Document 4a7p71jDORNM70dew1XKvwNKQ

' O-*- 5ei***** . , U&erOO'Y Ql Thr CrCy Unl*nily ol Maw Vom PROGRE- REPORT ?POM THE 1NSU; '1 fv USTJ LY HYGIENE RESEARCH PROGRAM PLAINTIFFS 4 EXHIBIT - Irving J. StliaoH, W.O Program Qir:e/ Vol.S.No.3 -xa>-' Pal! 197*1 The International Agency/or .Research on . interest in the health problems associated with Cancer of the World Health Organization has insulation work, we are devoting this entire just issued its Scienti/ic Publication No. S, issu$ to the publication of a paper/rorn the Biological Effects ofAsbestos, consistingofthe lAPC volume by Dr. Irving Selikoff, Dr. E. Proceedings of a Working Conference held at Cuyler Hammond and Mr. Herbert Seidman the Agency's Headquarters in Lyon, France, which treats the problem of cancer risk as October 2-6, 2372. In view of the considerable sociated with insulation work. Cancer risk of insulation workers -in the United States1 . I. J. SELIKOFF, E. C. HAMMOND & H. SEIDMAN In 1963, information became available which in on the possible relationship between asbestos ex dented that asbestos insulation workers employed posure and the occurrence of gastrointestinal cancer. ,.i the construction industry were subject to a sig nificant cancer hazard associated with their work (Selikoff et al.t 1964). Studying the mortality ex perience of the 632 members on the rolls of the New The data were insufficient, however,, to resolve a number of questions which are important for r. full evaluation of the problem and to provide guidance for the establishment of needed administrative and York-New Jersey branches of the insulation workers' industrial control measures. To obtain such ad union on 1 January 1943, for the 20-year period end ditional information, we have continued and ing 31 December 1962, it was found that death from cancer was almost three times as frequent as ex pected. The death rate from cancer of the bronchus and pleura was 6.S times as high as that for the gen eral US white male population, both age and date being taken into consideration; and cancer of stomach, colon and rectum was found to be three widened our investigations. CURRZhfT studies Using approaches and methods previously re ported (Selikoff et al., 1964; Selikoff et ah, 196ab), we have studied the experiences of three cohorts of workers exposed to asbestos insulation materials. times as common. Attention was called to the occurrence of pleural and peritoneal mesotheliomas (Selikoff a aL, !965a). These studies were subsequently extended to in vestigate the interrelationships between asbestos exposure and cigarette smoking in the production of lung rn.ocer, the significance of mesothelioma "'elikoff et a/., 1970) and to obtain further evidence 1. New York-New Jersey cohort. We have con tinued our surveillance of the New York-New Jersey insulation workers. By 31 December 1962, 261 had died; thus, 370 were still alive on 1 January 196?. Each has been followed upuntil 31 December 197k Two additional groups of workers have been investigated: 2. In one study, we registered the entire member, 1 From iht Environmental Cancer Research Project of the American Cutvccr Society and the Mount Sinai School of Medicine of the City University of New Yor!c ship of the insulation workers union in the United States and Canada2 on 1 January 1967 (17,UJ'J; including members of the New York-New Jcr'-. y 'wa|s mentioned above), and recorded many charac cancer risk first identified in 1963. Ey 31 December teristics for each man including date of birth and onset of insulation work. Each man has since been observed until 31 December 1971. 