U MASS/ AMHERST 312066 0374 7038 9 ^VERWAENT DOCUtKN COLLECTION SEP 2 2 1986 V.' .crsity ot (Viassachuser depository Copy Ar. Investigation into Pancreatic Cancer Mortalitv in Peabodv, Massachusetts Massachusetts Department of Public Health June, 1984 Table of Contents Page Ir.t-oductior. 1 Section 1. Car.ce" of the Parc-eas 2 1.1 Risk Factors 3 1.2 Human Studies 3 1.3 Occupational Health Studies 6 1.4 Animal Studies 7 Section 2. Methodology 7 2.1 Selection of Cases 9 Section 3. Survey Results 9 3.1 Residential Clustering 14 Section 4. Environmental Survey and the Potential for Exposure 18 4.1 Environmental Testing 19 4.1. a Tests of Peabody Drinking Water 20 U.l.b Testing of Specific Sites 21 Section 5. Conclusions and Recommendations 27 Bibliography 31 i 1 0 » Appendices Page A. Peabody Mortality Statistics for Cancer of the Pancreas A.l 3. The Leather Industry B.l C. Statistical Evaluation of Clustering C.l D. Industrial History D.l E. Peabcdv Water Supply System E.l F. Peabody Drinking Wate1* Test Results F.l Z. Dumpsites in Peabody G.l H. Eastman Gelatine H.l I. Salem Acres 1.1 J. Pierpont Street J.l K. Strcngwater Brook K.l L. Peabody and OCR Landfills L.l ii List of Tables Page 1. Tvpes if Information Obtair.ed During Interviewing 2. Presence >f Personal Risk Factors ir. Peabodv Pancreatic Car.ce- Cases Compared to Other Studies 13 A.l Standardized Mortality Ratios for Cancer of the Pancreas in Peabody, Massachusetts A. 2 3.1 Outline of Major Steps in the Manufacture of Leather B.2 3.2 Chemicals Associated with Various Operations ir. the Tannery Process B.3 C.l Place of Residence by Census Tract in 1950 of Peabody Pancreatic Cancer Cases C.2 C.2 Place of Residence bv Census Tract in I960 of Peaboay Panc-eatic Cancer Cases C.3 C. 3 Place of Residence bv Census Tract in. 1970 of Peabody Pancreatic Cancer Cases C.4 D. l Peabody Industries in 1930 and 1983 D.2 F.l Example of Water Supply Analysis F.2 F.2 City of Peabody Water Testing Results - T-ihalomethane Concentrations F.3 F.3 City of Peabody Water Testing - State Pu»-geable 0-ganics Testing Results F.4 F.4 Inorganic Chemical Concentrations F.5 F. 5 City of Peabody Water Testing Results F.6 G. l Peabody, Massachusetts Dumpsites G.3 H. l Eastman Gelatine Lagoon Water Testing by DEQE H.4 J.l Soil Analysis by Skinner & Sherman of Pierpont Park, Peabody, Massachusetts J. 2 J. 2 Soil Testing Results of Goldberg & Zoino at Pierpont Park Peabody, Massachusetts J. 4 K.l Strongwater Brook Sampling Results, Peabody, Massachusetts K.2 iii 0 List of Figures Page 1. Age Distribution of the Peabody Pancreatic Career Cases at the Time of Death 10 2. Place of Residence of the Peabodv Pancreatic Cancer Cases in 1950 by Census Tract 15 2. Place of Residence of the Peabodv Pancreatic Car:ce'- Cases in I960 bv Census T^act 16 4. Place of Residence of the Peabody Pancreatic Carce" Cases in 1970 by Census T-act 17 A.l Place of Residence of the Peabody Pancreatic Cancer Cases at Time of Death A. 4 D.l Location of Tanneries in Peabody, Massachusetts in the year 1930 D.3 D. 2 Location of Tanneries in Peabody, Massachusetts in the year 1978 D.4 E. l Peabody, Massachusetts Downtown Water Supply System E.2 G. l Peabody, Massachusetts Dumpsite Locations G.2 H. l Eastman Gelatine's Waste Lagoons H.2 iv • • Acknowledgements The Department expresses its gratitude to the following students of the 3oston College School of Social Work who assisted us greatly in car*"ving out this study: Andrea Cohen, Liz Hirsh, Alisa MacFail, and Jerry Marx. Their ha'-d wo""k, commitment, and patience are greatly appreciated. We would also like to thank Professor Richard Bolan for his interest and support. Manv Peabodv officials and -eSicier.ts representing various organizations provided information vital to this study: the Mayor's Office, Peabody Water Department, Department of Human Services, Community Development Department, Coalition for a Safer Peabodv, Peabody Post Office, and the Office of City Clerk. The Department of Environmental Quality Engineering, particularly the Woburn regional office, supplied us with environmental data on Peabody, and its help is greatlv appreciated. The names of individuals who supplied us with information follow the bibliography. Finally, we thank the Division of Health Statistics and Research of the Department of Public Health for providing the raw data on pancreatic cancer mortality in Peabody. V Executive Summary The Massachusetts Depa-tmer.t of Public Health ( DPH) investigated ar. elevation ir. mortality from career of the pancreas ir. Peabody for the period 1974 to 1978 and ir. one census tract (CT 2106) within Peabody for the period 1974 to 1982. The goals of the investigation were to determine whethe- mortality f>-om pancreatic cancer in Peabody and CT 2106 was related to environmental contamination in Peabody and to identify risk factors associated with the disease. Informants for those Peabody residents who died of pancreatic cancer between 1974 and 1982 were interviewed using a questionnaire developed by the DPH. Forty-one cases were included in the studv. Two informants refused to participate in the study and four could not be -eached by phone or letter. The -espor.se -ate of the study was 87 percent. Information on residential history, occupational history, medical history , personal -isk factors (such as smoking nabits), and exposure to environmental contaminants was collected. In addition, existing environmental data for Peabody, including air, wate-, and soil testing -esults, were collected to identify potential environmental -isk factors . The -esidential history information collected during interviews was analyzed to determine if the cases were concentrated in any particular area of Peabody and could have been exposed to environmental pollutants from living in the neighborhood. Since most cancers have a latency period of 20 to 40 years after initial exposure to the carcinogenic agent(s), the location of -esidences of the cases for the years 1950, 1 960, and 1970 were examined to investigate past exposures to environmental contaminants. vi • j No concert -atior. of cases in any particular area of Peabodv was seer, ir 1950 or 1960. Ir fact, ir 1950, at least ore-half of the cases lived outside :f Peabocv, and ir 1960, ore-quarter of the cases were rot Peaoodv -esi- derts. Ir 1970, more parcreatic cancer cases lived ir CT 2106 thar would be expected giver the distributior of the Peabody population. Therefore, the apparer.t clustering in CT 2106 is a relatively recent phenomena. One possible explanation for this firding is that older people have ter.ded to move into CT 2106 because of its -esidertial characteristic and its proximity to Peabody certer . While a statistically significant excess of deaths f>-om parcreatic career occur-ed over the five-year period 1974 to 1978, the parcreatic career mortality expe-ier.ee of Peabody for the two periods of 1969 to 1973 and 1979 to 1982 was as expected, based upon statewide rates. It appears that mortality from parcreatic cancer in Peabody has returred to expected levels. The potertial for exposure of the cases to air, water, and soil contami- rarts fourd at various locations withir Peabodv was assessed. It was concluded that actual exposure of the cases to environmental contaminants at these locations was highly unlikely. Because environmental data on Peabody exists back only to the early 1970s, exposures may have occurred in the 1940s, 50s, or 60s which were rot possible to document. In summary, the DPH concludes that pancreatic cancer mortality in Peabody is probably not related to living in Peabody. That is, no common exposure to environmental pollutants was found among the pancreatic cancer cases. One recommendation, however, is made in the report. The recommendation, although not related to the increase in pancreatic cancer mortality, was made because of citizer concern and information brought to DPH's attention during the investigation. DPH feels further analyses should vii • be conducted or. soil samples at Pie»-por.t Street Park. Preliminary soil ar.alvses pe^fo'-med ir 1983 "evealed high levels of lead ar.d chromium, -hil j-er. playirg ir. the pa»-k #mav be exposed to these toxic metals by ingestion and/or inhalation of soil panticles or by skin contact with the soil. viii ■ • • introduction In February 1983, the Department of Public Health (DPH), as part of its •-outir.e function of examining death and disease •-ates in cities and towns acoss the Commonwealth, began its initial review of health statistics for Peabody. In Julv, the DPH was contacted by three independent groups concerned about pancreatic cancer mortality in Peabodv. Those groups were: the office of Representative Theodore C. Speliotis, the Coalition for a Safer Peabody, and the office of Senator F-ede-ick Be^-v. P-eliminarv analysis of the period 1974 to 1982 showed a higher number of deaths f-om pancreatic cance*- for Peabody as a whole and significantly more aeaths from pancreatic cancer in one census tract (CT 2106) in South Peabody than were expected on a statewide basis. Appendix A contains these statistics. This -eport is a summary of the subsequent investigation of pancreatic cancer deaths in Peabody carried out by the DPH. Section one of this report is a general discussion of pancreatic cancer including -isk factors associated with it. Section two describes the survey car"-ied out with the relatives of those Peabody residents who died of pancreatic cancer between 1974 and 1982. Section three contains the results of the questionnaire survey. Section four is a summary of existing environ- mental data in Peabody as they relate to the potential for exposure of Peabody pancreatic cancer cases. Section five contains the conclusions and recommendations of the Department. Appendices A through L follow the report. 