Document b5zDnoEq9yRyd85QE4Ko9zX61

CONFIDENTIAL May 2, 1974 To: Ur, P. N. Shellenberger Mr. W. L. Sullivan Subject: Additional Epidemiological Studies on Workers Exposed to Vinyl Chloride A number of industry lenders, includingDiamondShamrock, Union Carbide, and PPG, have proposed.that^addi tional studies on the mortality of workers chloride bo immediately undertaken to esthbiisI^^h^^eiSla;:^ showing of "no increased risk* exposed at levels in the 10 been under review within Organization ResourcosftCo^^ one of our affiliated organizations, and _r ,, tendorsement from the r..u...b...be_r .i.n...d. u..s._t_r_y_. ^ I_t_^_is|j^pppip| number of sizable plants engaged'in the'extrus!i^i%]^ciSc^^|^fe^ fabrication of PVC, but not in monomer^or.ipblymejg^^p^^^" be enrolled in this study and that an attempt ^belmaSWiita ,, -elude all such plants which have been in operation more.i'thao^itawelve years and have more than 100 - 200 people ''" -- '',v":rt' posed protocol is attached. An unidentified company, has .commit exploratory phase of the study. We will',sbortTf Invitation to become one of the sponsors; a nd^tplind^is^he^iwP whether or not we have facilities which: If a sufficient number of sponsors canobe-;e'hlisjtlM^[iitW^&Wnnd< +thfeaat 4t- +thli/%e iwi/norrlkr uw* 4 i1l1l Yb\ro% asftnawrtfeodrt >bityt ORC*' oaWn#4dv'4*tKhaeWn'v;' overall umbrolln of SPI. It is likely that All ledi^- as>lf^i;nAUaanucv:lAaii r1fo r contribution would be $20,000 or loss,1 to be equally.diyfded between Fibers and ICD. We are providing this advance noticevSd|tlw^|;yoU;;Cnh be developing your thoughts on Allied's possible ;partldipa.tiori. Should you feel that we should not participate^ wejs - fully consider the effect of this decision on ourfrelations with the vther major companies in the industry. ' WSF:mn c: Ur. E. W. Callahan Mr. J. M. Quinn W. S. F rguson ASJ 00023603 Mr. T. D. K nt Mr. R. H. Sand f\pn i ca, y/H Mortality Among Workers Involved in Processing PVC The objective of this study is to review death certificates for the period 1964 to 1973 among workers who may have been exposed to PVC and died while working for or retired from the companies involved. The population under study will be limited to those for whom death certificates are available from life (insurance) underwriters of the companies in question. Two specific questions are to be investigated: (a) is there any evidence . . . of death due to angiosarcoma and was that death possiblyrelated:;tOtPVC?v:^}';'i:- 9 ^ w* #4 I U 1 m A vh ama i mi I % I a ^ r rt A a fflA V 4 I vll^^ntAilAO.VlIlk .1 necessary . within a four-month period and since it would not be possible the entire study in that time, the investigation will be:carr a series of sequential steps. * v> rv-^f. -.a *. There are an estimated 123,000 workers involvetf )1nfthe : of PVC after it leaves the manufacturing pi ant. ; We assume- m to Identify those plants or divisions within,companies)where# that PVC exposure could have occurred. It is further "assume<il ifS tfttP ." / il&J&hnSjW life underwriters will be able to provide us with death)cfertMcafiSf^lff#M workers from the plants or divisions where the PVC exposure^ very extensive, and it is very unlikely that anyj,sub]st^ can be completed in four months. By limiting the; to those certificates available from life underwifitere'.ytfiel^ , ,,cU i .. .. .. ; ' "leakage. Such a review would not Include any employee who5left the company ~z~ and died without coverage provided by the life underwri ters. Additionally, we would like to restrict the death certificate review to deaths occurring among all employees of the plants where PVC is used in fabrication, assuming the underwriters can provide death certificates on that basis. This could result in additional leakage if there is a tendency for employees to transfer within a company from a job in a PVC-using plant to a job in a non-PVC plant. It is hoped that some information may be available from the companies which would show that the latter leakage is minimal. It is estimated that there would be about 11,000 deaths for a group of 123,000 workers over a 10-year period. The strategy of this study will be to review the approximately 11,000 certificates and from these, select those with either cancer or liver disease mentioned on the certificate. This would amount to about 2,600. For these 2,600 deaths, it will be. necessary to go to the hospital where the death occurred and review the pathology report for any mention of angiosarcoma. (The number of deaths and the number selected for further review are, of necessity, gross