ ^-4>f t >* vJtM^ Gj*t tjUo oJLh w Lg-^'isv \*{<3jddh u^Uil toe* a>* tg c 4w . 3iOx^ UCC 004887 A ^ 'S f 'J l } -r A The Influence of Occupational and Environmental Asbestos Exposure on the Incidence of Malignant Mesothelioma in Connecticut by H. C. Levinsohn J. V. Meigs M. J. Teta UCC 004888 The Influence of Occupational and Environmental Asbestos Exposure on the Incidence of Malignant Mesothelioma in Connecticut. by H. C. Lewinsohn, J. W. Meigs, and M. J. Teta Introduction Bruckman et al. (1977a) reported that "the combined sex ageadjusted mesothelioma incidence rate (AAR) per 100,000 Connecticut population has exhibited a possible ten-fold increase since 1935" They acknowledge that the available statistics might be subject to diagnostic error but, nevertheless, postulate that the rapid increase in the State's mesothelioma incidence rate is closely related to the increase in the State's "cumulative asbestos consumption", which includes asbestos emissions from industrial sources, motor vehicles, and building demolition. One criticism of their study is the failure to make available complete occupational exposure histories for the 133 diagnosed (19351972) cases of mesothelioma drawn from the Connecticut Tumor Registry (CTR). The Air Compliance Unit of the Connecticut Department of Environmental Protection has monitored asbestos concentrations in Connecticut (Bruckman, 1977h; Bruckman, 1978) and proposed an ambient air standard for Connecticut of 30ng/m^, measured over a thirty-day interval. The rapid increase in the State's mesothelioma incidence UCC 004889 A 00 2. reported in 1977 is referred to as justification for this proposal. The present study has been designed to re-examine the identified cases of malignant mesothelioma recorded in the CTK and attempt to determine the role of various etiological factors, such as occupational and environmental asbestos exposures. In view of the acknowledged possible diagnostic errors, which our results verify, a thorough review of available histological material will be undertaken. Methodology The CTR has identified 229 cases of malignant mesothelioma of the pleura (147) and peritoneum (33), other mesotheliomas, including breast, ovary, synovial membrane, testis, spermatic cord, perivesical tissue, cervical glands, brain, esophagus, bone and sternum (49), as well as 38 other pleural tumors, not mesothelioma, which were diagnosed in the state between 1935 and 1977 (Table 1). Disease topology and morphology are coded according to the ICD-0 (World Health Organization, 1976) classification. Medical, demographic, and occupational data have been collected for the cases and for their respective spouses. Similar information has been gathered for a random sample of approximately 700 decedents (1935-76) aged 20 to 98 years from the Division of Health Statistics of the Connecticut Department of Health (Table 2). Descriptive epidemiology for this research pertains primarily to the forty-three year time interval. Statistical modeling and future UCC 004890 A> 3. case-control comparisons include cases (215) diagnosed after 195h (Table 1), and controls (605) whose deaths occurred during this same time period. This procedure yields a case-control ratio of approximately 1:3, while reducing sources of error resulting from the limited occupational and medical data prior to 1955 and. the increasing awareness of mesothelioma associated with the later time period (1955-1977). The Price and Lee City Directories, utilized as an occupational data source for the three populations, were available for approximately sixty-six percent of Connecticut towns from 1890 to date ( with the exception of smaller towns). Directories were searched for job title and name of specific employer or industry for cases, controls and spouses at 1, 10, 20, 25, 30, 40 and 50 years prior to date of diagnosis, death or until the subject was less than twenty years old. An occupational history search was attempted for spouses to coincide with these intervals for their corresponding cases. The three-digit 