Document mmobJpRLkyoDOw1vVdMqBo4bg

FILE NAME: Talc (TALC) DATE: 1982 June DOC#: TALC099 DOCUMENT DESCRIPTION: Medical Journal Article - The Mortality Experience of Upstate NY Talc Workers From (613D 998-3280 Order # 05543491DP04581959 Mon 3un 13 20:50:18 2005 Page 2 of 6 The Mortality Experience of Upstate ? New York Talc Workers All All W. T . Stille, D.P.H., and Irving R. Tabershaw , M.D. D eath for a 31-year period (1948 to 1978) were ana lyzed In a historical prospective cohort study o f 655 white male talc workers. Death rates from all causes, from cancer o f the respiratory system , and from nonmaUgnant respira to ry disease were n o t significantly different from those o f th e U.S. w hite male population. However, significant differences fo r these causes o f death were found among workers who had previous occupational histories. An analysis o f the latency periods o f die observed lung cancer suggests tita t exposure to an etfologlc agent during previous work experience m ay play a role In die development o f lung cancer. T alc pneumoconiosis was first associated with talc mining and milling by Thorel in 18961 and since that time this association has been repeatedly documented in studies from various geographic sites, However, there are a number o f contradictory reports regarding disability from taic pneumoconiosis and its effect on working-capaeity artd respiratory function. The contradiction in these reports has been ascribed to the differences in the mineral compo sition of the various talcs. Kleinfeld et al,2,5 in a series of reports, confirmed the increased Incidence o f talc pneu moconiosis and noted that among die talc workers who were 60 to 79 years o f age and had 15 or more years of exposure the incidence of lung cancer was four times that among the general population. Their follow-up study repeated these findings b u t noted that the observed deaths from 1945 to 1959 dropped to approach the expected mortality between 1960 and 1969. The authors attributed this improved mortality experience to better environ mental dust control.3 Brown and Wagoner,4 using a similar cohort front the same population base; l.e,, taic miners and millers In upstate New York, also found a statistically significant increase in the occurrence o f bronchogenic cancer. The average latency for these cancer deaths was 20 years. They noted that data on several variables, particularly cigarette smoking and previous work history, were not available. Thus, from the - work o f Kleinfeld et al2'3 and Brown and Wagoner, it From Taberjhaw Occupational Medicine Associates, 6110 Ex ecutive Blvd., Rockville, MD 20852. Dr. Stille was formerly Senior Epidemiologist, Dr, Tabcrshavf'ls Senior Occupational Physician. appeared that talc mining in upstate New York is associated with an increase in lung cancer. Selevan et ai,s in studying workers in Vermont talc mines and mills, showed an excess o f deaths due to nonmalignant respiratory disease among millers and an excess o f lung cancer mortality among miners. The lung cancer rate was similar to that found by Brown and Wagoner in their cohort of New York State workers. Selevan et al concluded, however, that agents other than talc, "either alone or in combination with talc dust, affect mine workers. The possible role of ration daughter exposures for this cancer mortality risk cannot be eliminated.1'3 The present study Includes all workers In one talc mine and mill (TMX) in upper New York State who were em ployed between 1948, when operations began, and Dec 31, 1977. M aterials and M ethods The work force consisted of 744 persons'employed between Jan 1 ,1948 and Dec 31,1977. Thirty-five women, primarily employed in office and administrative jobs, were not included because of the small numbers and a lack of exposure to talc. Nearly all employees were white and records indicate that to date all decedents have been white. The vital status o f 36 (5.1%) men remained unresolved after Social Security and New York State Motor Vehicle Bureau records were searched. The maximum cohort available for analysis consisted of 708 white men, 713 of whom were known to be dead by Dec. t 2 , 1978, which was established as the vital status cutoff date. Demographic d ata' (microfilmed from employment records) consisted o f the following: worker identifiers; sex; date of birth, employment and death; cause of death and Information