Document g24ddwjpyoXX6E5989eaxwz8Q

7MA) ises Original Articles Cancer Mortality of a Group of Canadian Workers Exposed to Vinyl Chloride Monomer G. Theriault, M.D., Dr.P.H., and P. Allard, M.D., M.Sc. (7 / . The present study was undertaken to find out whether there was an excess of cancer mortality from causes other than angiosarcoma of the liver among a group of workers heavily exposed to vinyl chloride monomer (VCM). The mortality of 451 workers exposed to VCM for more than five years was compared with that of 870 workers from the same company who had not been exposed to VCM. The relative risk for digestive cancer was significantly higher than 1 [6.25, confidence interval 2.69 to 14.52) in the ex posed group. The standardized mortality ratio (SMR) for digestive cancer was also higher (SMR 259.26 p < 0.01) than that of the general population. No other cancer was in excess. Since the exposed workers are known to have had a cigarette smoking experience similar to that of those who jvere not exposed, it is concluded that the association be tween lung cancer and VCM exposure, if present, is indeed rather small. Several studies have demonstrated that vinyl chloride monomer (VCM) produces angiosarcoma of the liver among people exposed at work,17 There is, however, much controversy about the capability of VCM to gener ate other types of cancer in man. One author,* using a proportional mortality ratio technique, described ex cesses of lung cancer (observed, 13; expected, 7.9; obs/exp 1 6) and brain cancer (observed, 5; expected, 1.2; obs/exp 4.2) in a group of 161 workers from two vinyl chloride plants.* Another author,* who studied the mortality of a group of 1,294 American workers exposed to VCM, re ported excesses of lung cancer (observed, 11 expected, 7.5; SMR 194, p < 0.05) and brain cancer (observed, 3; ex pected, 0.6; SMR 498, p < 0.05) among men who had had From the Department of Social and Preventive Medicine School of Medicine. l*val University, 5te*Foy. Quebec, Canada. OK 7P4 T hi* research was financed partly under Crant No 780396 from the Conseil U Recherche en Sante du Quebec L their first exposure more than 15 years before their death. While most authors have not been able to document a significant excess of any other cancer in their studies, they have noted that such an excess may exist. In order to better document the cancer mortality asso ciated with exposure to VCM, the present study of the workers of the Shawinigan vinyl chloride polymerization plant was undertaken. The main objectives of the study were (1) to compare mortality among vinyl chloride workers with the mortality of workers from a nearby industrial complex; and (2) to compare the mortality of the VCM workers with that of the general population. The Shawinigan vinyl chloride plant opened in 1943 as an independent company which produced both VCM and PVC (polyvinyl chloride). In 1958, it was merged into an industrial complex (to be used as the comparison group) of which it became the Cana dian Resins Division. In the late 1960s, the VCM produc tion ceased and only the polymerization process con tinued operation. In 1972, this division was sold and again became an independent plant. Over the years, the VCM and PVC productions fluctuated. The average annual work force at the plant was 225 men. In 1965, VCM pro duction was estimated at 90 million pounds and PVC at 60 million pounds."' In 1977, PVC production was 22 3 million pounds." Since several episodes of unconscious ness among workers were reported, the level of VCM in the air at the workplace is believed to have been high dur ing this period. Ten cases of angiosarcoma of the liver have been reported since then.7 The industrial complex from which the comparison group was drawn comprises several divisions: a carbide division which produces calcium carbide and acetylene black; a chemicals division which produces several chem icals derived from acetylene: acetone, chloral, butanol, acetic acid, acetaldehyde, several acetates, solvents and resins; a stainless steel division which consisted of a small foundry and a division formed in 1957, which produced Journal o( Occupational Medicine/Voi. 23, No. 10/0ctober 1981 671 ASI 00005520 f Fig 1. -- Population under itudy. sulfuric acid, chlorine and cyanide. This complex started operations in 1927; some of its divisions opened at a later date; others closed in the late 1960s or early 1970s. The annual number of male workers averaged 1,000.11 Methods Population Under Study. -- All production workers whose names appeared on the unions' lists or the com panies