Document 6w897rdb9zwkpdrnK26xE01z3

DEPARTMENT O iH E .TH & HUMAN SERVICES Public Health Service Centers for Disease Control National Institute for Occupational Safety & Health Robert A. Taft Laboratories 4676 Columbia Parkway Cincinnati OH 45226-1996 February 14, 1990 HETA 69-021 The Honorable Sherwood Guernsey Massachusetts House of Representatives 152 North S tre et - Room 330 P it t s f ie ld , Massachusetts 01201 Dear Representative Guernsey; The epidemiologic study of deaths among workers a t the General E le c t r ic Company, P it t s f ie ld , Massachusetts, conducted by Or. David Wegtnan and Dr. Sander Greenland has been concluded. The re su lts were presented to the public at a press conference In P it t s f ie ld , Massachusetts on January 24, 1990, As 1 understand i t , copies of the executive summary and appendices have been made availab le at several local lib r a r ie s . We received a copy of both documents the morning of January 24, 1990, and w ill retain these 1n our f ile s (refe r to HETA 89-021). Several appropriate recommendations were made to General E le c t r ic , The establishment of a panel of physicians to . . . "provide assessment and consultation to any employee who has further concern about health problem s..." is p a rtic u la rly useful. This should help workers who are le f t wondering..."So, what does th is mean for me, as an individual?" Not mentioned, but worth considering, are the selection c r it e r ia for physicians to the re fe rra l panel. I believe that the local union should be a c tiv e ly involved in th is process. L it t le w ill be gained i f the workers have no tru st in the physician panel. The recommendation to further examine the association between exposure to "re sin s11 and lung cancer should also be pursued. The f i r s t task would be to explore the a v a ila b ility of useful data. Several lim itatio ns to the study were emphasized by the in vestig ato rs. The fa c t that many personnel f ile s were unavailable lim ited the study design options. Even so, the researchers have done an extraordinary job with the available materials. No doubt, many questions remain unanswered and new questions have evolved. While these questions have merit and deserve attention, further epidemiologic evaluation, using th is data set, w ill probably not help. Further work should focus on solvable issu es. Current exposures to chemical hazards, Increased worker awareness, and concern regarding health ris k s may be addressed by in stitu tin g engineering controls to reduce exposures, worker education, and developing physician panels lik e the one recommended by Dr. Wegman. The costs, benefits, quality of available data, and time required should be considered before further epidemiologic studies are begun. Removing hazardous exposures prevents future disease and need not await the outcome of a prolonged study.  fe , E\ .J Page 2 - The Honorable Sherwood Guernsey In your recent le tte r dated February 7, 1980, you referred to several Inadequacies 1n Dr, Hegman's report. Local workers apparently believe that not a ll of those who died were included In the study or had been assigned to the correct job task. To address this you requested a set of tables "categorizing worker deaths by exposure le v e l, by job code, and by Individual b u ild ing s.B You were p a rticu la rly concerned with buildings 12 (12, 12X, 12Y, 12P, 12H in clu sive) and 100. I believe i t reasonable for General E le c t r ic , or it s contractor, to furnish a lis t in g of a ll workers in the study who worked in these buildings. The l i s t should include name, job codes, exposure le v e ls , and dates of employment for each job code. Since a ll study subjects had died, the release of names, death c e rtific a te information, and job h isto rie s should not infring e on privacy considerations. However, information from medical records might be considered private and probably should not be released. The General E le c tric Company should also provide a lis t in g of a ll deaths that were not Included in the study, giving name and cause(s) of death (or copies of death c e r tific a t e s ). The next three Items (#2, 3, 4.) focused on the deceased workers excluded from the study because of missing personnel f i l e s . The question is whether these workers differed from the study group in respect to exposure and cause of death and, i f so, whether th is affected the study re su lts, Drs. Greenland and Salvan examined th is 1n th e ir Appendix IV (pages IV 42*46), Listed are the numbers of cases and controls and cause of death by exclusion/inclusion status. The tables are discussed on pages IV-2 and IV-3, I t does not appear that the exclusion of these deaths seriously biased the study re s u lts . The "proportional an alysis based on national and state [proportions]", which Dr. Hegman agreed to conduct, would not add to the more d ire ct approach taken by Drs. Greenland and Salvan, The proportionate m ortality of the exclusions and the study group appear to be sim ila r. Comparing the proportionate m ortality of the exclusions to state and national m ortality would, In e ffe c t, be a proportional m ortality ra tio study. Such studies are generally considered less accurate than the case-control approach that was carried out. You also requested a comparison between Dr. Hegman's executive summary and an Interim report recently released concerning the m ortality experience of workers at Westinghouse E le c tr ic in Bloomington, Indiana. The two studies differed in design and cannot e a sily be compared. The General E le c tr ic f a c ilit y included multiple buildings and manufactured transformers and p la stic resin systems. The Hestinghouse f a c ilit y was housed under a single roof with few w alls separating the various process ares and manufactured only e le c tric capacitors. Types of exposures and the amount of exposure probably differed at the two f a c ilit ie s . However, polychlorinated biphenyls (PCBs) were used at both. i t &, n Page 3 - `The Honorable Sherwood Guernsey In the Westinghouse study, we Identified a ll workers, employed for one or more days between January 1, 1957 and A p ril 30, 1977. The workers were followed until June 30, 1986. The number of observed deaths from sp ecific causes was compared to the number of deaths expected from age-, sex-, race-, and calendar year-sp ecific United States rates. The en tire cohort of workers at Westinghouse had been exposed to PCBs; differences in PCB exposure level have not yet been factored Into the a n alysis, although we intend to do th is . We reported an increased m ortality ris k from malignant melanoma (SMR * 4,06, 9 B % Cl 1.75 - 8,01) and cancer of the brain and central nervous system [SMR 1,60, 95i Cl 0,58 - 4.22) for workers at the Westinghouse E le c tric Plant, The General E le c tr ic study found no association between exposure to PCBs (pyranol) and either cancer of the brain or malignant melanoma. Malignant melanoma was not studied by Drs, Wegman and Greenland since they did not observe more than ten deaths attributed to th is cause. (We found only 8 deaths with an underlying cause of nallgnant melanoma among 5600 exposed workers, representing a four-fold excess.) Two studies of capacitor workers, sim ilar to the Westinghouse study, did not confirm the increased m ortality from malignant melanoma or cancer of the brain and central nervous system that we found a t the Westinghouse f a c ilit y . An appendix appears to be missing from the report. The executive summary refers to an "Operations Descriptions Book1' included as an appendix (page 11-12), This appendix should be added to the reports already distributed. I have enjoyed working with you during the past year. Please convey my appreciation to Mr. Steve Hamilton of General E le c t r ic , Dr. Dayid Wegman, Dr. Sander Greenland, and the peer review panel for in vitin g me to attend th e ir meetings. I f I can be of further assistance, please feel free to contact me at (513) 841-4386, Thomas Sinks, Ph.P, Supervisory Epidemiologist Medical Section Hazard Evaluations and Technical Assistance Branch D ivision of S urveillan ce, Hazard Evaluations, and Fie ld Studies