on prior employers. Data on cigarette smoking were not available. The occupational history in cluded dates of work, lob codes, and work location codes. Records were updated for the dates and causes of death as such Information became available. Death certificates were coded according to the eighth revision of the International Classification of Diseases by a trained nosologist. Compari sons o f observed deaths by age and year of death with those expected in the U.S. white male population were done by means o f the modified iife-tabie methods described by Mcnson.9 All standard mortality ratios (SMRs) were com puted by this program. Killy-three workers whose records lacked dates of birth or other significant data had to be All Ail All Ext r.dropi Lcoho samp 48C Mortality am) Talc Workera/Stille & Tafrarahaw m Journ W E S 0 0 1 126 From (613) 998-3280 Order # 05543491DP04581959 Mon Jun 13 20:50:18 2005 Page 3 of 6 Tibia 1 - 0 (waved and Expected Numbersof Oastbs and SMRs Amoig TMXWork {665White Mats! With 11,350 Yaats at Risk). Causa of Cstii ICC,* Ito Revision Dtoll>1 Observed Expected SMR ork Is associated a!,5 In studying Ids, showed an ipiratory disease ancer mortality . similar to that ort o f New York , however, that ombination with tie role of radon !y risk cannot be In one talc mine e who were embcgan, and Dee rsons'employed Irty-fivc women, rative jobs, were rs and a lack of were while and have been white, ined unresolved s Motor Vehicle aximum cohort lite men, 113 of 1978, which was m employment rker Identifiers; 5 cause o f death ata on cigarette lonal history in : location codes. iuses o f death as certificates were die International rfogist. Comparideatii with those >n were done by ds described by MRs) were corn 's whose records data had to be All deaths 113 107 106 Tuberculosis (010-019)1 3 .7 414 Ail cancer 1140-209) 26 20b 122 Digestive system (150-159) e 5.7 105 Esophagus (160) i 5 216 Stomach (151) i 1.1 92 liver 1155-156) 2 .4 511 Pancreas (157) 1 1.1 89 Respiratory system (150-163) 11 6-8 163 Lung (182) 10 6.4 167 Prostate (186) 1 U 79 Kidney (189) 1 A 185 Breda (101-192) 1 ' .7 149 Lymphosarcoma (200) 1 j 206 Hodgkin's distate (201) 2 A 631 Leukemia (204-207) 2 A 229 AUlymphopoietic cancer (200-209) S 2.1 230 All circulatory system (390-488) 48 53.7 89 Myocardial Infarction (410-4131 33 38.3 68 Cerebrovascular disease (430-438) 5 7.1 71 All respiratory disease (460-519) to 8.1 164 Pneumonia (430-488) 4 2.2 182 Emphysema (492) All digestive system (520-677) 2 1,7 121 3 S.3 57 External causes of death (800-8981 11 12.6 87 Motor vehicle accidents (810-827) 4 4.3 93 Suicide (950-959) 2 28 70 * ICO Indicates International' Classification of Diseases t Humbert referto ICO coda dropped from toe life-table analyses. Thus, toe study i cohort consisted of 625 white males, a relatively small | r sample size and therefore a limitation on toe study. R esu lts Table t shows the numb of observed and expected l deaths in the study cohort o f 655 white mate talc workers. Although the SMRs are elevated, the numbers of deaths from all causes, from cancer of toe respiratory tract and lung from nonmallgnant respiratory disease, or from other causes of death are n o t significantly different from those occurring In toe U5>. white male population. Numbers of deaths from circulatory diseases, digestive tract diseases, and accidental and external causes are lower, but no t slg* : nlficantly so. The 5MR o f 106 for all causes of death has 9$% confidence limits from 87 to 128, thus th total talc worker population has an SMR within toe usual range of -variation from the national data. Since people who are chronically 111 or disabled are usually unemployable and have higher death rates, healthy working populations ^ typically have SMRs below 100. The SMRs in an occupatlonal group wito values over 100 Indicate either possible (le & Taberehaw M ' Journal of Occupational Medifcina/Vol, 24, No. 6/June 1982 Tibia 2 - Observed and Expected Numbers et Osati and SMRs Among Tdfc Workers With Known Prior Employment Bifore Work at the TMX (540White MalesWith - 8,188 Years at Risili. - j Caute of Death ICO * 8th Revhlan - - Deathi__________ Observed Expected SMR All deaths 90 Tuberculosis (010-019)1- 3 AH cancer (140-209) 22 Digestive system (160-159) 6 Enophagtu(IBQ) 1 Stomach (ISO 1 Liver (165-156) 2 Pancreas (157) 1 Respiratory system (180-163) 