1970 U. S. Census industrial and occupational codes (U. S. Dept, of Commerce, 1971) were assigned to the employment information ascertained from medical histories, death certificates, and City Directories. Industrial coding was determined on the basis of product descriptions for each specific employer which uere found in the manufacturing directories of Connecticut, published by the State Department of Labor (1957, 19&6, 1973) and from the Price and Lee Directories for non-manufacturing industries. From the complete listing of job and industrial codes a subset was selected by H. C. Lewinsohn, utilizing published sources indicating occupations A0045b UCC 004891 4. and products associated with asbestos (Hutchison, 1976; Levine, 1978; Weston, 1976). This will form the basis for probable asbestos expsoure classifications. The City Directory data over five decades provides information on duration of exposure. With a categorical response variable of occupational asbestos exposure, two fundamental research interests can be tested: 1. Whether cases and controls differ significantly with respect to job-related asbestos exposure (Relative Risk) and 2. The nature of the relative contributions of occupation, spouse occupation and geo graphical residence to the incidence of mesothelioma in Connecticut (Logistic Regression). In addition, the health experience of spouses of occupationally exposed cases and spouses of non-exposed cases can be compared. Results Of all cases diagnosed as pleural mesothelioma, twenty-one percent (31) had unknown staging (SEER, 1976) and sixty percent did not have an autopsy. The forty-nine cases of mesothelioma of other or unknown site have a mean age of fifty-two years and sixty-seven percent are without autopsies. The male-female ratio is approximately 1.5:1 (Table 3)* The question of the accuracy of classification of cases of malignant pleural tumors into mesothelioma and other pleural tumors naturally arises. The mean ages and the male-female ratios are similar for these two groups and the "other pleural tumor" UCC 004892 A-Jur h 0 5. group exhibits weak diagnostic confirmation, i.e., seventy-one percent positive histology and only thirty-five percent autopsied. A discriminant function analysis of these two groups of cases (diagnosed after 195*0 on the basis of age, sex, survival, time, stage, and number of tumors yields a sixty percent misclassification. This result suggests that: 1. Either these variables are not sufficiently powerful discriminators of pleural mesothelioma from other pleural tumors, or 2. The two groups, as a result of mi3-assignment of a substantial number of cases to the incorrect category, now lack the heterogeneity one would expect from distinct categories. If consideration is limited to cases identified after 1954, when recorded clinical detail in general improved, a3 well as an increasing awareness of this form of cancer, the number of cases with solid diagnostic confirmation and characteristics consistent with the natural history of mesothelioma are strikingly few. Of the 131 cases classified as pleural mesothelioma, 124 (95$) had histological material, (specimen from biopsy, frozen section, surgery, autopsy or D and C) examined and autopsies were performed on forty-five cases (38$). The proportion autopsied is disappointingly small. Table 4. shows the percentages of subjects with these two diagnostic critieria for mesothelioma cases of other or unknown site, for peritoneal mesothelioma, and cases of pleural tumors, not mesothelioma. UCC 004893 A00457 6. As a result of these findings, it is imperative that a slide review and a re-examination of hospital clinical reports for all cases of mesothelioma and other pleural tumors be undertaken. A commitment has been received from an independent pathologist for this aspect of the study. The following data analyses are subject to change subsequent to this procedure. Descriptive Epidemiology Between 1935 and 1977, the CTR identified 229 cases of mesothelioma, yielding an overall crude incidence rate of