1 • Section 1. Career of the Pancreas The parceas , a glard located behind the stomach, produces substarces, such as i r. s u 1 i r. , that -egulate the level of sugar in blood and that aid digestion. Jance- of the pancreas is the fourth leading cause of cancer deaths in the United States. It is exceeded onlv by colon, lung, and breast car.ee- (American Cancer Society, 1982). However, pancreatic cancer deaths account for only five pencent of the total number of cancer deaths (American Cancer Societv, 1980). For 1977, the U.S. age-adjusted death »-ates for pancreatic careen were: 10.9 deaths for every 100,000 white males; 6.9 deaths for every 100,000 white females; 13.3 deaths for every 100,000 nonwhite males, and; 8.5 deaths for everv 100,000 nonwhite females (National Cance«- Institute, 1982). Panceatic cancer, like many other types of careen, is mone common in olden people. It is quite nane in people unden 40 yeans of age. In Connecticut, which has had a Tumon Registny since 1941, the incidence nate of pancreatic cancer" in people 40 to 44 yeans of age is two cases of pancneatic cancc- fon eveny 100,000 people. Fon people between 80 and 84 veans of age, 100 cases occun annually fon eveny 100,000 people (MacMahon , 1982) . The annual incidence rate of parcreatic careen has increased slowly over the past 30 years. One partial explanation for this increase is the improved diagr.osis of pancneatic cancer. By the time it is diagnosed, the disease has usually pnogressed to advanced stages. Fon this reason, a low survival rate exists for pancneatic cancen . Most types of cancer, including pancneatic cancen, ane chanacterized by a long latency period. This means that typically 20 to 40 yeans elapse from the time of exposure to the cancer-causing substance on event and the mani- festation of the disease. In this study, the latency peniod or the critical time with -espect to exposure is, -oughly, the years between 1940 and 1960. 2 * 1.1 Risk Factors Marv factors are thought to contribute to the occur-erce of carce-. Epi- demiologic studies (studies of human populations) and experimental animal studies are conducted to identify factors which may be associated with certain diseases. These factors, called -isk factors, may increase a person's chances of developing cancer. A -eview of the health literature on pancreatic cancer was conducted to identifv -isk factors associated with pancreatic cance- . None of the factors discussed below, however, accounts for more than about thirty percent of the incidence of pancreatic car.cer (Lin et al , 1981). It may be that a combination of these factors increases a person's chances of developing cancer. Most pancreatic cancers are still unexplained. 1 . 2 Human Studies Smoking Several epidemiologic studies have shown that pancreatic car.cer is about twice as common in heavy smokers as in nonsmokers (MacMahon, 1982; MacMahon et al, 1981; Wynder, 1975; Wyr.der et al, 1973; Moolgavkar et al , 1981). Research- ers think that carcinogens in the tobacco are deposited in the pancreas by the bloodstream or, un.de>- certain conditions, by liver-produced bile which flows back to the pancreas (Wynder, 1973). Coffee A recent case-control study of 369 patients with pancreatic cancer and 644 control subjects (with various chronic diseases) has suggested an association between coffee drinking and pancreatic cancer (MacMahon, 1981). People who drank one or two cups per day were twice as likely to have 3 \ • parceatic car.ce" as ror.cof fee drinkers and people who d^ar.k three or more cups pe»- dav we^e about three times more likely. Howeve*-, it is possible that the sor.trols, marv with gastrointestinal diseases, ter.ded to d-irk less coffee tnar. the cases because of thei*- illnesses. Mo*-e studies reed to be dor.e to investigate this potential risk factor. In another hospital -based case-control study, pancreatic cancer cases d-ar.k significantly more decaffeinated coffee than people who served as cor.fols in the study (Lin, 1981). It is possible, however, that the high consumption of decaffeinated coffee •-eflected generally high coffee consumption bv the cases in the past. For this >-eason, caffeine cannot be '•uled out as a possible -isk factor. Alcohol Excessive use of alcohol has been suggested by some as a •-isk factor for pancreatic cancer (American Cancer Society, 1982). However, other researchers have concluded that it is unlikely that alcohol has any role in pancreatic cancer (MacMahon, 1982; Wynder, 1975). Medical History Some studies have shown increased frequencies (although not always statistically significant) of the following medical conditions in pancreatic cancer patients compared to controls: diabetes, gallstones, removal of the gallbladder, chronic inflammation of the pancreas, removal of the ovaries, spontaneous abortions, and uterine tumors (Lin, 1981; Fraumeni, 1975; Wvr.der, 1973). The role that any of these conditions pi ays in the development of pancreatic cancer is not clear at this time. Some researchers think that hormonal imbalance and chronic inflammation, characteristic of some of these conditions, may increase the likelihood of developing cancer (Soloway et al , 1966) . 4 i T--ihalornethar.es Ore studv investigated a potential correlation between carce- mo^talitv ard t-ihalomethar.es (THMs; organic chemicals found in chlorinated water supplies) in a public drinking water system (Ca^lo et al , 1980). Over 4000 cases with five different types of career, ircluding pancreatic cancer, were analyzed (using multiple regression methods) in terms of their source of drinking water ard level of THMs. Investigators found a correlation between THMs and pancreatic cancer in white men. The authors of the report question the significance of this result because no congelation was found between females or nonwhites and THMs. This study, because of its design, could not control for important variables, such as smoking, which are associated with parc-eatic carcer. Asbestos The discovery of very high levels of asbestos in the drinking water of Duluth, Minnesota prompted a long-term surveillance study of cancer rates in Duluth ( Sigurd son , 1 98 1 ) . For over 25 years, taconite wastes containing asbestos fibers were dumped bv a mining company into Lake Superior, the source of Duluth's drinking water. It is still too early to make a complete assessment because sufficient time has not elapsed to accommodate the length of exposure ard a -eaSonabie induction period for the development of cancer. However, a statistically significant excess of pancreatic cancer was noted in the Duluth population companed to residents of Minneapolis and St. Paul. Occupation Another case-control study of 109 cases with pancreatic cancer and 109 controls found that the proportion of male pancreatic cancer patients who had been employed in the dry-cleaning business or in occupations involving close 5 ♦ ♦ • exposu-e to gasoline (for example, service station and garage work) was significantly higher than fo- the male control subjects (Lin et al , 1981). The risk «as increased five-fold among mer who worked in these occupations for mo^e than ten yeans. 1.3 Occupational Health Studies One retrospective mortality study of 639 workers in a plant manufacturing beta-naphthylamine and benzidine showed a higher number of deaths from pancreatic cancer than was expected from statewide statistics (Mancuso, 1967). The number of cases was small, however, and the statistical significance of this finding was not sported. Occupations in which workers mav be exposed to these chemicals include dyeworkers, chemical workers, rubber-tire manufacturing industry workers, manufacturers of coal gas, nickel refiners, copper smelters, and electrolysis workers. A retrospective studv of the causes of death of 19^6 workers in a pigment plant (employed there between 19^0 and 1969) who were exposed to lead and zinc chromates also showed an excess of pancreatic cancer deaths (Sheffet et al, 1982). The number of pancreatic cancer deaths was small and statistically insignificant and the researchers concluded that people exposed to chromates should be observed more closely for the occurrence of pancreatic cancer . Peabody has been called the "Leather City of the World" because of its many leather processing industries. For this reason, the occupational health literature was reviewed to identify risks associated with employment in the 6 ♦ leathe- industry (see Appendix B). Several studies have shown ar. association between emplovmert in the leathe-" indusfv and bladde- car.ce-. I- addition, it is known that many carcinogens are used in leathe- processing and, depending upon the type of job, a worke*- may be exposed to any number of them. 1.4 Animal Studies Several chemicals have produced pancneatic cancen in laboratory animals ( Lor.gr.ecker , 1983; Wyrde-, 1 973 ; Mainz, 1974). Under a -esearch g-ant provided by the National Cancen Institute, Longnecker reviewed experimental animal studies of pancreatic cance- (Longnecker, 1983). He identified sevenal cnemicals that induced par.ceatic cance1- in laboratory animals in long-term studies. The implications of such findings in tenms of carcinogenic potential for humans are not known. The induction of a tumon in a particular organ, such as the pancreas, in one animal species does not mean that the same organ will be affected in man on any other animal species (Tomatis et al , 1973). Section 2. Methodology A detailed questionnaire was developed by the Depantment to collect information on those Peabody residents who died of pancneatic cancen between 1974 and 1982. The death centificate was obtained for each resident to verify the cause of death. The questionnaire included a medical, occupational, and residential history as well as questions on personal habits and environmental exposures. Table 1 lists the types of questions asked during the interviews. Questions on pensonal habits were developed after reviewing the health litera- ture to identify risks factors for pancreatic cancer. 