1971, 430 of the original 632 (New York-New Jersey cohort) had died, 421 after reaching 20 years from onset of employment. One hundred and 3. A third group consisted of the entire workforce eighty-nine (45% of the 421 total deaths) died of (]941-4S) of a factory in on eastern US city which cancer, whereas 47.2 such deaths had been expected manufactured amosite asbestos insulation materials (Table 1). ' (including insulating block and pipe covering, and The average age of the insulation workers through asbestos mattresses), for use by insulation workers out the United States and Canada (US-Canada in the construction industry, especially In ship con cohort) was much lower, and they therefore had struction and repair. The plant opened its doors in June 1941 and remained in business until November 1954. Very few of the production workers studied by us had had any prior occupational exposure to asbestos. Between 1941 and the end of 1945 a total of 933 men were hired; of these, some worked for as little as one day, while others worked until 1954 when the plant closed. Results of physiological studies of men in this plant have been reported (Sader el a!., 1961). Recently, we have traced the entire group up until 31 December 1971 in order to evaluate the considerably less work experience than that of the New York-New Jersey cohort (Hammond & Sellkoff1). Nevertheless, the excess of car.cer deaths was very much the same, these deaths occurring among the older men who had longer work experience. Relatively few deaths were found among the younger men, despite the large number in the cohort. 144.09 cancer deaths were expected and 459 occurred, representing 41% of the total of 1092 deaths observed in the period 1967-71 (Table 2). cancer risk among them. Eight hundred and seventyseven (94%) were fully traced, and the remainder were partially traced for varying periods. We have analysed their mortality experience in relation to the duration of employment, age at onset of employ ment and lapsed period from onset of employment. Lung cancer Bronchogenic carcinoma was found to be a leading cause of death among asbestos insulation workers, accounting for approximately 45% of fata', neoplasms. In the first two series reported, it was responsible for 20% of all deaths and was almoi*. ' FINDINGS IN INSULATION WORKERS IN THE CONSTRUCTION INDUSTRY three times as common as were pleural and peritonea; mesothelioma. Recent histological studies of tu mours from these patients show the distribution of The experiences of the two cohorts of insulation workers in the construction industry (New York-New Jersey and US-Canada cohorts) amply confirm the cell types to be that of lung cancer found ir. the general population (Kannerstcin Sc Ciurg, 2972;, and our clinical experience indicates that their prog nosis was equally poor. These findings stress 3 International Association of Heat and Frost Insulators urgency of preventive measures, such vs dust control and Asbestos Workers, AFL-CIO. lnsuiairpn Hygiene Progress Reports Vol.G. No. 3 Fall 1974 from I he Insulation Industry Hygiene Research Program Advisory Council of UHR? Irving J. Selikoff, M.D., Program Director and Chairman E. Cuyler Hammond, Sc.D., Vice President, American Cancer Society, New York. N.Y. Andrew Haas. General President. International Association of Heat and Frost Insulators and As* bestos Workers, Washington. O.C. Fred L. Pundsack. Ph.D..' Vice President. Re search and Development. Jolms-Manville Cor poration. Denver. Colo. Wlilur: W. J. Nicholson, Ph.D., Published ;t( th* f-.min.rmwntalSciences Laboratory (Jrvinjt J. ScliVoM. M.D.. Diiectnr). Mount Sinai School of Medicine of the City University ot New York. N Y. 10029 PURPOSES OF THE INSULATION INDUSTRY HYGIENE RESEARCH PROGRAM 1. Todeve'.op improved methods for minimizing exposure of insu lation workers todusis and furr.r*; encountered in their work. 2. To disseminate knowledge of these improved methods of dust control wherever they m:;y c-r applied advantageous!)* and to offer cooperation, advice ar._ assistance toward iheir univer:.:>.' adoption. Tnpie 16. .Expecteu5 and observed number of deaths among 623 New York-New jersey asbestos insulation workers V . .nuary 1 DO-31 December 1 971, twenty or more years alter onset of first exposure to asbestos 1943-51 1952-61 1962-71 Total. 1043-71 Sxpteied Observed Expected Observed Expected Observed Expected Observed ''oral cancer: all sites - ' 11.3 28 18.3 57 17.6 104 47.2 1 29 Lung cancer ' 1.5 13 3.5 23 5.1 48 10.1 .24 Pleural mesothelioma n.s.e 1 n.a. 2 n.a. 5 n.a. 0 Pcritonuol mesothelioma n.s. 1 n.a. 3 n.a. 20 n.a. 24 Cancer of stomach, colon, rectum. oesophagus 4.1 7 ' 5.0 18 3.9 16 13.0 41 Cancer all ocher sites 5.7 6 9.8 11 8.6 15 24.1 32 Asbestosis .- n.a.- - - . 