8 Luna(182) a Kidney (189) 1 Brain (191-192) 1 Lymphosarcoma (200) ! Hodgkin's disease (201) 2 Leukemia (204-207) 2 AH lymphopoietic cancer (200-209) 5 All circulatory system (388-468) 36 MyocardiBl Infarction (410-413) 27 Cerebrovascular disease (430-439) 3 All respiratory disease (480-519) 9 Pneumonia (480-486) 3 Emphysema (492) 2 Ail digestive system (520-577) 1 External causes of death (900-698) 11 Motor vehicle accidents (810-827) 4 Suicide (950-953) 2 60.6 149* .4 680* 11.6 182 3D 201 3 382 .0 180 .2 1,013* .6 184 4.0 228* 3.7 214 .3 325 A 206 A 325 2 349 .5 391 U 374* 29.0 129 20.3 133 3.1 97 2.9 307* 1.0 274 .7 278 35 28 9.B 112 3.4 118 2.2 92 * Indicatesstatistical significance at toe 1%level t Numbers in parenthesesrefer to ICO codas i Indicates statistical significance at the 8%level . work place hazards or possible confounding with rtonoccupatfanaJ exposures. One such noreoccupatlonal exposure Is smoking, and it is noted that the SMR of 163 for respiratory cancer in this population Is consistent with a smoking effect While toe. actual smoking experience of this population is not known, i insight into It is gained from knowing the smoking patterns o f current TMX workers, many of whom are Included In the study cohort. Among 151 workers given a medical examination at TMX In 1979, 69 (46%) were current smokers, 38 {2556) were past smokers and 44 (2996) were nonsmokers. The percentage o f current smokers Is higher than in the general population, b u t Is consistent with pat terns seen in the mining industry. The SWIR for all causes of death, 106, may also reflect In part inclusion in the cohort o f everyone ever employed during toe study period, regardless of their length of employment. Universal cohort membership generally tends to Increase SMRs, since short-term workers often have, a greater mortality experience than long-term employees. Most study designs exclude workers with less than one year of working experience. The practice was not followed r 481 W ES001127 From (613) 998-3280 Order # 05S43491DP04581959 Mo Jun 13 20:50:18 2005 Page 4 of 6 Table 3 - Observed and Expected Numbars of Deitbs md SMRs AmongTele Worten With No Known Work Prior to TMX Emptoyrnem<115Whits MilesWHfc2,184 Vein of Risk). Causa of Death ICO, 8th Revision -------KSSE-----Observad Expected SMR All deaths 23 All cancar (14D-2091T 3 Lung (162) 2 Prostate (18S) 1 All circulatory system (390-459) 12 Myocardial infarction (419-4131 G Cerebrovascular disease (430-438) 2 All respiratory diseases (469-519) 1 Pneumonia (489-486) 1 All digestive system (520-67?) 2 45.9 50* 9 4 33 2.6 76 .6 120 268 48 T8J0 33 4.0 SO 3.2 31 1.1 90 U 112 * Indicates statistical significance at the 1%level; note that this SMR is lower then expected with respect to the U.S. national t Numbers In parentheses refer to ICD codes this study in order to maximize cohort size. The use o f the U.S. white male population as the exter nal comparative group does not necessarily' describe the TMX mortality experience, since the mortality experience of the U.S. white male population is likely to vary substan tially from that o f this cohort. For example, the New York State cancer death rate is 199.24 compared to the national average o f 174.04, approximately a 13# difference.7 Since exposures prior to employment at the TMX could have included carcinogens and other substances hazardous > to the lungs, the results have been analyzed to investigate this possibility. Tabic 2 shows the observed and expected1 numbers o f deaths and the SMRs for all workers whose job histories show prior work before employment a lth e TMX. Deaths for all causes are significantly Increased, as are deaths for a number of cancers and for nonmalignant respiratory diseases. Since the cancers and lung diseases typically have long latencies, the possibility exists that exposures prior to work a t the TMX were responsible for a t least some of these diseases. ' Tabie 3 shows the observed and expected numbers of deaths for workers with no known employment prior to the TMX. Mortalities for alt disease categories do not differ significantly from those for the comparable U.S. popula tion. In fact, the SMRs from all cancers and ail respiratory tract diseases {33 and 31, respectively) were among the Jowest oFall the mortalities. It should be noted that workers whose employment records were devoid o f work history data were