7 t Table 1. Types of Information Obtained Durir.g Ir.terviewir. Personal Habits: Er.viror.mer.tal Exposures: smoking history coffee corsumptior tea consumption alcohol consumption hobbies source of drirking water taste of drinking water contact with ponds in Peabody contact with industrial waste proximity to industry neighborhood air pollution neighborhood odors gardening Occupational History Medical Histo-v: heart disease gastrointestinal diseases respiratory disease surgical history reproductive history family medical history Residential History 3 t » Section 2.1 Selection of Cases Initial -eview of the state's health statistics identified 49 Peabodv "•esider.ts who d i e a if parc-eatic career between 197 4 a r. d 1982. Set wee-. October 11, 1983 and November 1, 1983. fortv-five interviews were conducted with relatives of the deceased (informants listed on the death certificates). Information obtained during the interviewing of two informants caused the removal of two cases from the study; one case never lived in Peabody while the second case lived in a Peabody nursing home only fon a few months before her death. Two informants refused to participate in the study and four informants could not be "eached by phone or letter after several attempts. Finally, two interviews were conducted with relatives of people who were not among the 49 cases originally identified. One person with pancreatic cance" had lived on the Peabody/Lynnf ield line for ten years. The other person, lived in Peabody for thirty years but died in 1970, four years before the time period of study. Although these last two cases were not included in the statistical analysis, interviews we-e conducted in case additional information might be provided by these informants. The statistical analysis, therefore, included 41 cases. Section. 3. Survey Results Of the 41 cases, 22 were males and 19 were females. Figure 1 shows the age distribution, at time of death of the cases. Sixty-six percent of the cases were age 65 and over. Two percent of the cases were under 45, five percent were under 55, and thirty-four percent were under 65. 9 « Cx 1 U " 2" it: ^1 txj £H \M 1 i ^ 1_1 3« 3<3 HO *0 1/^ ^ TO T* 75 "J*? ?D 3^ fl'H- Aye. C year^ Figure 1. Age Distribution of the Peabody Pancreatic Cancer Cases at the Time of Death. 10 c Occupational histories were obtained during the interviews. Nine (22%) of the cases worked ir. a tannery at some time in their lifetime; five (12%) of the case?5 worked ir a tannery for 15 vea-s of mo-e. Four (10%) of the cases worKed in metal fabrication jobs. Two of tne cases were ,-eported to have had daily exposure to asbestos - one was exposed to asbestos daily for four years while in the Navy and the other was exposed daily for 41 years on his job as an enginee-. Eleven (27%) of the cases were housewives. One case worked as a foreman at a chemical company for 32 years. The remainder of the cases worked in miscellaneous types of jobs, such as clerical and maintenance work. Informants were asked about the occupations of the cases' spouses. Occupational health studies have shown that the spouses of workers in certain industries may be at increased rj.sk f0>» certain diseases associated with their spouse's occupation; the presumption is that workers may bring home harmful substances on their clothes and thus expose family members. Of the 3^ cases who were narried , the spouses of seven of them (21%) were employed at some time in a leather factorv. See Appendix B for the range of possible workplace exposures in the leather industry. One spouse worked at the Danvers 31eachery, three worked at an electronics firm, two worked in shoe factories, and one was employed in the dry-cleaning business. The remaircjer Qf the spouses worked at various jobs where exposure to toxic substances would have been highly unlikely. Several questions pertained to potential environmental exposures associated with the residence where the case lived the longest or spent most of his or her life. One type of question sought people's opinions on the quality of the air and water in the case's neighborhood. A second type of question was designed to ascertain whether the cases had any common link with the environment (such as swimming in a particular pond). I 1 i c Six (15%) of the 41 cases spent most of their lives outside of Peabody. Section 3.1 is an analysis of the residential history information collected du^ir.g the irte'-views. Df the 4: cases, or.ly tnree (72) were -eported as ever swimming in Browns, Spring, or Sidneys Pond. Only one case was reported to have ever fished in these ponds. Concern about potential contact with waste lagoons at Eastman Gelatine's facility has been raised. See Appendix H for a discussion of Eastman Gelatine. Six waste lime lagoons cover about 22 acres. Seven out of 41 (171) cases were reported as having some contact with the lagoons; contact included working near, playing near, or living near the lagoons. No one was "eported to have had actual phvsical contact. Thirteen informants (32%) reported unusual odors in their neighborhood. The reported sources of the odors were the tanneries, the lime pits (Eastman Gelatine's waste lagoons), piggeries, the North River, and Borden Chemical Company. Seventeen (42%) informants reported that their relatives had noticed an unusual taste in the drinking water. Seven informants described the water as tasting fishy. Thirty-three (80%) informants said that their relatives did not complain of neighborhood air pollution. Table 2 is a summary of information on personal risk factors collected during our interviews. Column one lists the characteristics examined and column two contains the number and percentage of total cases that possessed the characteristic. The frequency with which these characteristics appeared in cases in other studies of pancreatic cancer are shown in column three. Coffee and tea consumption patterns were generally similar among the cases in the different studies. A somewhat higher percentage of the Peabody cases were smokers than in MacMahon's study. 12 c Table 2. Presence of Personal Risk Factors in Peabody Panc-eatic Cancer Cases Compared to Other Studies Characteristic Studied Number of Peabody Cases Percentage of Peabody Cases Percentage Identified in Other Studies Medical History History of Diabetes History of Gallbladder Disease History of Gallstones History of Chronic Pancreatitis Removal of Gallbladder Removal of Ovary or Ovaries History of Ute-ine Tumor Coffee Consumption Drank more than 5 cups daily Drank 3 to 1 cups daily Drank 1 to 2 cups daily Drank less than 1 cup daily Never drank coffee Unknown 7 17 7.4* 3 7 4 10 20. 4* 1 2 6.5* 5 12 13.8* 3 7 35.0* 4 10 27.5* 10 24 24** 12 29 29** 12 29 42** 10 25 4 10 5** 3 7 Tea Consumption Drank 3 or more cups daily Drank 1 to 2 cups daily Drank less than 1 cup daily Never drank tea Unknown 5 12 13** 14 34 60** 10 25 9 22 28** 3 7 Smoking Habits Smoked less than 1 pack daily Smoked a pack or more daily Never smoked cigarettes Unknown 7 17 11** 14 34 23** 16 39 28** 4 10 * Lin et al, 1981. ** MacMahon et al, 1981. 13 Section 2.1 Residential Clustering For the period 1974 to 1982, an excess r.umber of deaths due to cancer of the pancreas occurred in census tract 2106 of Peabodv. A total of twelve deaths occurred while only four deaths were expected, based upon statewide mortality rates for pancreatic cancer. The likelihood of this occurring by chance is extremely >-are(9 out of 10, 000). As stated earlier, most human cancers (with the exception of childhood leukemia) have a latency period of 20 to 40 vears. For this -eason , it is important to look at where the Peabodv pancreatic car.cer cases lived 20 to 40 vears before the date of their deaths. Using the -esidential history information collected on the questionnaire, the place of -esidence in 1950, 1960, and 1970 of the pancreatic cancer cases was plotted (by census tract). Nine census tracts exist in Peabody now; they are numbered 2101 through 2109. In 1 950 and 1960, only five census tracts existed in Peabody. The maps on the following three pages illustrate where the cases lived in 1950, 1960, and 1970. U.S. Bureau of the Census popula- tion statistics were collected for Peabody's census tracts for these three years. To investigate whether the pancreatic cancer cases were concentrated in any particular area of Peabody during their latency periods (when exposure supposedly occurred), the following was computed: (1) the percentage of the total number of pancreatic cancer cases who lived in a particular census tract (2) the percentage of the total Peabody population who lived in that same census tract 14 4 ON o o 01 01 CO CO u 11 CJ C co 'J co V M "J C — — o JZ CO co u_ 3 3 co C 01 0) y u 3 >. — — co 3 2£ CT\ 3 C Ol 3 =0 •H -4 ^ 01 co ta u M CO CJ fi • CO CJ 4-1 CO 0) u u c CO 15 i O X) o i-l cj c CO — z CO • o> ^ CJ CO 1) — H u 3 O vj c 1) o> s ° u >> . — GO O 2£ CT\ 0 C "J 'J V u - ■A .