1 .. . n.a. 10 n.s. 22 n.a. v A'.l other causes - 65.1 44 89.4 94 73.3 G1 232.0 19sI Total all causes '` 73.4 73 107.7 161 95.9 137 220.0 421 Person-years of observation 3726 4406 2898 11.020 0 632 members were on the union's rolls on 1 January 1343. Nine died before reaching 20 years from first employment. these calculations upon reaching the 20-year from onset ol first exposure point. - 5 Expected rates are based upon age-specific death rate data of the US National Office of Vital Statistics from 1949-07. polated 1 943-48 from rates for * 949-55, and for 1 968--71 from rates for 1 961--67. 6 US heath rates not available, but these are rare causes of death in tha general population. All others cm*r?u Rates were extra 3nd the avoidance, of cigarette smoking (Selikoff vation that fibres and fibrils may be found in many er ah, 196S; Hammond & Selikoff1). . organs, admits the possibility that a neoplastic effect exists for sites other than those heretofore Identified. Ccstro-inieslina! cancer Examination of this hypothesis is hampered by the In oar initial reports (Selikoff et a!.f 1964; Ham infrequency of some such tumours in general and by mond et ai, 1965) attention was called to the then, the as yet inadequate number of observations avail unexpected finding of a moderate excess of gastro able. We do not at present find it useful, therefore, intestinal cancer among New York insulation r-to comment on this question, although we have been 'Orkers. Nevertheless, relatively few deaths were studied and firm conclusions were not considered warranted. We have now collected additional data, particularly interested in cancer of pancreas, brain and oro-pharynx, in genito-urinary cancer, and lymphoma and leukaemia. ' and these give the same indication very much at the same level of excess, i.e., at two or three times ex pected deaths (Tables 1 &. 2). These findings were made for cancer of the oesophagus, stomach, and of tusk with Factory exposure to asbestos INSULATION MATERIALS the colon and rectum. A similar excess has been found among insulation workers in Belfast (Elmes & Simpson, 1971). Taken together, these experiences suggest that the increases real. While they are not responsible for a major proportion of excess deaths (lung cancer, mesothelioma and asbestosis account In 1964, Smith (1965) called attention to the multi plicity of materials to which insulation workers were exposed. It seemed justified to seek evidence-on whether asbestos insulation materials per se were specifically related to various diseases. If not, in for more), the incidence of gastro-intestinal cancer is vestigation of other insulation materials would be nevertheless substantial: in the New York-New required. We therefore studied the work force of jersey group 13.0 deaths were expected and 41 the insulation materials factory noted above, in occurred; and in the US-Canada survey, 27.3 were which exposure was only to asbestos insulation, and exp-uesd and 55 occurred. Moreover the finding not to other insulation materials used in the con may be of considerable theoretical importance, in struction industry (e.g., fibrous glass, calcined diato- view of the dearth of useful hypotheses concerning maceous earth, rock wool, etc.) the etiology of gastro-intestinal cancer in general, Four hundred and eighty-four deaths are known nd of cancer of the colon and rectum in particular. to have occurred among the S77 men fully traced up Other neoplasms to 31 December 1971. The distribution of deaths strongly resembled that seen among insulation wor The knowledge that a number of tissues are subject kers who install such materials, despite the fact that to asbestos-induced cancer, coupled with the obser observation of this cohort has reached only 25-f v  ' . Total deaths Jancer: ail sites . Lung cancer Pleural mesothelioma Peritoneal mesothelioma Cancer of stomach Cancer of colon, rectum Cancer of oesophagus AM othor cancers Asbestosis All othur causes Number of men Person-years of observation ' Total Expected Observed 805.63 144.09 44.42 n.a.