Judged as having had no known prior employ m ent and thus were included in Table 3. If some of these workers did indeed have work experience before employ m ent a t the TMX, the effect would be to increase the SMRs. However, even with this possible compromising effect, Table 3 shows a remarkably healthy work force, with most mortalities at S0% or less o f those o f the comparable U.S. population. As noted earlier, however, the small sample size limits the conclusions that can be drawn from this study. Five o f the 12 lung cancer cases, which are included in Table S, were substantially older than the median age of 482 Tabla 4 - Ageat death and Segments of Work History ofVKcrkert at the TMXWitb Broeebegenic Carcinoma. Age at Case No./Deatb, yr Pre-TMX Employment Yean nr Days Employed ttin TMX Post-TMX Employment 1/39 2/49 3/53 4/53 5/54 6/55 . 7/59 8/62 9/84 10/79 10 4 7 8 17 6 8 47 days 27 13 56 days 23 12 3 21 21 3 13 16 345 days 24 26 17 1 25 17 days 21 39 8 days 22 the entire cohort at age of hire. For those workers wilh TMX-oniy employment, 63% were under 25 when hired, contrasted with 35% for those with previous experience. The difference in age at hire between the two cohorts may partially explain die deficit in disease mortality among TMX-only workers and the lung cancer experience of those with other work history. Discussion Workers with previous jobs before employment at the TMX were found to have high mortalities, as shown In Table 2. Workers with no known work history before em ployment at the TMX were found to have much lower mortalities (Table 3). Thus, It would seem th at exposures to health hazards occurred in these earlier jobs and that these exposures are causally related to the higher rates o f death. Obviously, mines and mining operations are not uniform with respect to the occurrence o f health hazards and, as previously stated, 5elevan et al9 have described increased rates o f lung cancer in some taJc miners exposed to radon daughters. No significant radon daughter exposures oc curred at the TMX.*'9 The low mortalities given In Table 3 are in agreement with the absence o f radon daughter and other related hazardous exposures at the TMX. However, mortality studies such as ours lend themselves to various conclusions. Lacking data on smoking and other personal risk factors, we sought to obtain confirmation for our Interpretations by an Independent analysis of lung cancer latencies. Data for an analysis o f the lung cancer latencies are given In Tabic 4. For each of the ten cases of lung cancer, the employee's years of occupational history are given as the following: pre-TMX employment (employment was deter- TTTTTTT S ID YEARS EMPLOYED AT. TMX Fig. 1 - Numiiw of lung cancer cam by yean of employmentet TMX. Mortality and Talc Warkers/Stilte & Tabersbaw Taktes i Eaten N <10 1Q-K 15-1 20-24 26-21 30-34 36-30 40-44 46-40 50-54 55-59 >60 Total Latem i * Frr mined ai f: TMX; at | diagnosis I'Table 4, noted. T , 47 after ; he died i \ were 29 ' years po ^`.The sect ; date. His | TMX, Z To ex |. colum n. employe . ure. The are inclu the first Inverse d less OCM : response, , agreemen ; TMX em , having fui Know! 1determin >processes be usefu causal ex W E5001128 From (613) 998-3280 'W Order # 05543491DP04581959 Table 8 - latencies of Bronchogenic Carcinomas in TMX Workers MX merit > with hired, fence, tsmay among f those a t the >Wn in ire cm- lower urcsto it these death, iniform and, as creased i radon res ocTable 3 ter and awever, various rersonai for our ; cancer re given cer, the i as the s detef- 25 rnwntst Latency, *yr <10 10-14 15-19 20 24 25-29 i 30-34 35-39 40-44 45-49 50-54 55-59 >sa Total Latency Bronchogenic Carcinomas in TMX Wqrkers Whose Exposure is Hypothesized to Bogin On Employment at the TMX At Age 18, Prior to TMX Employment 1 1 2 5 1 3 T 1 - 4 - 3 - 1 ' - - - - - 1 12 12 199 38.2 * Fromensat of exposures diagnosis or death mined arbitrarily to have begun at age 18); tenure at the TMX; and the period from TMX employment until the diagnosis of lung cancer or death. Although not included in Table 4, two additional persons with lung cancer should be noted. The first was surgically treated for lung cancer at age 47 after 137 days o f employment at the TMX. A t age 57 he died of pneumonia. His significant occupational periods were 29 years pre-TMX, i 37 days of TMX tenure, and zero years post-TMX to the