-I CM CM LO -4 CM a; w CO U u o> CJ C CO CJ u CO 01 1- l o C CO 2- , 16 < 0> C 00 c CO V ■J) CO U V CJ to (U M -J — "3 O -2 CO 0) 09 - 3 3 ■ £ jj y "J U — 3j ^ O C u CO 0) 3 =0 0) CO CO CJ u a> CJ C CO U CO - o c CO 17 1 (3) the likelihood of observing (bv char.ce) that pe-cer.tsge of cases in that particular census tract (using tne binomial distribution statistical test) A percentage of cases in a particular census tract significantly greater than the percentage of the total city population living in that tract (in a specified year) suggests an elevated number of cases in that tract. Appendix C contains the data used in this section and a brief description of the statis- tical test applied to the data. In 1950 and 1960, no elevation in pancreatic cancer mortality in any particular census tract appeared. That is, the percentage of cases in any one of the census tracts, based on the population distribution of Peabodv, was as expected for those two years. However, in 1970, more cases lived in Census Tracts 2106 and 2109 than were expected. Therefore, during the vears critical to the development of the cancers, the cases did not tend to live in any particular area of the citv. This means that common exposure of the cases to environmental pollutants probably did not occur. Section 4. Environmental Survey and the Potential for Exposure Many factors may contribute to the occurrence of a disease. These factors include occupational exposures, nutritional status, genetic predisposition, personal habits such as excessive alcohol consumption and smoking, and exposure to environmental pollutants. This section reviews possible exposure of the cases to environmental pollutants. 13 For an environmental pollutant to exe-t an adverse effect or a pe-sor , it is 'ece?sarv for that person to be exposed to the .substar.ee ar.d to abso-b it into the body. Three possible "outes of exposure exist: inhalation, ingestion, and skin contact. Sources of exposure include air pollutants (taken in by breathing); water contaminants (ingested in drinking water or absorbed through the skin while swimming); soil contaminants (absorbed through the skin upon contact or ingested as dust particles); ar.d food contaminants (ingested in the diet) . Due to the latency period of cancer, it would be ideal to have environmental data for each of these -outes f*-om at least 1950 onward. Most of the data for Peabodv, however, a^e only from the early 1 970s to the present. Lacking earlier information, past and present industries and dumpsites in Peabody are analyzed to determine whether contaminants associated with them could have contributed to the incidence of pancreatic cancer. 4.1 Environmental Testing This section has two parts. The first part discusses resuj_ts of testing performed on Peabody's drinking water. The second part addresses specific sites where exposure to contaminants may have occurrea. Only summaries of environmental data are presented here. Detailed results are given in the appendices . 19 < 4 4.1.a Tests or Peabodv D'-ir.kirg Wate- Some -esiderts ir Peabodv have private wells ar.d do rot get their water f -om a public supply. Only two of the forty-ore cases, however, used private well water. Nearly all the other cases used wate>- from the downtown svstem of the Peabody wate- supply (see Apperdix E for a more detailed description of Peabody 's water system). Testirg for ervirormertal cortamirarts ir this water is the-efore of particular ir.terest. Wate" quality data for Peabodv exists f»-om 1975 to the presert (see Ap- perdix F). Tests have beer, conducted or samples f*-om Sp'-ir.g Por.d, Surtaug LaKe, zap wate", ar.d the Ipswich River (downtown system); and f*~om Wir.or.a Pond, the Johnsor. St-eet well, ar.d the Pine Street well (West Peabody System). Aralyses of stardard water quality parameters required by the federal Safe Drirking Water Act and the Drirking Water Regulations of Massachusetts wee car-ied out. These irclude pH, turbidity, inorganic solutes, certair. pesticides, microbiological contami rant s and others (See Apperdix F). With -egard to organic and inorganic chemicals (including metals), levels found in the drinking water were usually within federal standards. Ir some samples, iror concentrations exceeded the federal secordary stardard of 0.3 ppm. Sodium levels in the tap water slightly exceeded the state stardard of 20 ppm ir several samples. 20 i 7-ihalomethares (THMs) have also beer tested ir. Peabody wate-. THMs are present ir e ve-v chlorinated drinking wate" suppiv. T'nev are formed as a by-product of the chlorir.atior process (used to control microbial contamination) when chlorine reacts with naturally occurring substances in the water. In tests done in 1976, 1980, and 1982-1984, THMs were within the acceptable levels designated by the EPA (100 ppb averaged over 4 quarterlv samples) (see Appendix F). F'-om the results of testing, Peabody's wate*- supply shows ro unusual level of any contaminant. However, thee have been numerous complaints by local residents and informants for the cases ^egardir.g the qualitv of the drinking wate". It would be useful to know which substances, if any, were p-esent in the drinking water 20 to 40 vears before the mid to late 1970s. Since this information is not available, it is not possible to draw firm conclusions -egarding the relationship between drinking water and pancreatic cancer . 4.1.0 Testing of Specific Sites This section addresses specific sites in Peabody where exposure to environmental contaminants may have occurred. For each site, possible routes of exposure are discussed, as well as what is known about exposure to any of the cases. For the location of the sites, see Appendix G, Figure G.1. 21 I. gastmar Gelatine Lagoors - The lagoor.s, located off Wheele- Street ir -T 2106, cove- a- a-ea of 22 ac-es. Eastman Gelatine, a manufacturer of high-quality photog-aphic gelatines, has dumped wastes f-om thei- ope-atior.s ir.to the lagoor.s fo- over 50 years. The lagoor.s also contain waste dumped by glue factories which occupied the site for over 100 years prior to Eastman Gelatine . According to our survey, no case had any actual physical contact with the lagoon site. This rules out the dermal route of exposure from the site itself. Seepage f-om the lagoons has entered both Sidney's Pond and a potential aquife- found ir. the Meadow Pond area (C.T. Main, 1983, PP. 4-10). Acco-ding to the survey, however, none of the cases ever swam or fished in Sidney's Pond. In addition, the aquifer has never been part of any water supply system. The lagoons are also dowr.st-eam from Spring Pond, which is part of the Peabody water system. Thus, even if wastes from the lagoons entered various waters ir. the area, the cases were not exposed to the wastes via thei- drinking water. Area -esider.ts have complained of odors p-oduced from the lagoor.s. This -aises the question of whether exposure to toxic substances in the ai- may have contributed to the elevation of pancreatic cancer. Ir. April 1983 , Eastman Gelatine hired consultants, TRC, to do air sampling over the lagoons. Samples were also taken upwind and downwind from the site. The tests detected 4 kinds of amines and ammonia over the lagoons themselves, but not at the up- wind and downwind sites. The source of the amines and ammonia is probably the decomposition of proteinaceous material. These substances are quite basic, have a fishy odor, and are very irritating to tissue. 22 i Amines, ur.de1" neutral or acidic conditions ar.d ir the presence of nitrogen oxide compounds, have the potential converting to r.i trosamir.es . Nic-osamines a*-e suspected to be carcinogenic. Amines are, howeven, ubiquitous in the environment. Given the lack of past air data from the lagoons and the complexity of nitrosamine formation, it is impossible to predict how much, if any, conversion took place. Furthermore, a review of the residences of cases in 1950 and in I960 showed only six of the forty-one cases lived within a quarter mile radius of the site. Thus, the amines detected over the lagoons do not appear to be an important ^isk factor in this study. 2.5alem Aces - Arother environmental concern for residents is Salem Ac-es, located in Salem, just over the border from the Eastman Gelatine lagoons. Closed in 1 969 » this dumpsite, consisting of four sludge pits, was used for many years by the South Essex Sewerage District (SESD). Midnight dumpers are also suspected of dumping tannery wastes in Salem Acres (EPA Potential Hazardous Waste Site Identification and Preliminary Assessment, 1983). A consultant for EPA, the NUS Corporation, conducted tests at Salem Acres. NUS sampled soil from the sludge pits, air above the pits, and surface wate»- and groundwater ir. the area. Breathing zone samples of ai- above holes drilled in the sludge pits indicated "nothing unusual" was found (Panaro). Since Salem Acres is down- stream f»*om Spring Pond, it does not impact on Peabody's water supply system. Grit samples taken directly from the pits show levels of about 800-900 ppm 23 chromium ir two of the fou>- pits, ar.d about 25 ppm chromium ir the othe" two pits (Par.aro). These pi-elimir.a,-v results a-e still beirg evaluated. The 800-900 ppm measurements a»-e elevated levels (the mean ir urcortamir.ated soils is 100 ppm, with a range of 5-3000 ppm). Ir. summary, Salem Acres car probably be ruled out as a source of exposure of environmental contaminants to the cases. Salem Acres does rot impact on Peabody's water supply, ard for exposure to the chnomium found in the pits to occur, actual skin contact would have had to take place since chromium is rot volatile. 