* n.a. 6.62 17.51 ~ 3.21 ' 72.33 n.a. 661.54 1092 459 213 26 51 16 25 13 114 73 555 17.800 86,300 . Distribution by duration from onset of exposure Loss than 20 years 70 yeers and mom Expected Observed Expected Observed 173.94 26.31 7.03 n.a. n.a. 0.97 2.51 0.44 15.26 . n.a. 152.63 211 51 22 2 1 3 1 19 -5 155 626.69 117.7C 37.39 n.a. n.a. 5.63 15.00 2.77 56.97 n.a. 503.91 - 291 403 191 24 42 15 07 12 95 7b 400 12.681 02,673 5.119 22.627 o Expected deaths are based upon age-specific death rate data of the US National Office of Vital Statistics. Rates for 1968-71 were extra- pointed from rates for 1 961 ^>7. b US death rates not available, but these are rare causas of death in the general population. years from onset. Thus, lung cancer accounted for 73 deaths, although only 11.41 were expected; and gastro-intestinal cancer for 26, with 12.36 expected (including oesophagus). Mesothelioma caused seven deaths, and asbestosis 27 (Table 3). . These data indicate that asbestos insulation ma terials are capable of causing the disease patterns seen among insulation workers; there is no need to invoke the influence of other insulation materials. EPIDEMIOLOGIC VARIABLES Lapsed period from onset of exposure It has long been inferred, on the basis of rather limited data, that asbestos cancer largely occurs more than twenty years from onset of exposure. Experiences in the cohorts studied indicate that lung cancer may occur in considerable excess as early as 10-14 years from onset, with 0.65 expected deaths versus 10 observed among those working in the amosite factory for more than one year (Table 5b). Moreover, death rates for lung cancer were signifi cantly increased in the US-Canada insulation wor kers cohort 15-19 years from onset (4.S6 deaths were expected and 18 occurred) (Table 4). . On the other hand, in terms of numbers; most deaths from lung cancer occurred among insulation workers 30-39 years from onset of exposure, or Tablft 3. Expected* and observed deaths among 933* amosite asbestos factory workers first employed 1941--15, and observed to 31 December 1971 Before 1952 1952-61 1962 -71 . Total. 19 41-71 Total deaths Total cancer: all sites Lung cancer pleural mesoihelioma Paritont.il mesothelioma Cancer ol stomach Cancer of colon, rectum Cancer of oesophagus All other cancers Asbestosis AM other causes Expected 69.98 9.81 1.49 n.a.e n.a. 1.45 1.55 0.27 5.04 n.a. j 60.17 Observed 88 15 3 1 0 2 2 0 6 3 70 Expected 103.27 18.72 2.71 n.a. n.a. 1.24 2.C1 0.45 10.11 n.a. 89.55 Observed 146 44 22 0 0 4 5 0 12 7 95 Expected 121.24 21.63 6.21 n.a. n.a. 1.29 2.83 0.51 10.74 n.a. 93.61 Observed 250 84 47 2 /. e. 8 0 19 17 149 Expected ' Obr.cn/cd 229.49 50.16 11.41 n.a. n.ji. 4.62 7.05 1.22 25.89 n.a. 243.22 . ! i I ' i j | ! | I i i f pl .' 142 72 j 1* VJ e 27 27 314 * r--nct?d rates are based upon age-specific death rate data of the US Notional Office of Vital Statistics from 19*19-67. Rates .vers extr?dated for 1941--43 from rates for 1949-55. and for 1953-71 irom rates (or 1 961-67. * 922 rr.cn were employed. In 5 cases, ages were not known and these men have been excluded Irom these calculations. 377 men -* traced to death or to 31 December 1971. 51 men were partially traced and remain in the calculations until lost to observation. c US death rates not available, but these are rare causes of death in the general population. J*u!lv*the effects of asbestos exposure unless observa tion was possible for at least 40'years from the onset of work* exposure. Age at onset ' \ number of methodological difficulties are en countered in studying the effect of age at onset on lung cancer rates. Not the least are the secular changes in cigarette, smoking habits in the past 50 years (Hammond, 1966), and the concurrent in creases in lung cancer death rates in the general population, which must be {aken into account. With these in mind, there seem to be no striking variations in the incidence of lung cancer as a result of differences in age at which asbestos work exposure first began (Table 6). . among 17.200 osoestos insulation workers in trie US tme Canada. 1 January 1937-31 December 1071: relation to elapsed period from onset of work exposure Years from Lung cancer Pleural mesothelioma Expected Observed Ratio Observed deaths* deaths deaths <10 0.43 0 0 10-14 1.69 /, 2.4 0 15-19 4.85 10 2.7 2 20-24 7.55 25 3.3 <e 25-29 8.50 41 4.2 7 30-04 6,24 44 7.1 4 35-39 40-44 3.53 4.04 23 24 6.55.9 1 45-49 3.72 17 4.3 ' t, 50+ . 