payrolls of the VC plant and the industrial complex between January 1,1948, and December 31,1972, consti tuted the population under study. Several methods (old addresses, telephone books, Quebec motor vehicles bureau, inquiries with fellow workers and families, searches in homes for retired persons) were used to trace these men, 18% of whom were still working for the same companies. Once identified, the worker himself or his next of kin (spouse, child or close relative) was questioned at home or at work by a trained interviewer. The ques tions permitted the reconstruction of a detailed occupa tional history within and outside the companies under study. Questions were also asked regarding smoking habits. For each deceased person believed to have been exposed to VCM, the occupational history was confirmed with the help of a fellow worker. The men whose number of years of work for thev companies, as established through occupational histonedid not exceed five were removed from the stud\ The nmaining population was then subdivided into thm groups. The first group (known as the unexposed cohort consisted of men who stated they had not been invoked with the production of VCM or PVC, or both, for a perioc greater than five months. The second group (constituting the exposed cohort) comprised men who declared the\ had worked on the production of VCM-PVC for more than five years. Finally, the third group consisted of the men who had worked on the production of VCM-PVC tor a period of six months to five years. This group was ev eluded from analysis because it neither fulfilled the definition of the exposed nor the unexposed cohort Data Sources. -- The vital status of these men established as of December 31, 1977. Causes of death were determined from death certificates. They were coded according to the 8th revision (ICDA) by one of the authors, who was unaware at the time of the coding as to which group the man belonged. For cancer cases, a histopathological confirmation was sought on autopsy, btops' or cytology reports contained in the medical files at the hospital. Information on Quebec's population and roof- ufey by cause, age Q*ida for the year of the results, only the death certifi Statistical Analysi t*"man-years at ri nonber of person-' lereech five-year at between 1948 and ' The mortality in with that in the i method of relative wfti those of Mar The observed mort. Ulity expected fro Quebec according mortality ratio (SM tfte one described Quality Control flonal history was posed within the V I detailed profile within its ma tudy, it was pre-t Interviewers wh Wning by the sat the entire study. 1 slewed weekly b> nation was soug computer were s\ nel questionnaire; tearfa As shown in Fig M U) \fi Eipoturt O o o (Tir) * 5- 9 o 10-14 15-19 20-24 -- <x 25-29 30+ Total 672 Mortality of Workers Exposed to Vinyl Chloride Monomer/Th&iault and Allard ^WTBl of Occupa Hij#' 3^ h less the ; for these al histories, jdy. The reinto three -ed cohort) n involved or a period onstituting lared they * for more ued of the M-PVC for up was ex.Ifilled the :ohort - men was ; of death They were one of the ading as to es, a histoosy, biopsy files at the i and mor- t and Allard Table 1. -- Aga Distribution ot tha Man In the VCMExposed and Unexposed Cohorts at of Dseembar 31,1977 (Deceased Excluded). Age 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70*74 75 + Total Expastd No. % 0-- 0-- 1 0.3 28 7.1 75 19.1 96 24.5 102 26.0 53 13.5 30 7.7 6 1.5 1 0.3 392 100.0 Unaxposod No. S 1 0.2 3 0.5 6 0.9 45 7.1 55 8.6 64 13.2 112 17.6 110 17.3 113 17.7 65 10.2 43 6.7 637 100.0 tality by cause, age and sex was secured from Statistics Canada for the year 1971. In order to ensure the validity of the results, only the causes of death which appeared on the death certificates were used in the comparisons. Statistical Analysis. -- The results were analyzed using the "man-years at risk" method as described by Hill.'* The number of person-years of observation was established for each five-year age group and for each five-year period between 1946 and 1977. The mortality in the exposed cohort was compared with that in the unexposed cohort according to the method of relative risk. The tests of significance used were those of Mantel, Haenszel'415 and Miettinen.'4 " The observed mortality was also compared with the mor tality expected from the population of the province of Quebec according to the method of the standardized mortality ratio (SMR).'