time o f diagnosis of lung cancer. The second Individual died of lung cancer after the cutoff p date. His significant occupational periods were 15 years preTMX, 23 years of TMX tenure, and six years post-TMX. To explore the nature o f the dose-response relationship, column 3 in Table 4, which lists numbers of years or days employed at tire TMX, is graphically represented in the Fig : ure. Tire two additional cases of lung cancer just described ; are-included. The clustering of nine o f the 12 cases within the first five years of employment at the TMX shows an Inverse dose response, i.e., higher risks of lung cancer with . less occupational exposure to talc. The lack of 4 dose ; response, In terms of years of exposure at the TMX, is in . agreement with our findings that workers with "exclusive" : TMX employment seem to be at no considerable risk of . having lung cancer develop. . Knowledge o f the latency period of a disease permits ' determination o f the probable time at which pathological processes began. Thus, the analysis o f disease latencies tan be useful in attributing or refilling evidence as to when causal exposures occurred. Armenian and Liiienfeld10 have Mon Jun 13 20:50:18 2005 Page 5 of 6 reviewed the latency periods o f a number o f cancers, 'hoy present evidence that leukemias in different populations and under different conditions have median latency periods and latency variabilities that re quite similar. They also re viewed data (hat indicate (tut the intensity of exposure has very little effect on latencies of other neoplasms. Hcncc, wc hypothesize that if occupational talc exposure results in pulmonary exposure to a carcinogen, then the distribution o f TMX workers lung cancer latencies should be similar to those reported for lung cancer in other studies. However, If such exposure is n o t carcinogenic to the lung, then the distribution o f TMX lung cancer latencies would be dis similar to lung cancer latencies reported elsewhere. To test the hypothesis that carcinogenic exposure began at TMX, the median lung cancer latency o f 36.5 years, as calculated by Armenian and Liiienfeld,10 was compared to the latency period from the onset o f TMX employment to the diagnosis o f each lung cancer. As can be seen from Column 2 in Table 5, the average latency period of 19.9 years for lung cancer among the TMX workers is almost half the 36.5 years noted above. The Chi Square goodness of fit test provides a quantitative basis for rejecting the hypothesis at the 5% level o f statistical significance. if the occupational talc exposures to the TMX are not carcinogenic, then the question shifts to what other ex posures occurred that may have caused tire lung cancers. Since each or the subjects with lung cancer included in Table 4 had between eight and 39 years o f possible occupa tional exposure from the time o f his 18th birthday to em ployment a t the TMX, this prior period o f exposure may be related to subsequent cancers. To test this possibility, we hypothesized th at the causal exposure for the lung cancer occurred before employment at the TMX. The data in the last column o f Table 5 reveal a much more reasonable relationship to the lung cancers reported by Armenian and Liiienfeld. This second hypothesis, that the causal exposure began before TMX employment, yields an average latency period o f 3 8 .2 -years, which is only slightly longer than the Armenian and Liiienfeld median of 36.5 years. The Chi Square goodness o f fit Is almost, zero, indicating a very close agreement between the lung cancer worker latency periods and those o f talc workers when exposures prior to TMX employment are hypothesized as .carcinogenic. Hence, the initiation o f cigarette smoking in the late teenage years, World War II exposures, excessive exposures to mineral dust, in the past, diet and nutritional status, absence of complete work history, preexisting diseases, and oilier unknown etiologic agents could contribute to the etiology o f the observed (ung cancers, while exposures at TMX seem lobe noncarcinogenic. I t should be noted that nine of the 12 employees .with lung cancer worked underground In TMX, which suggests that their other Jobs also were likely to have been underground. Review o f the death certificates tends to confirm the observation that these men worked in occupa tions known for their dusty environment. The employment