3. Pierport Street Park - This location, was or.ce the site of a rumbe- of leather comparies. The site is believed to have beer used for dumpir.g tarnery wastes ir. the past (Peabody Commurity Developmert Departmert, local -esider.ts ) . About 10 years ago, a large fire destroyed the existing factories. Since then, the area has beer a vacart lot used bv neighborhood childrer. . In 1981, the site was developed into a playground. No question about contact with Pierport Park was or the questionnaire . Nonetheless, two points can be made. First, before 1973 (about the time of the fire), the area was occupied by various factories, makirg contact with the area by the cases unlikely. Also, no case worked ir either B.M. Moore or Cent-al Leather, which were located on the site. Secondly, rot mary cases lived rear the area. In 1950 and again in 1960, only one case lived within a quarter mile -adius of the site. Ir 1970, orly two cases lived withir a quarter mile radius of the site. Thus, it is highly unlikely to be a -isk factor for parcreatic cancer in our study. 24 * In p»-epa»-at ion for developing the site into a playground, a numbe*- of soil analyses we'-e conducted at Pie^pont Street Park. These -esults a-e shown in Appendix J. Preliminary testing showed high levels of chromium and lead in the soil. Lead is a neurotoxin and chromium in its hexavalent state is an irritant and a carcinogen. The soil analyses did not determine whether the chromium was in the hexavalent state or in its less toxic trivalent state. 4 . Strongwater Brook - Strongwater Brook originates in South Peabody and Salem, flowing in a northerly direction through parts of CT 2106, by Pierpont Street Park, and into the North River. The interviewers did not ask the informants whether cases had ar.v contact with the brook. However, only six of the forty-one cases lived within a quarter mile of the brook in 1950 and again in 1960. Furthermore, since it is downstream from Spring Pond, Strongwater Brook does not impact on Peabody 's watei" supply. Therefore, the brook is not a likely risk factor in the elevation of pancreatic cancer in Peabody. 5.Peabodv Landfill and GCR Landfill - The Peabodv landfill has been used by the city since the 1950s. After Salem Acres was closed in 1969. the SESD began disposing of its grit and scum wastes at the Peabody landfill. Due to concern about leaching from the landfill into surrounding surface and groundwaters, a consultant to DEQE, Tighe and Bond/SCI, conducted testing in 1980-1981 of groundwater, surface water, and leachate at the landfill. The Peabody landfill is in a different drainage basin than Suntaug Lake, and thus poses no -isk of contamination of that water supply. Surface and subsurface waters in the basin where the landfill is located eventually discharge into Goldthwaite Brook (EIR, Tighe and Bond, 1981, pp. 4-9). 25 The GCR landfill is adjacent to the Peabodv lardfill. Testing was done ir September 1983 bv GCR Er.giree-ir.g or leachate ard surface and groundwaters arour.d the lardfill. As with the Peabodv landfill, the GCR landfill does not impact on Peabody's water system because surface water drainage flows towards Cedar Pond and Cedar Swamp, which drain eastward into Goldthwaite Brook (GHR.1983, pp. A-10). Net groundwater flow from the landfill is toward the southeast (GHR.1983. pp. A-12). No plausible -oute of exposure between contaminants in the landfills and our cases could be identified. Furthermore, no cases lived near the site. Therefore, it is highly unlikely that exposure to contaminants in the landfills contributed to the development of these cancers. 6. Other Areas- Some citizens have expressed concerns over flooding of homes in the vicinity of Remis Industries. Residences of cases were reviewed to determine whether thev may have lived in the homes which were flooded. In 1950, no cases lived near Remis Industries. In 1960, one case lived in the vicinity (near the corner of Lynnfield Street and Farm Road), and in 1970, no case lived in the area. Othe1- flooding has occurred in homes on Fountain Street, near Pierpont Street Park, when St -or.gwater Brook overflows its banks after very heavy -airs. No case, however, lived on Fountain Street. 26 Section 5. Conclusions and Recommendations The goals of this studv we-e three-fold: o to investigate whethen mortality from pancreatic cancer in Peabody is related to environmental con- tamination in Peabody o to collect information on »"isk factors associated with the disease o to investigate an apparent concentra- tion of cases of pancreatic car.cer in CT 2106. The mortality >-ates for cancer of the pancreas were examined, for three time periods. The trend in pancreatic cancer in Peabody is as follows: from 1969 to 1 973. r-o significant excess of pancreatic cancer cases was seen; from 1974 to 1978, a statistically significant elevation in pancreatic cancer mortality occurred; and, between 1979 and 1982, no significant excess was noted. Thus, for Peabody as a whole, the significant elevation appeared during the five-year period 1974 to 1978; it appears that the incidence of pancreatic cancer may be returning to expected levels. Existing environmental data for Peabody, including air, water, and soil testing results, were collected and reviewed. The potential for exposure of the cases to environmental contaminants was considered. Based upon a careful review of the testing data and the Peabody environment, actual exposure of the cases to the industrial or environmental contaminants identified in the* "eport seems highly unlikely. Although legitimate concerns exist regarding past waste disposal practices in Peabody and potential threats to groundwater 27 4 4 and surface waters, r.o correction between environmental contamirarts ard the irciderce of par.c-eatic car.cer could be made. Howeve*-, because of the lorg latercy period of most careers, exposures could have occurred back ir the 1940s and 1950s which could not be documented. Information on personal risk factors was collected durirg the questionnaire survey. In general, no characteristics or personal habits of the cases were noteworthy. Due to the large rumber of leather tanneries in Peabody, employment in the leather industry was of interest. Twenty-two percent of the cases (and 21 pe'-cert of the spouses) were employed ir a tannery at some time ir their lifetime. Howeve'-, ir. 1968, less thar ter percent of the cases worked ir the leathe*- irdustry while twelve percent of Peabody's labor force was employed ir leather at this time. Similarly, while sixteen percent of the Peabody labor force was employed ir leather ir 1960, less than fifteer percent of the cases were workirg in leather then. Therefore, the percentage of cases who worked ir the leather irdustry was approximately the same as the pe*-certage of Peabody -esiderts who were employed in leather during those times. Ir 1950 and 1960, the time of the latency period, no clustering of cases in any particular area of Peabody was seen. In 1970, more pancreatic cancer cases lived in CT 2106 and 2109 than would be expected given the distributior of the Peabody population. Therefore, the apparent clustering in CT 2106 is a -elatively recent phenomena. It could be that persons who were later to develop pancreatic cancer tended to move into CT 2106; perhaps for convenience and proximity to Peabody center or because of neighborhood characteristics. In CT 2109, a plausible explanation for the elevation is the large elderly population in that area. 28 4 / € Fifteen percent of the cases spert most of thei- lives outside of Peabody. Ir 1950, at least or.e-half of tne cases lived outside of Peabodv ar.d ir. 1960 or.e-quarte»- of the cases were not Peabody -esider.ts. Seventeen informants said that their relatives who died of pancreatic cancer had -eported an unusual taste in their drinking water. According to city water officials, the taste is due to algae, which are nonpathogenic organisms commonly encountered in water impoundments. They acknowledge a long-standing problem with algae (which have occurred in large "blooms" in the past) in the city water supply and have taken steps to alleviate the problem. For example, some point sources, such as drains discharging high phosphate waters, which encourage algae growth, have beer, eliminated. The Ceda*- Grove Reservoir was cleaned and disinfected in September 1983. and Winona Pond will be aerated this spring. In the last couple of years, the-e have been only minor algal blooms. Several limitations exist in this investigation. Environmental sampling data and exposure information for the latency period was lacking. The quality of the information obtained by interviewing informants for the pancreatic cancer cases depended on the closeness and relationship of the informant to the case and the length of time which had elapsed since the death of the case. Reporting bias may have occurred if the informant felt that pancreatic cancer mortality was or was not related to living in Peabody. Because the etiology of pancreatic cancer is still not well understood, it is possible that information on important but unknown risk factors was not collected. Although 29 r deductions could be made about contact of the cases with Pie»-por.t Street Pa-k ar.d St'-orgwace- Brook, the questionnaire did not contain specific questions about either area. Finally, instability and lack of statistical power associated with the small number of cases characterize this investigation. Based upon the analysis of data collected in this study, the Department feels that pancreatic cancer mortality in Peabody is probably not related to living in Peabody. However, one recommendation is made below and is prompted by information brought to the Department's attention through this study. The Department recommends that further analvses be conducted on soil samples at Pierpont Street Park as soon as possible. Although exposure to this area is not a likely >-isk facto** fo" pancreatic cancer in Peabody, more study is -ceded to verify preliminary findings of elevated levels of lead and chromium in the soil of the park. Furthermore, the tests should' determine whether the chromium is hexavelent or trivalent. Potential exposure of children playing on the site by ingestion or inhalation of contaminated soil particles or skin contact is of concern. 