estimates but should give us some starting point for estimating workload. These numbers assume we can limit our review to plants involving PVC workers. If we must go to divisions or entire companies, the job will be significantly greater.) An additional element of the investigational strategy will be to contact the pathologist in any hospital visited and inquire whether there are any_ angiosarcoma cases in the pathology records and, if so, to determine whether or not it may be related to PVC exposure. Such a step in the investigation should cut considerably any "leakage" of the second type mentioned above (i.e., transfer of PVC plant to non-PVC plant within the same area. The hospital would provide coverage for a fairly wide AS I 00023605 geographic area and could easily encompass workers of other company plants in a nearby geographic area.) Invqstigational Procedure (Full Study) 1. Identify all plants among involved companies where PVC exposure may have occurred. These are the plants where the 123,000 workers are employed. 2. Identify life underwriters for the companies by whom these plants are owned. 3. - Request all death certificates for deaths occurring from 1964-1973 among a. active employees who died while employed at those plants, and b. retired employees who worked at those plants prior to retirement. These certificates are to be : a. Identified and separated from all other deaths in the covered companies, b. copied (Xeroxed), and c. sent to ORC. 4. 'Certificates will be reviewed by a nosologist retained for that purpose. 5. Death certificates on which cancer or liver disease is mentioned will be separated. Tor 1964-1967; cancer includes any of the causes of death identified by rubrics 140-205 of the 7th revision of the International List. Liver disease will include rubrics 580-586. Tor 19C3-1S75. acmeer is identified by rubrics 140-209 of th< 8th revision of the JCDA and liver disease will include rubrics 570-576.] 6. From the information on the death certificate, hospital of death will bi crprtained. AS I 00023606 .7 Hospitals will be visited to review pathology records for employee deaths identified in Step 5. It is assumed that the involved companies v/ill be able to provide quick access to all hospital records which must be reviewed. 8. While at the hospital, the record reviewer will contact the hospital pathologist to determine whether or not the pathologist is aware of any cases of angiosarcoma. 9. Should any cases of angiosarcoma be turned up in Steps 5-8, complete occupational histories will be obtained for those workers. 10. Analyses in the form of PMR's (Proportionate Mortality Ratios) will be carried out on those deaths identified in Step 5 while the hospital review is going on. 4 4 4'. ******** ROTE: We ar`e requesting that death certificates be sent to ,ORC in order to take advantage of specialized talent available in Washington for death certificate review. It is unlikely that on short notice we can obtain an experienced nosologist to travel to the insurance company offices or* in fact* to train someone. Further* it is imperative that ve have copies of death certificates mentioning cancer or liver disease on site at OF.C. Since we do not want the underwriters doing that screeening* it is necessary that we obtain copies of all certificates as outlined in Step 3. Investigational Prot.edure (Phase I) 1. Same as 1 above. 2. Same as 2 above. ASI 00023607 .3 Select the ?-5-? largest underwriters who are likely to cover approximately 30-35% of the workers. From these, request death certificates as outlined In Step 3 above. 4. Same as 4 above. 5. Same as 5 above. 6. Same as 6 above. 7. Same as 7 above. 8. Same as 8 above. 9. Same as 9 above. 10. Same as 10 above. 44444444444 BOTE; At this point, there remain at least Z alternatives for proceeding on to the hospital review. Me can do this one insurance company at a time (which is likely, since the records are likely to corns from the companies one insurance company at a time. Or, if all certificates are available at one time and very quickly, we car. select a sample for initial review. This would necessitate visiting some hospitals more than once). Because of the pressing time faczcr, it will he necessary to select that alternative which shows premise of quickest results and we won't know that until the study is underway. There are, of course, a number of additional details for each of the above steps, e.g., abstract forms and instructions for the hospital review, and these details will follow. LC/ms Leonard Chiazze, 0r. Sc.D. ASr 00023608