* The test of significance used was the one described by Bailar and Ederer." Quality Control Procedures. ~ A thorough occupa tional history was developed for workers who were ex posed within the VC plant The questionnaires presented a detailed profile of the plant while specifying possible fobs within its major units. Before the beginning of the study, it was pre-tested with former employees. Interviewers who contacted the workers had received training by the same person and were supervised during the entire study. The completed questionnaires were re viewed weekly by one of the authors and missing infor mation was sought by telephone. Data entered in the computer were systematically checked against the origi nal questionnaires. Results As shown in Fig 1, the population under study originally comprised 1,659 workers. Of these, 48 men were untraceable, leaving the number in the study population at 1,611 (or 97.1 % of the original number). After the removal of the 156 men who had worked at the companies for a period not exceeding five years and the 134 men who had been exposed to VCM for a perod extending from six months to five years, the exposed cohort comprised 451 men and the unexposed one, 870 men. In the exposed cohort, the proportion of men who responded to the questionnaire for themselves was 84.3%. In the unex posed cohort, this proportion was somewhat lower at 68.2%. The age distributions of the workers in the exposed and unexposed cohorts as of December 31,1977, reveals that workers in the unexposed cohort (Table 1) were somewhat older. Since the industrial complex had been in operation for several years when the VC plant opened, the finding of older people in the latter cohort was predictable. Table 2 shows the distribution of the exposed workers according to their length of exposure and their observa tion period (time between first employment and end of follow-up). Among these, 81 % (365 out of 451) had an ob servation period of 15 years or more. For 23% of them, the time elapsed between first exposure and the end of the follow-up period exceeded 29 years. The mean obser vation period for the exposed cohort was 22.6 years. Table 3 gives the diagnosis at death and the diagnosis after histopathological review of all the cancer cases found in the exposed group. On the death certificates, eight cancers of the liver were reported. The histopatho logical review rejected one of them as being a cancer of the sigmoidal flexure of the colon, but added another as an incorrectly diagnosed cancer of the peritoneal cavity. All eight cancers of the liver turned out to be angiosar coma of the liver. It is noted that two more angiosar comas of the liver were certified as cirrhosis on the death certificates and were considered as such throughout this study. Table 4 compares the mortality of the exposed workers with the mortality of the unexposed workers. The con fidence interval (95%) of the relative risk indicates that there is a significant excess of mortality due to cancers of the digestive system in the exposed group (RR 6.25, con fidence interval 2.69 to 14.52). There is no excess in the other causes of death. Table 5 shows the standardized mortality ratios (SMRs) for the exposed workers in comparison with the entire population of the Province of Quebec. The SMR for all causes is 83.02, thereby showing no excess of death in the exposed cohort Although the SMRs for all cancers, respi ratory system diseases and digestive system diseases ex- Table 2. -- Distribution of Exposed Workers by Length of Exposure and Observation Period. Exposure (Years) 5- 9 10-14 15-19 20-24 25-29 30 + Total Oburvstlon Period (Yiars) 5-9 10-14 15-19 20-24 25-29 30 + 30 28 7 40 2 4 28 26 19 17 5 -- -- 17 35 19 8 ------ 22 24 15 -- -- -- -- 34 -------- -- 33 38 30 56 50 116 96 103 Journal of Occupational Medicine/Vol. 23, No, 10/October 1981 Total 111 95 79 61 67 38 451 673 ASI 00005522 Table 3. -- Age at Death, Diagnosis on Death Certificate and Diagnosis on Histological Review ot the 20 VCM-Exposed Workers Who Died of Cancer. Casa Numbsr 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Daath Certificate Cancer of the liver Cancer of the intestine Carcinoma ol the trachea Angiosarcoma of the peritoneal cavity Cancer ol the lung Cancer of the liver Cancer of the liver Cancer o< the prostate Angiosarcoma of the liver Cancer of the pancreas Adenosarcoma of the intestine Hepatoma Cancer of the liver Cancer of the pancreas Chondrosarcoma Hepatoma Primary cancer of the liver Cancer ot the digestive track Cancer of the biliary ducts Lymphoid leukemia Histological Review Cancer of the sigmoid flexure of ihe colon Metastatic cancer of the colon Carcinoma of the trachea Angiosarcoma of the liver (no review) Angiosarcoma of the liver Angiosarcoma ot the liver Cancer ol the prostate Angiosarcoma ot the liver Epithelioma of the head of the pancreas Adenosarcoma ot the intestine Angiosarcoma ot the liver Angiosarcoma of the liver Anaplastic epithelioma of the pancreas Chondrosarcoma Angiosarcoma of the liver Angiosarcoma ot the liver Cancer ot the cecum Cancer of the head of the pancreas Lymphoid leukemia Age at Death (Years) 73 56 50 41 54 42 57 64 62 40 53 53 53 59 43 48 53 52 54 48 Total Exposure (Years) 15 16 15 11 16 19 17 26 18 8 25 24 22 29 7 22 5 26 12 21 Latency Period (Years) 28 17 15 11 16 19 23 29 23 13 28 24 23 30 10 22 12 26 12 21 ceed 100, they are not statistically significant. On the other hand, deaths by accident and violence are signifi cantly lower (p < 001). As for specific cancer deaths, a significant excess is noted for cancer of the digestive system only. This excess is accounted for exclusively by cancer of the liver (8 cases observed versus 0.14 expected, SMR 5,714.29). No other cancers are in excess. The cancer mortality among the workers who were still alive 15 years after their first exposure shows the same results as the mortality in the entire group, namely that only cancers of the digestive system are in excess in the exposed cohort (Table 6). Discussion Difficulties encountered in comparing the mortality of a group of workers with the mortality of a whole popula tion are well documented. They are termed: "healthy population selection effect," "survivor population effect,"20 "ethnic/cultural/social characteristics." They are believed to underestimate the actual mortality of a group of workers and therefore to underscore the effects of the risks to which these individuals are being exposed at work. There are no easy ways to avoid such difficulties One method would be to compare a group of workers with another group of workers with the same characters tics except for the exposure to the risk under study. The problems encountered in so doing are no easier. Although the group most similar to industrial workers is probably another group of industrial workers from the same town it is almost impossible to think of a group of industrial 1 Table 4. -- Deaths Observed In the VCM-Exposed and Unexposed Cohorts, Age-Adjusted Relative Risks and Confidence Intervals of the Relative Risk. Cause ef Death Ail cancers (140-209) Mouth and pharynx (140-149) Digestive (150-159) Respiratory (160-163) Bone, skin, connective (170-173) Urogenital Eye, CNS (185-189) (190-192) Leukemia (200-209) Other cancers Cardiovascular (390-458) Respiratory (460-519) Digestive (520-577) Accidents and violence (800-899) Others Total "Significant according to Miettinen's test*? fOne cause unknown Z7A No. el Casas Exposed Unoxposad 20 52 02 14 9 2 20 22 17 02 14 06 25 124 6 15 4 12 2 15 2t 15 59 233 Relative Risk 1.48 -- 6.25 0.36 1.67 0.83 1.16 0.80 0.81 1.44 0.51 0.57 1.07 95% Confidence Interval of the Relative Risk 0.84- 2.61 -- 2.69-14.52* 0.08- 1.58 0.21-12.99 0.10- 7.10 -- 0.07-17.80 0.47- 1.36 0.44- 9.35 0.49- 4.25 0.21- 1.24 0.14* 2.39 0.79- 1.45 J ASI 00005523 - lln workers not c In the presen exposed wori group of un among the ur found to be ir with that of t The comp, the unexpose reached by ti The nuVnber c cantly higher in excess and tically signific Considering posed to VO of the liver b (skin, lung t glands)2' 22 25 on VCM-expc cancer,24 the are surprising cems the asc tive studies c ords it someti or leave get I *Journal fif (\rr ) *- Latency Period 17 15 11 16 19 93 29 23 13 28 24 23 30 10 22 12 26 12 21 hole populaed: "healthy population cs." They are ity of a group effects of the ' exposed at h difficulties, p of workers ^^aracteris^Bpdy. The iW^lthough s is probably e same town, of industrial fidenee si ths Risk 2.61 4.52* 1.58 2.99 7.10 7.80 1.36 9.35 4.25 1.24 2.39 1.45 fault and Allard Table 5. - Observed and Expected* Numbers of Deaths in the VCM-Exposed Cohort and Standardized Mortality Ratios (SMRs). Cm* of De*th All cancers Mouth and pharynx Digestive Respiratory Bone, skin, connective Urogenital Eye. CNS Leukemia Other cancers Cardiovascular Respiratory Digestive Accidents and violence Others Total (140-209) (140-149) (150-159) (160-163) (170-173) (185-189) (190-192) (200-209) (390-458) (460-519) (520-577) (800-999) Observed 20 0 14 2 2 1 0 1 0 25 6 4 2 2t 59 Expected 16.37 0.64 5.40 5,78 0.38 1.33 0.60 1.67 0.54 31.67 3.21 3.85 10.58 5.40 71,07 SMR 122.17 -- 259.26t 34.60 526.32 75.19 -- 59.88 78.94 186.91 103.90 18.90T 37.03 83.02 `Expected numbers obtained from 1971 males Quebec death rates applied