30 r Bibliography Ame-icar. Car.ce- Society. Carce- Statistics, 1980. Jar.uary/Feb-uarv 1980. American Car.ce" Societv - Massachusetts "Division . Car.ce- Manual, Sixth Edition. 1932. Carlo, George L. et al . Car.ce- incidence and tr ihalomethane concentrations in a public drinking water system. American Journal of Public Health 70:5: 523-525. May 1980. Colton, Theodore. Statistics in Medicine. Little, Brown and Company. Boston. 1974. Decoufle, Pier-e. Cancer risks associated with employment in the leather and leather products industry. Archives of Environmental Health. January/February 1979. Eastman Gelatine Corporation. Neighborhood Odor Patrol Program - 1983. April 1984. Environmental Protection Agency. Potential Hazardous Waste Site Identification and Preliminary Assessment - Salem Acres. Prepared 1/26/83. F-aumer.i , Joseph F.Jr. Cancer of the pancreas and biliary tract - epidemiologic considerations. Cancer Research 35: 3437 - 3446. November 1975. Garabrant, D.H. and Wegman, D.H. Cancer mortality among shoe and leather workers in Massachusetts. American Journal of Industrial Medicine 5:303-314. 1984. GHR Engineering Corporation. Final Environmental Impact Report - GCR Landfill, Peabody, Mass. October 1983. Green Engineering Affiliates, Inc. Industrial Discharge Survey of North River and Tributaries, Peabody, MA. April 1974. International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Humans - Volume 25. World Health Organization. 1981. IARC. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans - Supplement 4. October 1982. Lin, Ruey et al. A multifactorial model for pancreatic cancer in man. Journal of the American Medical Association 245: 147-152. 1981 Longnecker, Daniel S. Carcinogenesis in the pancreas. Arch Pathol Lab Med 107: 54-58. February 1983. MacMahon, Brian et al . Coffee and cancer of the pancreas. New England Journal of Medicine 304:11: 630-633. March .12, 1981. 31 MacMahon, Brian -. Risk factors for cancel of the pancreas. Cancer 50: 2676-2680. 1982. Main, Chas . T., Inc. Eastman Gelatin Corporation - Lime Lagoon Study Phase 1 Report. Boston, MA. October 1983. Mainz, D. and Webster, Paul. Pancreatic carcinoma - a review of etiologic considerations. Digestive Diseases 19:5:459 - 464. May 1974. Mancuso, T.F. and El-Attar, A. A. Cohort study of workers exposed to betanaphthylamine and benzidine. Journal of Occupational Medicine 9:6: 277-285. June 1967. Massachusetts Department of Commerce and Development. Monograph on Peabody. 1983. Moolgavkar, S.H. and Stevens, R.G. Smoking and cancers of bladder and pancreas: >-isks and temporal trends. JNCI 67: 1: 15-23. July 1981. National Cancer Institute Monograph 59. Cancer Mortality in the United States: 1950-1977. April 1982. Rounbehler, I.S. et al . Exposure to N-Nitrosodimethy lamine in a leather tannery. Food Cosmet . Toxicol . 17:487-491. 1979. Sigurdson, Eunice E. et al. Cancer morbidity investigations: lessons from the Duluth study of possible effects of asbestos in drinking water. Environmental Research 25: 50-61. 1981. Sheffet, A. et al . Cancer mortality in a pigment plant utilizing lead and zinc chrcmates. Archives of Environmental Health 37:1:44-52. January/February 1982. Soloway, H. et al . Endocrinopathy associated with pancreatic carcinomas. Annals of Surgery 164: 2: 300-304. August 1966. Tighe & Bond/SCI & Jason Cortell & Associates Inc. Final Environmental Impact Report - City of Peabody, Mass. Municipal Landfill Expansion. 1981. Tomatis, L. et al. The predictive value of mouse liver tumour induction in carcinogenicity testing - a literature survey. International Journal of Cancer 12:1-20. 1973. Wynder, Ernest L. et al . Epidemiology of cancer of the pancreas. Journal of the National Cancer Institute 50: 645-667. 1973. Wynder, Ernest L. An epidemiological evaluation of the causes of cancer of the pancreas. Cancer Research 35: 2228-2233. August 1975. 32 ( INDIVIDUALS WHO SUPPLIED INFORMATION PEABODY Peter Ar.geramo, Director, Department of Human Services, City of Peabodv Jack Barrv, Board of Health, City of Peabody Jack B»*enran, Civil Engineer, Department of Public Services, City of Peabody F. Michael DiGianno, Director, Department of Community Development, City of Peabody Nick Morello, Assistant Director, Department of Community Development, City Peabody Pete1* Smv^-nios, Chemist and Operator, West Peabody Water Treatment Plant, Department of Public Services, City of Peabody Daniel Dousette, Peabody Historical Societv Maurice Hallinan, -etired Sewer Inspector, Department of Public Services, City of Peabodv John Dowd, Co-Chairman, Coalition for a Safer Peabody Sandra Innis, Co-Chairman, Coalition for a Safer Peabody Louise Hamilton, Coalition for a Safer Peabody MASSACHUSETTS DEPARTMENT OF ENVIRONMENTAL QUALITY ENGINEERING Northeastern Regional Office, Woburn William St. Hilai^e, Regional Environmental Engineer Victor Karian, Supervisory Sanitary Engineer, Division of Hazardous Waste John Fitzgerald, Senior Sanitary Engineer, Division of Hazardous Waste John Keating, Senior Sanitary Engineer, Division of Water Supply David Erekson, Senior Sanitary Engineer, Division of Water Pollution Jack Mullins, Senior Sanitary Engineer, Division of Water Supply Robert Tanzer, Senior Air Pollution Control Engineer, Division of Ai" Qualit Margaret Thornton, Junior Sanitary Engineer, Division of Hazardous Waste Loretta Oi , Junior Sanitary Engineer, Division of Air Quality Ed Pawlowski, Principal Sanitary Engineer, Division of Hazardous Waste 33 i Ma»*jorie Aleo, Junior Air Pollution Engineer, Division of Air Quality OTHERS Leo^a-d Pag^otto, Di*-ecto«-, Division of Occupational Hvgiene, Massachuset Department of Labc- and Industries Jeff Mathews and Scott Summers, Eastman Kodak Company, Rochester, N.Y. John Panaro, Richard Dinitto, and Paul Clay, NUS Corporation, Bedford, Massachusetts Mark Hoey, Tighe and Bond/SCI, Easthampton, Massachusetts 34 ( Appendix A Peabody Mortality Statistics for Cancer of the Pancreas This appendix contains a summary of the statistics on cancer of the pan- creas in Peabody and CT 2106 for various periods. Table A.1 is a summary of these statistics. Column two contains the observed number of pancreatic can- cer deaths and column three contains the number of deaths by cancer of the pancreas expected if statewide mortality »*ates applied. Column four contains standardized mortality ratios (SMRs) - the ratio of the observed number of cases to the expected number, multiplied by 100. An SMR over 100 indicates that more deaths occurred than would be expected. An SMR below 100 indicates that fewe»* deaths occurred than expected. For example, an SMR of 136 is interpreted as an excess of 36% of deaths over the expected number. Column five refers to the statistical significance of the SMR - whether the difference between the observed and expected number of deaths can be explained by chance (random variation of the disease within the population) or is most likely an excess of deaths by cancer of the pancreas. A yes in column five means that the difference is statistically significant and cannot be attributed to chance occurrence. The P value is the probability that the difference is due to chance; a small P value (usually below 0.05) is, by convention, interpreted as showing statistical significance.' A. I Table A.I. 3tar.da»-di zed Mortality Ratios (SMRs) for Car.ce- of the Par.c-eas ir. Peabodv, Massachusetts I. Peabody (As A Whole) II. Census Tract 2106 Period Covered 1969-1973 1974-1978 1979-1982 1974-1978 1979-1982 M* F T M F T M F T T T Observed Expected 9 12 21 15 14 29 12 8 20 6 6 10.4 8.8 19.2 10. 1 8.7 18.8 8.8 8.8 17.6 2.4 2.0 SMR 87 136 109 149 161 154 137 91 114 250 300 Statistical Si gr.i f icar.ce ( P Value)** no r.o r.o r.o no yes (0.02) r.o no r.o yes (0.04) yes (0.02) * M = males; F = females; T = both sexes combined ** Level of statistical significance at P < 0.05 A. 2 ( « The ""esiderices of the cases at tneir times of death we^e plotted on Figure A.1. The concentration of cases appeared to be east of Route 128, in the urban center of Peabody. To the west of Route 128 are the more surburban and rural parts of the city. Based upon the population distribution of the city, the location of elderly housing projects, and discussions with city officials and concerned citizens, two possible clusters of cases were identi- fied - CT 2106 and CT 2105. Even though CT 2107, 2108, 2109 appear to have elevations, the population of Peabody (including the elderly) is concentrated here. Aftc adjusting death >-ates for both age and sex, it was determined that CT 2106 (and not CT 2105) experienced a higher mortality rate for pancreatic cancer than would be expected (based upon statewide statistics). ( Appendix B Leather Industry The International Agency for Research on Cancer (IARC) has studied extensively the health risks associated with employment in various industries (IARC, 1981). Following is a summary of IARC's evaluation of the leather tanning and processing industry: "The chemical complexity of the tanning process and the wide variety of finishing agents used will almost certainly result in worker exposure via inhalation and dermal contact to multiple and changing chemical pollutants. Employment in tanneries may entail exposure to a number of chemicals for which there is evidence of carcinogenicity in humans and/or laboratory