per million. Age-adjusted Incidence rates per 100,000 population (using 1950 U. S. population as a standard) for pleural and peritoneal mesothelioma consistently increase and exhibit a rapid rise beginning about 1955 - 04 {1935 - 44); .07 (1945 - 54); .18 (1955 - 64) (Figure 1.). We are in the process of generating these rates through 1977- Of these cases, 195 (85^) were reported after 1954. The male-female ratio is approximately 2:1; the mean age at diagnosis is fifty-nine years. Survival time from date of diagnosis to date of death is approximately ten months. A logistic regression model of the form: Log < -E- ) = A + \ Xi + b2 \ + B3 X3. was fit in order to examine the incidence of pleural mesothelioma (binary response variable) as a function of the categorical explana tory variables - time, age and sex (X^, X^, X^). Connecticut population figures supplied by the Connecticut State Department of Health were A used for denominator data in the estimate of disease probability (P). A00458 UCC 004894 7. The three and two-way interactions of time, age and sex were found , to be unimportant and the main effects model provided a reasonably good fit (p> 0.1). AAA From the estimates of the model parameters (B^, B^, By, changes in relative risk (here approximated by the odds ratio) over time intervals, age categories, and sex groups are approximated. Persons exposed between 1975 - 77 carry 3-5 times the risk of pleural mesothelioma as those exposed between 1955 - 64. With progression across the age intervals: 45-54, 55-64, 65-74,...the individual risk of this disease is 1.5 times greater than in the preceding ten year age group. As one might expect, males have three times the risk of females. In assessing these results, the poor diagnostic confirmation outlined in Table 3 should be borne in mind. The geographical distribution of cases of mesothelioma throughout Connecticut shows evidence of urban clusters in the five largest cities where 30$ of the cases resided at the time of diagnosis. Since these locations have comprised from 37$ to 20% of Connecticut's population (1940 - 1977), the suggested urban effect reduces to merely being a factor of population density. These five large cities are also centers of Labor Market Areas (LMA) of which there are eighteen in the State. Application of the CochranMantel-Haenszel Test of Average Partial Association (Landis et al., 1978) between these five LMA's and time of diagnosis (1935-77), while controlling for population age differences, suggests the following: UCC 004895 A00453 8. 1. Incidence trends (i.e., the manner in which an area's cases distribute over given time intervals) are primarily the same for four out of five LMA's tested. One region suggests a greater proportion of Its cases appearing in more recent years than the others (p=.04). This is readily explicable by the substantial in-migration experienced by this one area and its increasing industrialization as opposed to the other highly industrialized areas whose populations have gradually decreased over time and are, therefore, more comparable for analysis. 2. With the exclusion of this one LMA, the others indicate a similar distribution of their cases of mesothelioma over time (p=.l?) This possibly implies similar patterns of asbestos exposure for the four large industrial regions of Connecticut. Until such time as the mesothelioma incidence has been adjusted for the previously mentioned diagnostic weaknesses and possible misclassifications and until the occupationally exposed cases have been identified, it is not possible to determine the role of environmental asbestos exposure, if any, in the remaining cases. Summary Medical, occupational, and demographic data were collected for 267 cases of malignant mesothelioma and other pleural tumors, their spouses, and 605 controls. Methodology was developed for classification of subjects into probable asbestos exposure categories on the basis of product and job descriptions. Although disease incidence rates , . A j 14 d 1/ exhibit a rapid increase from 1955 