to person-years in each age group tp< 0.01 tOne cause unknown workers not exposed to some kind of carcinogenic risks. In the present study, the authors decided to compare the exposed workers with both the general population and a group of unexposed workers. It must be noted that among the unexposed workers, no causes of death were found to be in excess when their mortality was compared with that of the general population. The comparison of the mortality of the exposed and the unexposed workers gave results similar to the one reached by the comparison with the general population. The nu*mber of cancer of the digestive system was signifi cantly higher in the exposed cohort No other cancer was in excess and there even seemed to be a deficit (not statis tically significant) in respiratory cancers. Considering the fact that animals experimentally ex posed to VCM have developed not only angiosarcomas of the liver but also malignant tumors of several organs (skin, lung, bone, Zimbal glands, kidneys, mammary glands)*' ** ** and since several epidemiological studies on VCM-exposed workers have found an excess of lung cancer,*4 the results of this study of heavily exposed men are surprising. Three questions can be raised. One con cerns the ascertainment of the population. In retrospec tive studies conducted from old company or union rec ords it sometimes happens that records of people who die or leave get lost These losses could account for the low cancer rates observed. But this is a remote possibility, since the approach followed was a historical-prospective retrospective one, which began with old lists of workers composed in 1948 and 1952 and followed almost every one through time, adding new workers its they first ap peared on more recent lists. This was done without knowl edge of exposure to VCM. A second reason for low lung cancer rates could be a bias in the classification of persons as exposed or not ex posed. The classification was made essentially on the basis of personal recollection for living persons and by family members with confirmation by co-workers for deceased persons. To minimize such bias, much attention had been given to the occupational history (with indica tion of the buildings the subject had worked in) and workers with VCM exposure shorter than five years were rejected. There remains the possibility that some of the men who died of liver cancer, had they lived longer, might have developed lung cancer. The small size of the population under study increased the probability of not detecting an excess that was actual ly present This was a case where "the number of person/ years at risk may be too small to detect an increased risk of cancer."*1 It was particularly true for rare cancers such as cancer of the brain and cancer of the lymphatic system. Inasmuch as cancer of the lung was concerned, Table 6. -- Relative Risks and SMRs for Cancer Deaths Among Men Having an Observation Period of IS Years and More. Cancers All cancers Mouth and pharynx Digestive Respiratory Bone, skin, connective Urogenital Eye. CNS Leukemia Other cancers (140-209) (140-149) (150-159) (160-163) (170-173) (185-189) (190-192) (200-209) Retattv* Ne. Risk 15 1.30 0-- 11 5.68 2 0.35 0-- 1 0.82 0-- 1 0.59 0-- 9SS Confidence Interval of tho Rotative Risk 0.97- 1.74 2.72-11.58 0.25- 4.73 -- -- -- -- *P<0.01' Journal of Occupational Medicine/Vol. 23, No. 10/0ctober 1981 SMR 129 -- 281* 47 -- -- -- -- 675 the situation was somewhat different. When one applies the Quebec lung cancer death rate after adjustment for age to the exposed cohort the probability of finding an excess twice as high as in Quebec was 73%. This proba bility would have been 99% had the excess been three times as great in the exposed group." In view of the fact that the smoking experience of the exposed workers was similar to that of the unexposed and considering that the subjects in this study are believed to have been heavily exposed to VCM, it seems reasonable to conclude that the excess of lung cancer associated with VCM exposure, if present, is indeed rather small. 7. Delorme F and Theriault C. Ten cases of angiosarcoma of the liver in Shawinigan, Quebec. I Occup Med 20:338-340.1978. 8. Monson R. Peters j, and Johnson M: Proportional mortality among vinyl chloride workers. Lancet 2.397-398, 1974. 9. Waxweiler |R, Stringer W. and Wagner J|: Neoplastic risk among workers exposed to vinyl chloride. Ann NY Acad Sci 270:40-48.1976 10. Special Report by the Editors: PVC aims lor a bigger share in plastics. Can Chem Process 49:41-44.1965. 