Document DGMnmq0dkVxNZYeKJQjrvMzbO

/ Environmental Health Associates, Inc. 4 October 1954 Dr. Michael Cardin Union Oil Company of California 461 South goylston' Street Lor Angela*, CA Dear Dr. Cardin: A draft of our review of the Department of Health Services (DHS) document, "Part 5: Health Effects of benzene" was sent to you on 24 September 1964. We are in the process of preparing a final report of our review, which should be ready In several weeks. In the Interim, we would like to summarise the major points of our review in this cosnunleatlon. First, contrary to the DHS document, we do not find any epidemiologic evi dence of association between benzene and lung cancer. Second, in our opinion, the DHS document does not provide a convincing argument for the DHS choice of a particular, also the highest, risk estimate based on an animal study. Third, we have carefully reviewed the three epidemiologic studies used by EFA's Car cinogens Assessment Croup (CAG), and found the GAG had moat likely overestimated the risk associated with benzene exposure. Thus, contrary to the DHS assessment, risk estimates based on epidemiologic studies are lower than those from the ani mal studies reviewed by DHS. Finally, the statement In the DHS document that "occupational studies based on exposed males will underestimate the effect of - benzene when the results are extended to the general population because women and children weigh less Is speculative and not supported by current evidence. As mentioned above, the DHS document suggested lung cancer as another site that might be associated with an increased risk as a result of benzene exposure. The document stated, "only one epidemiological study, that of Environmental Health Associates [1963), examined all causes of cancer in male workers possibly exposed to benzene. Interestingly, this study demonstrated a positive trend for higher overall cancer Incidence and higher lung cancer Incidence In exposed / SAL 000002525 2150 Shattuck Ave. Suite 4U Berkeley. CA 94704 [415) 548-1888 Import - 10/04/64 Page 2 workers**." Thl* statement has a number of inaccuracies* Since we conducted the study, we feel that it would be appropriate for ue to clarify the iaeue of lung cancer and benzene exposure* In addition to the EHA 1963 report, there are other publications reporting all causes of death for individuals exposed to benzene* Examples include Ott et al (1976), Decoufle et al (1963), and Tsai et al (1963). Many studies based on rubber workers and refinery workers also reported all causes of death* Examples include Monson and Nakano (1976), McMichael et al [1976], Sehottenfeld et al (19811, Environmental Health Associates (1964), and Kanls et al (1962). Regarding the EHA (1963) benzene study, we want to point out that it did mot examine incidence; it was an historical eohort mortality study* More impor tantly, the EHA (1983) study did not demonstrate a positive trend for either all cancers combined or for lung cancer* The followlr. -ble shows the Standardized Mortality Ratios (SMRs) for all cancers, and lung cancer by cumulative benzene exposure* Cause Cumulative Exposure fppm*-months) <160 180-719 720+ All Cancers Lung Cancer 102.3 123.4 129.4 173.7* "Statistically significant at the 0.06 level 77*6 64.8 It is obvious from these SMRs that neither cause of death shows a positive trend. Xn fact, the group with the highest exposure (720+ ppm-months) experienced considerable deficits in all cancers and lung cancer* The group with 160-719 ppm-months of exposure hsd a statistically signifi cant SMR for lung cancer when compared to the U.S. males. However, the excess could be attributed to a plant located In a county with unusually high lung SAL 000002526 Report - 10/04/64 Page 3 * 4 cnctr rates* This information was transmitted to DH5 by Chemical Manufacturers Association (CMA) (see Attachment!, but was not Included In the OHS document* To further examine the issue of lung cancers and benzene exposure, we reviewed some other epidesdologic studies and the following is a tumury of our review: Author Ott et al [1976] Tsai at al (1983J Decoufle et al (1963) Hants at al [1962] Schottenfeld at al (1981) EHA (1984) Monson and Nakano (1976) McMiehaal at al (1974) Exposura Benzene Benzene Benzene Refinery Refinery Refinery Rubber Rubber Lung Cancer SMR (Observed/Expected) 111 (7/6.3) 32 (2/3*88) 31 (1/3*21) 91 (60/66.0) 68 (37/83.37) 67 (124/183.42) 92 (234/233.1) 84 (91/109*3) Based on the above discussion, we do not feel that there is any epidemio logic evidence supporting the argument for a positive association between expo sure to benzene and lung cancer* The OKS argument was based on a single statistical finding for a subgroup in the EHA [1963] study, which, as explained to OHS previously, was primarily due to the unusually high lung cancer rate in a county where one of the participating plants is located (see Attachment). The OHS suggestion that benzene might cause lung cancer has profound ramifications for the public health. Therefore, that suggestion must be examined In detail* In summary, wa feel that the risk estimate provided by the DHS document, which Is baaed on a vary selected risk estimate from an animal study, represents an overestimate, and is not consistent with findings from existing epidemiologic studies. SAL 0000QZ527 Import 10/04/84 Page * tf you h*vo any questions About this iunr/ or our draft report, please feel free to coll on ae. OW:Jdd ce: Robert Harrison Enclosure SAL 00000252Q