4 1977, there remains a serious question of diagnostic reliability. A case review will be undertaken. UCC 004896 9- Acknowledgements The authors wish to acknowledge the technical assistance of Linda Mowad of the Connecticut Cancer Epidemiology Unit and Kathleen Pinto of Raybestos-Manhattan, Inc. UCC 004897 H L-- `t- i i 10. References Bruckman, L., Rubino, R. A., and Christine, B. (1977a-) Asbestos and Mesothelioma in Connecticut. APCA Journal, 27 > 121-126 Bruckman, L. (19776) A Study of Airborne Asbestos Fibers in Connecticut. Paper presented at the Workshop on Asbestos: Definitions and Measurement Methods. National Bureau of Standards, Gaithersburg, Maryland Bruckman, L. and Rubino, R. A. (1978) Monitored Asbestos Concentrations in Connecticut. APCA Journal. 28, 1221-1226 Connecticut Labor Department: Directory of Connecticut Manufacturing and Mechanical Establishments (1957) Wethersfield, Connecticut Connecticut Labor Department: Directory of Connecticut Manufacturing and Mechanical Establishments (1966) Wethersfield, Connecticut Connecticut Labor Department: Directory of Connecticut Manufacturing and Mechanical Establishments (1973) Wethersfield, Connecticut Cox, D. R. (1969) The Analysis of Binary Data; London; Methuen and Company, Ltd., pp. 14-29 Hutchison, M. K. (1976) A Guide to the Work-Relatedness of Disease. Washington, D. C., U. S. Department of Health, Education and Welfare (U. S. Government Printing Office) UCC 004898 A00462 11. Landis, R. J., Cooper, M. M., Kennedy, T., Koch, G. G. (1978) A Computer Program for Testing Average Partial Association in ThreeWay Contingency Tables (Parcat) Biostatistlcal Technical Report #18 Levine, R. J. (ed.) (1978) Asbestos: An Information Source. Washington, D. C., U. S. Department of Health, Education and Welfare (KTEW Publication Number (NIE) 79-1681) Price and Lee City Directory (1890-1977) The Price and Lee Company, New Haven, Connecticut D. S. Department of Commerce, Bureau of the Census: 1970 Census of the Population. Alphabetical Index of Industries and Occupations (1971), Washington, D. C,, TJ. S. Government Printing Office D. S. Department of Health, Education and Welfare, National Cancer Institute Cancer Surveillance Epidemiology and End Results Reporting. SEER Program (1976), Washington, D. C., U. S. Government Printing Office Weston, R. F. (1976) Technical Feasibility and Economic Impact of OSHA Proposed Revision to the Asbestos Standard. Prepared for the Asbestos Information Association/North America. Washington, D. C., Figure 2-1, p. 2-3. World Health Organization: ICD-0 International Classification of Diseases for Oncology. (1976) Geneva, Switzerland UCC 004899 A004s3 12. Hilton C. Lewinsohn,Corporate Medical Director, Raybestos-Manhattan, Inc. and Lecturer, Yale School of Epidemiology and Public Health (address: Raybestos-Manhattan, Inc., 100 Oakview Drive, Trumbull, Connecticut 06611, D. S. A.) J. Wister Meigs, Director, Connecticut Cancer Epidemiology Unit and Clinical Professor of Epidemiology, Yale School of Epidemiology and Public Health (address: 30 College Street, New Haven, Connecticut 06520, U. S. A.) Mary Jane Teta, Associate in Research, Connecticut Cancer Epidemiology Unit, Yale School of Epidemiology and Public Health (address: 30 College Street, New Haven, Connecticut 06520, U. S. A.) UCC 004900 A00464 Figure 1 Age-adjusted Incidence Rates by Decade of Diagnosis, Pleural and Peritoneal Mesothelioma, Connecticut, 1935 - 197^ UCC 004901 A 00 4 6 b da8 o o ooo b-p~ o 00 to ON roo r-focr DECADE OF DIAGNOSIS UCC 004902 A00 Table 1. CTR Reported Cases of Mesothelioma and Other Pleural Tumors Diagnosed Between 1935 and 1977 Table 2. Sources of Data and Variables Table 3- Sex, Age and Diagnostic Characteristics of Malignant Mesothelioma Cases and Other Pleural Tumors Diagnosed in Connecticut, 1935 - 1977 Table k Diagnostic Evidence for Disease Groups UCC 004903 Aar 4 Table 1. Diagnosis Period 1935-77 Primary Pleural Tumors (T6l) 185 Mesothelioma (M9O5) Other 1^7 38 Peritoneal Mesothelioma (T58, M905) Mesothelioma