3.31 17 4.5 1 Total 44.42 j 213 4.3 25 * Expected deaths are based upon age-specific death raio data of the US National Office of Vital Statistics. Ratos tor 1 062--71 ware extrapolated from data for 1261--67. Table 5a. Age at onset of employment, 1941--45. of 933 workersin an amosite asbestos Insu lation factory Duration of employment Age at employment Total . <3 months 3-11 months 1 + years 15-19 20-24 ' 25-29 ' 30-34 . 35-39 40-44 . . 45-49 . 50-54 55-59 CO-04 65-69 70-74 . 75-79 . Unknown 94* 114 120 114 106 101 83 90 55 30 , 16 4 1 5 42 35 28 32 27 .... 30 ,. 29 2G 14 6 6 1 -- 2 30 34 48 40 40 38 21 27 21 12 4 3 -- 3 21 45 - 44 42 39 33 33 37 20 12 6 -- 1 -- Total 933* 278 * Includes one man with unknown length of employment. 32T 333 Table 5 b. Lung cancer among 326 amostte asbestos workers employed for 1 year or more starting 1941-45, and observed to 31 December 1 971 .* Expected and observed deaths b . Attained years from first employment 1-9 10-14 19-19 20-24 25-29 Total Exp. Obs. 0.68 2 0.G5 0.72 10 11 1.03 1.01 4.09 8 14 45 <40 Exp. Obs. 0.02 0 0.02 0 0.01 0 ---- ---- 0.05 0 Attained ages 40-49 50--59 Exp. Obs. Exp. Obs. 0.03 0 0.30 2 0.07 0 0.19 4 0.07 0 0.20 3 0.10 2 0.29 5 0.06 1 0.27 4 0.39 3 1.25 18 60-69 Exp. Obs. 0.22 0 0.29 5 0.26 4 0.39 1 0.37 7 1.53 17 70--72 ' t 80-29 Exp. Obs. | Exp. Obs. 0.05 0.08 0!--!-- \' 1_ oo o 0.17 r. 0.22 0.26 0 i 0.02 i 0 | 0.05 j C fJ 0.78 11 1 0.09 0 j .> Scvsn were omitted, four because of prior occupational exposure to asbestos and three who had not baen traced after tho first year. Rive were pa/'iaJiy traced and remained in the calculations only to the date of last observation. All others entorod the calculations a.'tir lust year i/om onset. * Expected rates are based upon age-specific death rate data of the US National Office of Vital Statistics from 1949-67. Ratos were extrapolated for 1041^18 from raws for 1249-55. and for 1968-71 from rates for 1961-67. 8. 'Expected and observed deaths of Jung cancer an.*.*g!7,20Q asbestos insulation workers, 1 January 1967 31 December 1971 ; distribution by age at onset of exposure A? at onset of exposure <25 25-34 25-44 45 + Deaths of Jung cancer Expected* 10.19 .15.20 0.73 2.25 Observed 102 56 30 7 Ratio 5.G1 4.34 4.33 2.11 ' 0 Expected deaths are baaed upon age-specifiedaath rete data of the US National Office oI Vital Statistics. Rates for 1 96^71 were extrapolated from rates for 1961-^57. ;/iif<w; u- I ..,/uJi.* t In one cohort, it was possible to study the effect of differences in length or exposure. While all the in sulation materials factory employees stance! work between 3941 and 1945, approximately one-third of these worked for less than three months, another third for three to eleven months, and the remair.irv.; third for a year or more (Table 5a). All worked in the same factory and had similar exposure to the same asbestos starting at the same point in time. The only significant variable was duration of expo sure. All were traced to the end of 1971. Our findings indicate that duration of exposure, and thus presumably total dose, has an important influence on lung cancer rates. Those with less Table 7. Expected and observed deaths of Jung cancer among 876 amosite asbestos factory workers, first employed 1941--45, and observed to 31 December 1971. Distribution by duration of employment Duration of employment <3 months 3-11 months 1 + years Total Number of men 256 294. 326 876 Person-years of Observation 5.369 6,153 6,912 18,339 Deaths of lung cancar Expected9 Observed Ratio 3.55 3.58 4.09 13 15 45 3.66 4.19 11.00 11.22 73 6.51 a This table excludes 57 man. Ten died during first year of employment, 39 could not be traced after the first year, 7 had prior occupational exposure to asbestos and 1 had employment of uncertain duration. Seventeen men of the 87S were partially traced and remained in the calculations until lost to observation. Expected rotes are based upon age-specific death rate data of the US National Office of Vital Statistics. 