11. Conseil des Sciences du Canada: Vue d'ensemble des dangers de la contamination par le chlorure de vinyle au Canada. Quebec Science (Suppl) :1-24, 1977. 12. laRochelle F. Shawinigan depuis 75 ans. Ateliers de Timprimerie Publicity PSquet Enr., 1976. pp 497-525. 13. Hill ID- Computing man years at risk. Br) Prev Med 26-132-134 1972. IV II 14. Mantel N and Haenszel W. Statistical aspects of the analysis of The authors wish to express their gratitude to the following who contributed to this study workers, union representatives and management personnel of the companies Canadian Resins and Chemicals, Shawinigan Chemicals. B F Goodrich and Gulf Canada, Drs H Hurra and I Fabia, epidemiologists, and Or F Delorme, pathologist, for their help and advice, the severs) hospitals con tacted, and the several persons who assisted with the collection of information and the writing of this article data from retrospective studies of disease. I Natl Cancer Inst 22 710-748, 1959, 15 Mantel N. Evaluation of survival data and two rank order statis tics arising in its consideration. Cancer Chemother Rep 50.163-170 1966 16. Miettinen OS Standardization of risk ratios. Am ) Epidemiol 96 383-388, 1972. 17. Miettinen OS: Simple interval-estimation of risk ratio Am / 1? Eliza Epidemiol 100:515-516,1974. References 16 Gaffey WR; A critique of the standardized mortality ratio. J Oc 1, Creech |L |r and Johnson MN. Angiosarcoma of liver in the man cup Med 3.157-160,1976 ufacture of polyvinyl chloride I Occup Med 16:150-151, 1974. 2 Tabershaw IR and Caffey WR: Mortality study of workers in the manufacture of vinyl chloride and its polymers. / Occup Med 16:509-518, 1974. 19. Bailar JC and Ederer F: Significance factors for the ratio of a Poisson variable to its expectation. Biometrics 20.639-642.1964 20 Fox Al and Collier PE: Low mortality rates in industrial cohort studies due to selection for work and survival in the industry Br) Prev n and r a rut. 3. Nicholson W), Hammond EC, Seidman H. et al Mortality experi Soc Med 30.225-230, 1976. the P- ence of a cohort of vinyl chloride polyvinyl workers. Ann NY Acad Sci 246 225-230, 1975. 4 Duck BW, Carter JT. and Coombes E|- Mortality study of workers m a polyvinyl-chloride production plant Lancet 2-1197-1199. 1975. 21. Viola PL- Cancerogemc Effect of Vinyl Chloride. Vol. 29. X Inter national Cancer Congress, Houston, 1970. 22. Viola PL, Bigotti A, and Caputo A: Oncogenic response of rat skin, lungs and bones to vinyl chloride. Cancer Res 31-516-519, 1971 prevr June. USi 5 Byren D, Engholm G, Englund A, and Westerholm P: Mortality 23. Maltoni C and Lefemine C: Cancerogenicity bio assays of vinyl deatt *i ; and cancer morbidity in a group of Swedish VCM and PVC production workers. Environ Health Perspect 17:167-170,1976. chloride. Current results Ann NY Acad Sci 246:195-218,1975 24 Wagoner )K: Vinyl chloride and pulmonary cancer. / fnv Pathol male ing re 6 Fox Al and Collier PF: Mortality experience of workers exposed to vinyl chloride monomer in the manufacture of polyvinyl chloride in Toxicol 1.361-362, 1978, 25. Pell S, O'Berg MT, and Karrh B. Cancer epidemiologic surveil ph)* Great Britain Br; Ind Med 341-10.1977. lance in the DuPont Company. / Occup Med 20:725-740,1978. from serve. !I 3.3 e obser ) rubbe from Reducing Financial Risks place tiona There is too little evidence that the nuclear industry, which was never very good at calculating its own interest, is seriously contemplating the effect of another Three Mile Island on its future -- or its fortunes. The industry has already returned to complaints that the safety bureaucracy is nitpicking it to death, running up construction costs and delaying licenses. The latest complaint is that, in the aftermath of the TMI accident, NRC diverted safety reviewers to deal with operating reactors and fell behind in ap proving new licenses. may : clem decre dart serve and v mult. There is a kind of tunnel vision in the failure to see the connection between safety and the special financial risk involved in nuclear investment To use the current energy crisis to avoid safety requirements is to ignore the fact that safety is essential to protec In ting the heavy investment in nuclear power. Some hope lies in the fact that, although the industry thinks safety requirements are a pain in the neck, they may be forced to look at them as an investment in financial public relations to alleviate the bankers' skepticism expe; for fi Were 1974 -- Victor Cifinsky. Commissioner, the Nuclear Regulatory Commission, as quoted in amor "Notable and Quotable," in The Wail Street lovrn&l. May 21.1961 Fror Huvtr Supt ftSI 0000SS2S kqondh-