Elsewhere 33 49 Site not pleura or peritoneum Site unknown (T99) 37 12 Period 1955-77 151 131 20 29 35 27 8 267 267 215 215 UCC 004904 A00468 V a ria b le s Dem ographic, d ia g n o s tic , c lin ic a l, fo llo w -u p , employment m, 43 P h O T3O O o 22 O a-P P ft o 43 & Po a Q XJ V 43 P a O8 03 m 89 o O P* CO o aP p 8 a ft s b0 o O j--i S ex CD a Q O P P O O* <Q ft O <H 09 8 0 *H P a. Pft o CO CQ ft p os 8o T5 P o ex a 0 to 03 ft 0o ft <H TJ CQ 8 o H 1 8 P cx Pft CO u CQ o03 ft o8 P a Pi P 03 o b0 * rH cd IS fot 5 ft P Pi P PO Pp a P c (D * a0 a a 33 rH o P P rH (X st PaQ s 8 CQ 0 P O O h p i p M * 03 P O o p o p CD CO xjo o <r P opulation Case3 Source C t. Tumor R e g is try Tape & Paper F ile s 03 CQ 03 rH rH rH fOt oft fot Pp p 88 8 OOO Oa o CO CO 03 03 * s to 00 03 CO 2 8 8 8 8 a GO o O 03 o CX a) Pi cd Pi CO O CO O CO o CO * # p pa 'a 8 p P ft H Pa *> a<m aO 3 cad E-* 8ft 3 8 -- O 8 PoO PH P PCO rH P >P 43 PP P 4O3 Pa 0< s o P8 pft CO co*\- p ir\ cOnn rH P O P<H PPoft 43 P a p 8 o P p o g p o o Cm O Pft O Po ft p m r- c On P 8 Pf8t Po NNPOOOen* <3 3 rLT*NOPn Pft O P O ft P o p p o .3 nd 89 o ppf-t UCC 004905 CO rH O ft P c o o CO 03 Si cd o CO T3 8 P Pft P 28 T? 8 P 8 T? O O8 CO* 8>* o p 8oH P rOON- p ooo&8 Cm S' CD P 3 o Jt Vi. Vi. lT\ Ov m CO cO cO " H P w. V*. v. coP n rn rH H *H Ov t- Ov m sc cou CO CO ba>l < C vO vCOM CO crv CM LTV aO) X co o H P a3 rH J-} LTV B Ji-lt -* mo rH CM ON LTV CoM *11--i CoM -OifJ o CM ON OJ Lf\ c-- soi CoOn rmo O^ vO OJ *8 03 P a) a cj o a a aB o o oO a rH O 5 q09 H *H rH df *h tH V rH O rH rH *H ,2 fl oCO rH 0 rH 03 q O a) 43 0$ 43 O JS 43 p 3 o 3 o S Mp MP -C -P H O PP O cd 0) 01 01 H rH 0 rH s> m 09 43 p P h Q a. X CU X CL, S s o rH P O E-h UCC 004906 o v4~ o o -=I ON P r-H0H \&H TOC3D O vO r- ON stwwo Table 4. Diagnosis Peritoneal Mesothelioma No. 29 Other mesothelioma (not pleural or peritoneal) 35 Pleural tumors, not mesothelioma 20 "Positive Histology11^' Autousy 96% 88% 88% ko% 85% 38% 1. SEER, 1976, Code/Field Number 19 UCC 004907 A0047 1 lustration for Typing Abstract Since the abstracts are to be reproduced directly from the author's own typescripts in photo-offset, reed instructions carefully before typing abstract 1. Do not mak6 carbon copies when typing on this form. 2. Use a typewriter, preferably electric, with ELITE type, black ribbon (preferably carbon), insertions should be written in BLACK Ink. 3. Do not erase. Your abstract will appear in print EXACTLY es submitted. 4. ENGLISH should be used for submitting of any abstract. 5. Please make the abstract as informative as possible including methods, results and conclusions reached. It is not adequate to state: *The results will be discussed*. 6. Your entire abstract must fit within the box outlined. 7. Use abbreviations as lar as possible to describe the names of laboratory or reference journals, for example. Biochem. Div.. Natl. Inst, for CaneSr Res.. Biochem. Dept., Unrv. ot Oxford J. Biol. Chem. 235,10 (1960). B. Your abstract should be sent to the Secretary and should be received before Aprii30 1979 NOTE ESPECIALLY: Entire abstract In 1 paragraph typed single-spaced and no margin at top or left side. * Title in CAPITAL LETTERS ( Author names underfined Laboratory where research was done, city, state and country tudies on $ea-smkitte nomst'cmreins a AND b FROM/LATICAUDA SEMIFASCIATA VENOM. N. Tamiya* k H.Arai**. "Chen. Dept., Tohoku Univ.,Japan and Med.School, Tokyo Red. and Dent. Univ., Tokyo,Japan. The toxic principles in the venom of the sea-snake Laticauda seraifasciata were separated into two components by CM-cel 1 ul ost- chromatography and obtained in crystalline forms by salting out with (NH^gSO,. They were named 'erabutoxins a and b'. The homogeneity4of each toxin was shown by rechromatcgraphy, by disc electrophoresis.by ultracentri fuging, by toxicity measurements before and after repeated crystallizations and by N-terminal amino acid analysis. They had molecular weights of about J000 as measured by ! (Abstract may be extended to fill the rectangle.) A j ... 4 4 0 UCC 004885