policy o f TMX to preferentially hire experienced workers appears to have resulted in the confounding o f prior hazardous exposures with employment at 1 MX. Sum m ary and Conclusions A historical prospective mortality study of workers at an upstate New York talc mine demonstrated elevated mortali- sentan ' Journal of Occupational Medieme/Vol. 24, No. Q/June 1882 483 WES001129 From C613) 998-3280 Order # 05S43491DP04S81959 Mon 3un 13 20:50:18 2005 Page 6 of 6 ties but no significant Increases In the numbers of deaths from lung cancer, from nonmallgnant respiratory disease, and from all causes. However, workers with exposures in other jobs prior to work at the TMX were found to have extessive mortality from lung cancer .and from nonmalig nant respiratory tract disease. Exposures in jobs held prior to TMX employment also were Implicated In an Indepen dent analysts o f lung cancer latencies. A cknowledgm ent Technical and editorial contributions were made to this com munication by Maureen A, Vogt and M. fames Sharpe. Juliet A. Foster, Pamela J, Trotter, and Deborah Rogers aided In data col lection and processing. R eferences 1. T hortl Cr Die spcckstelnlunge (ein beitragzur pathaloglschen anatomic der straublungen). Beltr Patti A nt 28:85-101,1896. 2. Klelnfcld M, Mosslte J, Kooyman O, at at: Mortality among talc miners and milters In New York Stale. Arch Environ Hewn, 14:663-667,1967. 3. Klelnfeld M, Messlte ), Zaki MH: Mortality experiences among talc workers: A follow-up study./OM 16:346-349,1974. 4. Brown DP, Wagoner JK: Occupational exposure to talc con. taJnlng asbestos: III. Retrospective cohort study o f mortality, u.S Dept, o f HEW, (NIOSH) Pub. 80-115:31 -33,1980. ' 5. Selevan $G, Dement |M , Wagoner JK, et al: Morality, patterns among miners and millers of non-asbestlforra talc: Pr. Ilmtnary rsport.7 Environ Pathol Toxicol 2:273-284,1979. 6. Monson RR: Analysis o f relative survival and proportional mortality. Comp Biol Res 7:325-332,1974. 7. Mason T |, McKay FW: U.S. Cancer Mortality by County- 19SC-1969. HEW (NIH) Pub. 74-615:685,1974. 8. Health and Safety Report, Health and Safety Spot Inspec tion (Radiation), American Mine, Gouvarneur Talc Co,, Inc., Sal- mat, St. Lawrence County. N.Y., Feb 28,1973. U.S. Department of Interior, Bureau of Mines Health and Safety Activity. 9. Health and Radiation Inspection Report, Gouvemeur Talc Co., Mine No. t and Mill, Balmat, N.Y., March 24, 1976. U.S, Department of Interior, Mining Enforcement and Safety Adminis tration. ' 10. Armenian HK, Lilienfeld AM: The distribution of Incuba tion nedodsof neoplastic diseases. A m ; Epldemol 99:92-100,1974. S eiet froir Thonu D epart r i Cl i Gr t. Wi kI De Jar Ce Jan Cln Chi Cin Shaping Moral Neuters Common to the various '*moral education" approaches offered today is the Idea that there is nothing right or wrong and that to teach any substantive ethical percept or idea is to "indoctrinate" students. In one approach, students are Invited to "clarify" their own values; the focus is on "rationality,'"'"creativity," "autonomy" and "process." In another approach, students are asked to dwell on ethical dilemmas such as when lying can be Justified, or on public policy questions such as the morality of racial quotas. The problems with these approaches are obvious to anyone raised in the basic morality. In the values approach, no one need fear a teacher's Judgment if his values are wrong, because there are no wrong values (other than, o f course, the value of opposing "values" education). Upon learning that all choices are equally valid, a student will be educated into skepticism about morality, fertile soil for wrong action.. . . Today's "moral education" fails in the mast critical way because it addresses the intellect but-n o t the emotions. No one is taught to dislike and certainly not to hate anything. If young Americans are shaped, they are unfortunately shaped to be moral neuters.' The inculcation of the basic morality in the minds and hearts o f youngpeopie, as given In American education since the T7th Century', has for the most part ceased to be in the latter part o f the 20th. - From "Teaching Morality In Public Schools" by Terry Eastland In The WallStreetJournal, Feb. 22,1982. Em T. / Neu J,' Geo A lkt Ep[< \ Phil) Pitts Ergc E. R. Now Suzal Roch l l' Expc M. P. Linde 484 Mortality and Talc Worlcsrs/Stills St Tebersbaw iJournal of WES001130