1949-67. Rotes were extrapolated for 1341--43 from rotes for 1943--55. and for 1968-71 from rates for 1961--67. Table 8. Mortality "vpenenco of 337 asbestos insulation workers 1 January 1 963--31 December 1 971, analysed by dura tion of shipyard employment . Cause of death Total cancer: all sites Luny cancer Pieurai m-'-iothelioma Hcniontji intsothelioma Cancer of stomach, colon, rectum, oesophugus Cancer ail other sites Asbs>tost> Ail other causes Total deaths None (33) Observed Expected11 25 3.8 11 1.2 n n.a.9 6 n.a. 3 0.8 3 1.8 4 n.a. 8 16.2 37 20.0 Shipyard employment Up to two years (112) Observed Expected 28 4.8 9 1.4 2 n.a. 6 n.a. 2 1.0 9 2.4 11 n.a. 18 20.3 57 25.1 Three or more ysars (121) Observed Expected 29 5.2 17 * r 0 n.a. 4 n.o. 5 1.2 2 2.5 5 n.a. 21 2U.C 55 22.4 Extracted deaths are based upon age-specific death rate data of the US National Office of Vital Statistics. Hates for 1950-1971 wire extrapolated from rates for 1961-1367. # US death rates not available, but these are rare causes of death in the general population. inu'-.U)' v*wi.\. January 1Uo/-->i 'jccuiiUhi *>/! * ' Workmen registered in union locals in south-west, mid-west and central states Expected Observed Total deaths ' Total cancer: all sites . 316.28 56.2A Lung cancer 17.47 Pleural mesothelioma n.a.* Peritoneal mesothelioma n.a.- Cancer of stomach 2.56 ' Cancer of colon, rectum ' "* -6.75 - -- Concur of oesophagus . 1.26 All other cancers 1 28.30 Asbestosis n.a. All other causes - 259.94 Number of men Person-years of observation 7.289 35.310 446 172 79 no 13 6 12 6 46 26 248 Workmen registered in shipyard locals Expected 39.56 7.91 2.67 n.a. n.a. 0.26 0.98 0.20 3.70 n.a. 31.65 Observed 462 2.244 24 1C 9 2 1 0 0 1 3 3 15 " Expected deaths are based upon age-specific death rate data of the US National Offiee of Vital Statistics. Rates lor 1963--71 were extrapolated from rates for 1961-57. 6 US death rates not available, but these are rare causes of death in the general population. . . than three months of work showed a definite increase lation workers in the New York-New Jersey group in lung capctr risk (13 observed versus 3.55 expected), according to their duration of shipyard employment and those with three to eleven months' exposure (ship insulation being fairly important in the New much {he same increase; while workers employed for York harbour area, especially between 1939 and one year or over had a much more substantial in 1945). Of the 370 men in the union on 1 January crease (45 observed, 4.09 expected) (Table 7). .1963 who were survivors of the 1943 cohort, v/e were r able to obtain histories of shipyard employment by Fibre type .. Crocidolite was not used for insulation work in the United States during the period covered by our personal interview from 337. The resuits indicate that the mortality experience in men With shipyard employment, with its greater potential for amosite exposure, did not differ substantially from that of study. Indeed, very little crocidolite was imported men in the same union, at the same period of time, into the United States until after the Second World without such employment (Table S). War (Selikoff et ah, 3970). Thus, exposure to this In the US-Canada survey, each union member fibre cannot explain the cancer risk observed. was registered in a specific union local, several of Both chrysotile and amosite have been used in which were engaged entirely in shipyard work. insulation materials in the US, the former since the Others, in the central and mid-western pans of the turn of the century (Selikoff et a1965b) and the country, had no such experience. This is not to latter since the mid-1930's (Selikoff et ah, 1972). say that individual members of these locals did not, Amosite has been used principally in ship insulation in the past, do some shipyard work, but it Is unlikely and chrysotile in general construction and industrial to have constituted a major part of their work- work, although it is also found in some shipyard materials. It is difficult to compare the relative disease poten tial of the two varieties of fibres in insulation work, since cohorts which differ in this respect may also differ in other respects, including type and intensity of exposure, duration of exposure and lapsed period om onset of exposure. Nevertheless, two sets of observations are available which may be of interest. history. We have analysed the mortality experience of men in the two groups oflocals between 1967 and 197!. There does not seem to have been any significant