animals ... A positive association between employment in the leather industry and bladder cancer is supported by a number of studies. Suggested associations between employment in the leather industry and cancer of the lung, larynx, buccal cavity, pharynx, kidney, and lymphomas exist." (IARC, 1981, pp. 243-244) The >-isk faced by any particular worker depends on the degree and type of exposure, which in turn varies with specific jobs and work areas in tanneries. Additional factors may also influence a person's chances of developing cancer, such as smoking habits, genetic factors, and health status. Table B. 1 contains an outline of the major steps in the manufacture of leather and the types of chemicals typically used in the various processes. Table B.2, taken from the IARC report, lists specific chemicals associated with various tannery operations. B. 1 Table B.1. Outline of Major S Major Steps I. Preliminary Processes 1. Sorting and trimming hides and skins 2. Soaking 3. Liming 4. Unhairing 5. Fleshing 6. Bating 7. Pickling 3. Degreasing II. Tanning 1. Vegetable 2. Chrome 3. Other nonsynthetic tannages 4. Synthetic organic tannages III. Final Processes 1. Splitting and shaving 2. Dyeing 3. Fatliquoring, oiling, stuffing 4. Mechanical softening 5. Application of final finishes 6. Boarding Source: DeCoufle, 1979- in the Manufacture of Leathe- Chemicals Used in the Process Sulfur compounds, aliphatic amines Inorganic chromium compounds (hexavalent and trivalent), tannins Coal tar dyes, pigments B.2 ( Table B.2. Chemicals Associated with Various Operations in the Tannery Process Beam House Hydrogen sulfide Ammonia Tanyard Sulfuric acid Chromium (hexavalent) Chromium (trivalent) Hydrogen sulfide Sodium 2,4,5 -trichlorophenate Retan, colour, fat liquor Formic acid Chromium (trivalent) Copper Cobalt Ammonia Benzidine Benzidine 2, 2-disulphonic acid 1-Naphthalene sulphonic acid 3, 3-Dimethoxybenzidine Finishing Petroleum distillates Butyl cellosolve Tetrachloroethylene Xylene Methyl ethyl ketone Butyl acetate Toluene Methyl isobutyl ketone Acetone Formaldehyde Isopropyl alcohol 3enzene Source: IARC, 1981 B.3 c Another chemical which has been identified in tannery environments is N-nitrosodimethylamine or NDMA (Rounbehler et al,1979). NDMA is an animal carcinogen (IARC. 1982). The source of NDMA in tanneries is not known but it is thought that dimethylamine used in the unhairing process reacts with nitrogen compounds in the air to form NDMA. Rounbehler et al reported that limited testing of the air immediately adjacent to tanneries has led them to believe NDMA is probably present in and near tanning facilities. The results of a proportional mortality study of 2,798 shoe and leather workers in three Massachusetts towns (Brockton, Haverhill, and Peabody) were recently -epcted (Garabrant et al , 1984). A statistically significant excess of stomach cancer was present among male leather workers while a nonsignificant excess was also seen among female leather- workers. However, the authors felt that this excess would disappear if the expected proportion of deaths from stomach cancer was based upon Massachusetts stomach cancer rates instead of U.S. rates. Examination of the causes of death of 289 leather workers according to the type of job they performed (hide preparation, tanning, finishing, or maintenance) was also carried out. Lung cancer was associated more strongly with leather tanning than with the other processes ; however, the number of cases was small and therefore the results should be interpreted cautiously. B.4 c Appendix C Statistical Evaluation of Clustering A brief description follows of the statistical test applied to the data on residential clustering described in Section 3.1 of the report. As discussed earlier, the place of residence in 1950, I960, and 1970 of the pancreatic cancer cases was plotted (by census tract). The percentage of the total number of pancreatic cancer cases who lived in a particular census tract was then compared to the percentage of the total Peabody population who lived in that same census tract. If the percentage of cases was higher than the percentage of the population in the census tract, then the percentages were tested to see if they were statistically significantly different. (See Tables C.l, C.2, and C.3.) The binomial formula was used to compute the probability that the observed number (or greater) of cases in a particular census tract, as a proportion of the total number of cases in Peabody, could be explained by chance. This proportion was compared to what would have been expected, under a binomial distribution (a statistical distribution which describes the data). Below is the binomial formula; for a more detailed discussion of this topic, see Colton et al. C.l i Table C.l. Place of Residence Bv Census T-act ir. 1950 of Peabodv Par.c-eatic Car.ce- Cases 13 14 15 16 17 Census Tract (2109) (2108) (2107) (2101,02,03,04) (2105,2106) Number of Cases ir. Census T-act 2 2 4 7 9 (5.06±2)» (6.24±2.15) Number of Cases in City ** 24 24 24 24 24 Census Tract Population 3369 3955 4636 4789 5896 Total City Population 22,645 22,645 22,645 22,645 22,645 Percentage of Cases in Census Tract 8.3 8.3 16.7 29.2 37.6 Percentage of Total City Population in Census Tract 14.9 17.5 20.5 21.1 26.0 Apparent Elevation Yes Yes Level of Statistical Significance (P value) - 0.229 0.147 Statistically Significant Elevation No No ^Expected number of cases plus or minus the standard deviation. **From a total of 43. The remainder were not living in Peabody in 1950. C.2 i Table C.2. Place of Residence Bv Census T-act ir I960 of Peabociv Pancreatic Car.cer Cases 13 14 15 16 17 Census T-act (2109) (2108) (2107) (2101,02,03,04) (2105,2106) Number of Cases ir. Census Tract 434 11 10 (3.84+1.8*0* (8.26+2.48) Number of Cases ir. City *« 32 32 32 32 32 - Census Tract Population 3863 3391 4148 12,496 8304 Total City Population 32,202 32,202 32,202 32,202 32,202 Percentage of Cases in Census Tract 12.5 9.4 12.5 34.4 31-3 Percentage of Total City- Population ir. Census Tract 12.0 10.5 12.9 38.3 25.8 Apparent Elevation Yes Yes Level of Statistical Significance (P value) 0.55 0.30 Statistically Significant Elevation. No No •Expected number of cases plus or minus the standard deviation. **From a total of 43. The remainder were not living in Peabody in I960. C.3 \ Table C.3. Place of Residence Bv Census T-act ir. 1970 of Peabodv Par.c-eatic Car.ce- Cases Census T-act 2109 2108 2107 2106 2105 2104 2103 2102 2101 Numben of Cases in Census Tract 7259 28 211 (3.±1.7)» (3.1±-7) (4. ±1.8) (5.6+2.2) Number of Cases 37 37 37 37 37 37 37 37 37 in City ** Census Tract Population 4102 3551 3932 4786 5259 7348 5657 6371 7074 Total City Population 48,080 (same) Percentage of Cases in Census T-act 18.9 5.4 13.5 24.3 5.4 21.6 5.4 2.7 2.7 Percentage of Total City Population in Census T-act 8.5 7.4 8.2 9.9 10.9 15.3 11.8 13.3 14.7 Apparent Yes Yes Yes Yes Elevation Level of Statistical Significance (P value) 0.034 0.183 0.009 0.196 Statistically Significant Elevation Yes No Yes No *Expected number of cases plus or minus the standard deviation. **From a total of 42. The -emainder were not living in Peabody ir. 1970. C.4 1 The binomial formula states that: the probability of observing x or more cases = |n| 6X (1-5) n~x where x = the number of cases in a census tract n = the number of cases in the city 9 = the probability of observing a case in the census tract As is standard practice, a probability of less than 0.05 was considered to be statistically significant. C.5 ( r 4 Appendix D Peabody Industrial History Peabody has been called the "Leather City of the World" with tanneries dating back to pre-Revolutionary times. The industry expanded greatly at the beginning of this century, making a section of Main Street one of the largest leather processing districts in the world. In 1919. there were 91 leather establishments. The industry gradually declined, with the number of firms decreasing to 68 in 1955, and to 50 in 1968. By 1968, tanneries employed only 51} of the manufacturing employees (3,026 employees) in Peabody as opposed to 80% only a few years before (Monograph on Peabody, 1983). Today, leather establishments are still the largest source of manufactur- ing jobs in Peabody, with 36 firms employing 26% of manufacturing workers. Chemical and allied products follow, with 11 firms employing '\9% of manufac- turing employees (see Table D . 1 ) (Monograph on Peabody, 1983). Figures D.1 and D.2 show the location of tanneries in Peabody in 1930 and in 1978. A number of tanneries were formerly located on Foster Street in the northeast corner of CT 2106. Also located on Foster Street was the Danvers Bleachery which closed in the 1950s. The bleachery was engaged in the bleaching, dyeing, and finishing of cotton goods. Today, the only major industry in CT 2106 is Eastman Gelatine Corporation, located on Washington Street and Allen Lane. Prior to Eastman Gelatine, the site was occupied by two glue factories, the Upton Glue factory from 1808 to 1 89-4 • and the American Glue factory from 1894 to 1930. D . 