difference between them (Table 9). At least as re flected in these two sets of data, construction indus try insulation work, with its preponderance of chrysotile exposure, was no more or less hazardous than was shipyard work, with its wider use of amositu We have compared the mortality experience of insu materials. * , iA serious cancer risk has been Demonstrated among It would be difficult to evaluate fully the effects of asbesto: 'asbestos insulation workers in the United States. Ap exposure without observation for at least 40 years from proximately one death in five, an extraordinary incidence, onset of exposure. has been the result of lung cancer. Gastro-intestinal Data have been presented indicating that these risks cancer was more than doubled in incidence, and meso arc associated with the asbestos insulation mater:;:! . thelioma was responsible for 7% of all deaths. perse. An Increased incidence of lung cancer was seen as little . We were not able to find evidence that the use or as 10--14 years from the onset of work exposure; we sus chrysotile in insulation work was associated with a great ,*r pect that this tends to be associated with more intense risk than was amosite, or vice versa. V7e have no kno v. exposure, as among factory workers. However, the ledge of the comparative effect which crocidolite might greater increase occurred 30-45 years from onset, especi have; it was not used in US insulation work. ally among the less intensely exposed insulation workers. ACKNOWLEDGEMENTS We wish to express our indebtedness to the skilful assistance of our staffs. We are particularly grateful to Mrs Janet S. KafTenburgh, Mrs Selma Annenberg, Mrs Frances Perez, Mrs Dorothy Perron, Mrs Shirley Levine, Mrs Kayla Margoles and Mr Charles V. Nolan for the successful tracing of the subjects in these epidemiological studies, and to Mr Lawrence Garfinkel and Mrs Doris Fleisher for their devoted help in many aspects of the work. ' This work was supported in part by US Department of Health, Education and Welfare research grants OH 00320 and ES 003.53, and by research grant U 1272 of the Health Research Council of the City of New York, to the Mount Sinai School of Medicine. . REFERENCES Bader, M. E., Bader, R. A. & Selikoff, I. J. (1961) Pul monary function in asbestosfs of the lung. An alveo lar-capillary block syndrome. American Journal of Medicine, 30, 235-242 Elmes, P. C. &. Simpson, M. J. C. (1971) Insulation wor kers in Belfast. 3. Mortality 1940-1966. British Journal of Industrial Medicine, 23. 226-236 l-Iammond, E. C. (1966) Smoking in relation to the death rates of one million men and women. In: Epidemio logical Study of Cancer and Other Chronic Diseases, Bethesda, National Cancer Institute Monograph 19, pp. 127-204 Hammond, E. C., Selikoff, I. J. & Churg, J. (1965) Neo plasia among insulation workers in the United States with special reference to intra-abdominal neoplasia. Annals of the New York Academy of Sciences, 132, 519-525 . Kannerstcin, M. 3c Churg, J. (1972) Pathology of car cinoma of the lung associated svith asbestos exposure. Cancer, 30,-14-21 Selikoff, I. J., Churg, J. & Hammond, E. C. (1964) Asbestos exposure and ncopiasia. Journal of the Ameri can Medical Association, 183, 22-26 Selikoff, I. J., Churs, J. Sc Hammond, E. C. (1965a) Relation between exposure to asbestos and thelioma. New EnglandJournal of.Medicine, 272, 560 565 Selikoff, I. J., Churg, J. Sc Hammond, E. C. (1965b) 're occurrence of asbestosis among insulation worker: the United States. Annals of the New York Academy o Sciences, 132, 139-155 Selikoff, I. J., Hammond, E. C. & Churg, J. (1963) Asbes tos exposure, smoking and neoplasia. Journal of d:T American Medical Association, 204, 106-112 Selikoff, I. J., Hammond, E. C. <L Churg, J. (1970) Men tality experiences of asbestos insulation workers;. - Shapiro, H. A., ed.. Pneumoconiosis. Proceedings . the International Conference, Johannesburg, Cav Town, Oxford University Press, pp. 130-136 Selikoff, I. J., Hammond, E. C. u Churg, J. (1972; O' cinogtnicity of amosite asbestos. Archies: of '.< vironmer.tal Health {Chicago) (in press) Smith, KL. W. (1965) Trends in the health of the asb-.-tt; worker. Annals of the New York Academy of Scivr.c-: 132, 635-690