1 Table D.1. Peabody, Massachusetts Industries ir. 1930 and 1983 Type of Industry Number of Units 1930 Leather 48 Chemical and Allied Products 4 Glue and Gelatine 2 Miscellaneous 6 Source: Maurice Hallinan, "etired Sewer Inspector, Peabody Department of Public Services 1983 Leather and Leather Products 36 Machinery, except electrical 14 Chemicals and Allied Products 11 Fabricated Metal 10 Printing and Publishing 7 Plastic and Rubber Products 5 Miscellaneous 21 Source: Monograph on Peabody, Massachusetts Department of Commerce and Development, 1983 D.2 D.l. Location of Tanneries in Peabody , Massachusetts in the year 1930. X = Location of Tannery D.3 Figure D.2. Location of Tanneries in Peabody, Massachusetts in the year 1978. X = Location of Tannery D.4 f Apper.dix £ Peabody Water Supply System The following information was supplied by Jack Brennan and Peter Smyrnios of the Peabody Wate** Department. Figure E.I shows the location of various parts of the water supply system. The system has historically been divided into two systems: the West Peabody system, which serves the western section of Peabody, and the Downtown system, which selves the downtown area and South Peabody. West Peabody wate- is treated at the Winona Pond treatment plant, which was constructed about 10 years ago. Prior to that, West Peabody received its wate»- from the Johnson Street and Pine Street wells. After the treatment plant was built, the wells were closed for a few years. In the mid to late 1970s, the wells were ""eopered, and their waters are being pumped to Winona Pond to supplement the West Peabody system. Except for the addition of the Ferry Lane Reservoir (1960) and the Ipswich River Pump Station (1971), the Downtown system has remained essentially the same since 1905. Wate" for the system is pumped from the Ipswich River (prior to 1971, the city of Lynn's pumping station was used) to Suntaug Lake. It is also chlorinated at the Ipswich River Pump Station. From the "iver to the lake, the water travels in pressurized water mains. F^om Suntaug Lake, the water travels in mains by gravity to Spring Pond. Adjacent to Spring Pond is the Coolidge Pump Station where the water is chlorinated a second time. From the Coolidge Pump Station, the water is pumped into the downtown distribution system. This distribution system contains two elevated $ * U CU o E.2 i f holding tanks, the Ferrv Lane Reservoir and the Ceda>~ G»-ove Rese'-voi r . Wate»" cnac is not used in the initial distribution is sent to the reservoirs for storage and to help maintain pressure in the system. Since Peabody's wate>- travels in a pressurized system, water will leak out and not into pipes where small leaks may occur. In the late 1950s, a line from the MDC system was hooked up to the Peabody system. While MDC water has not always been mixed with Peabody water, up to a million gallons a day of MDC water has been mixed into the downtown system at various times during this period. Today, a part of South Peabodv is using only MDC water. The switch was made in November or December of 1983, and included the area south of the junction of County and Lynn Streets. Peabody Water Department officials consider this use of MDC water to be more efficient. Spring Pond is a protected water source. This means that all potential land use activities which could threaten the quality of the water have been identified and action has been taken to protect against those threats. In the case of Spring Pond, no residential or industrial drainage enters the pond. A drainage line that runs adjacent to the pond (since 1932-1933) serves to send surface »*unoff downstream from the city water supply. Both Eastman Gelatine's waste lagoons and Salem Acres are located downstream from and at a lower elevation than Spring Pond, thereby ruling out contamination of the pond from these sources. The Cedar Grove Reservoir is also upstream from the lagoons and Salem Acres. Suntaug Lake is not a protected watershed on the Lynnfield side of the lake because some homes border the lake. E.3 ( There has been some oontroversv ove>- the fact that an old line exists between Cedar Pond and Spring Pond. Apparently, in 190 5, a pipe connecting Cedar Pond, with the gravity main running between Suntaug and Spring Pond was put in place. While the main was being completed, Cedar Pond water was used for the water supply. However, according to an article from the Peabody newspaper, dated December 20, 1905, after the main was completed, the town would use Suntaug Lake water exclusively. Cedar Pond water could be used in an emergency, as apparently happened in 1916, when Peabody had low rainfall activity and a high demand for water. In another article, dated February 5, 1920, reference was made twice to the fact that the town's supply was Spring Pond, Suntaug Lake, and the Ipswich River. City water officials have also stated that the pipe from Cedar Pond to the main water pipe was closed off, probably in the 1920s. Thus, it appears that water from Cedar Pond has not been used since at least the 1920s. E.4 0 Appendix F Peabody Drinking Water This section contains results of testing that has been conducted on Peabody drinking water. Included is a sample form (Table F.1) that shows parameters routinely tested. Trihalomethane results are also included (Table F.2). In addition, some tests for volatile organic compounds have been done bv DEQE (Table F.3). Annual tests a-e also performed for certain inorganics (Table F.4), and about every three years for certain pesticides (er.drir. , lindane, metho xychlor , toxaphene, 2,4-D, and 2,4,5-TP). A DEQE official indicated the only test results on pesticides for Peabody showed that none were detected (personal communication , Keating ) . Table F.1. Example date*' Supply Analysis (mg/i) Source A Source B Source C Source D Source E Source F Spring Pond near Gatehouse -229-01S Suntaug Lake near Gatehouse-229-02S Ipswich River, Raw water at Intake - 229-03S Winona Pond Res., Raw - 229-04S Winona Pond Res., Finished - 229-04 . EPA Drinking Water Standards Sample No. 563865 Date of Collection 3/29/33 Date of Receipt 3/29/83 866 867 868 869 Turbidity Sediment Color QP.olinity- Total (CaC03) 1 . 1 0 23 7.2 18 0.7 0 30 7.2 20 0.7 0 .70 6.7 15 1.4 0 45 7.2 21 0.2 0 0 8.0 22 1 TU (up to 5 TU) 15 color units 6.5-8.5 Hardness (CaC03) 33 Calcium (Ca) 7.8 Magnesium (Mg) 3.2 Sodium (Ma) 17. Potassium (K) 1.5 Iron (Fe) .04 Manganese (Mn) .04 Sulfate (S04) 16 Chloride (C1) 30 Spec . Cond . (micromhos/cm) 172 Nitrogen (Ammonia) .01 Nitrogen(Nitrate) 0.4 Nitrogen< Nitrite) .003 Copper (Cu) .03 37 9.5 3.1 21 . 1.6 . 1 1 .04 17 37 198 .01 0.3 003 .01 26 7.5 1.6 15. 1.3 .17 .01 17 25 146 .01 0.2 004 .00 35 7.1 4.1 15. 1.4 .30 .06 14 27 162 .01 0.4 004 .03 36 7.8 4.0 23. 1.5 .02 .03 24 30 200 .01 0.4 000 .00 250(guideline) 20. (state) 0.3 0.05 250 250 10 1 F.2 Table F.2. City of Peabodv, Massachusetts //ate- Testing Results T-ihalomethar.e Concentrations (ppb)* Locatior 4/76 6/80 1 1/82 2/83 5/83 8/83 1 1/83 2/84 Winona Plant (finished) 33.9 W.Peabody Fire Station 55.8 98 73 59 100 64 20 Brown School 33 27 32 79 13 7.5 Central Fire Station 35 21 46 83 27 17 25 Lenox Road 44 33 58 88 90 16 Proctor House Restaurant 1 12 56 . 65 97 55 21 Burk School 86 60 66 1 12 67 25 McCarthy School 91 49 41 109 59 20 Lowe Market 104 18 59' 98 63 20 Johr.son St. well 0.7 Pir.e St. well 0.4 Wate- Dept. Tap 75 48 53 96 3y definition, the THM meets the standard if the average of four quarterly samples is < than 100 ppb. The levels of THMs vary during the course of the year, tending to be highest in August or September. F.3 (0 4 Table F.J.City of Peabodv, Massachusetts Wate1- Testi-g - State Pu*-geable Ogarics Testing Results (ppb)(Julv 1980) EPA Johnson St. Well Pine St. Well Health Advisory Methylene chloride nd nd 150 (long-term ) 1,1 - Dichloroethylene nd nd 70 (long-term) 1,2 - Transdichloroethylene nd nd 270 ( 10-day) 1,2 - Dichloroethane nd nd none 1,1,1 - Trichloroethane 2.7 3.2 140 (long-term) Carbon Tetrachloride nd nd 20 ( 10-dav) Trichloroethylene 12.7 0.4 80 (long-term) Tetrachloroethylene nd nd 20 (long-te»-m) * nd = not detected F.4 4/ t Table F.I. Inorganic Chemical Concentrations (pprn) (Peabodv, Massachusetts Wate*" Dept. Tap - Dowtowr Svstem) 1979 1981 1982 MCL* Arsenic 0.0003 0.000 0.000 0.05 Barium 0.00 <0. 10 0.000 1.0 Cadmium 0.00 0.00 0.000 0.01 Chromium 0.00 0.00 0.000 0.05 Lead 0.00 0.00 0.001 0.05 Mercury 0.001 0.0005 0.0000 0.002 Selenium 0.0004 0.000 0.000 0.01 Silver 0.00 0.00 0.000 0.05 Fluoride 0.2 «).2 0.3 1.4 - 2.4 Nitrate (as N) 0.6 0. 1 0.2 10 Sodium 26 19 20 EPA sets a maximum contaminant level (MCL) for certain contaminants drinking wate"-. In setting the MCL, which should protect against adverse health effects, EPA assumes that a male adult drinks 2 lite»- of water/day over a lifetime. ** Massachusetts DEQE standard for sodium is 20 pprn. ( Table F.5. Citv of Peabodv, Massachusetts Wate" Testing Results* (February 29, 1 98 a ) Winona UNITS Pond Total Organic Carbon ppm 22. Total Organic Halogens ppm 0.13 Arsenic ppb <1 Cadmium ppm <0.002 Chromium ppm < 0 . 0 1 Copper ppm <0.01 Lead ppm <0.08 Mercury ppb <1 Selenium ppb <1 Silver ppm <0.005 Zinc ppm 0.01 Antimony ppm <0.07 Beryllium ppm <0.01 Nickel ppm <0.02 Sodium ppm 14 Suntaug Upper Spring Lower Spring Lake Pond Pond 11 45 4 0.085 0.046 <0.020 X X <1 X X <0.002 X X <0.01 X X 0.02 X X <0.08 X X <1 X X <1 X X <0.005 X X 0.01 X X <0.07 X X <0.01 X X <0.02 X X 22 * According to John Seites, Director, Peabody Department of Public Services, the data is subject to analysis and verification. t I Appendix G Dumpsites in Peabody Figure G.1 indicates present and former dumpsites in Peabody. Table G.1 lists the types of wastes deposited in the dumpsites. With the exception of the Eastman Gelatine lagoon area, none of the dumpsites are located in CT 2106. Many sites contain waste from the tannery industry, which include chromium, copper, lead, zinc, grease from leather scraps, and hide wastes. Methane can accumulate as hides decompose, and hydrogen sulfide can be emitted, causing a strong foul odor, especially in hot or damp weather. Today, most tannery wastes are discharged into the sewer system. G. 1 0« — « CO • a> 4-1 rH ■ l-l C 4-1 J2 o - 4-1 • I-l o ■u a c CO o CJ -o O c 3 CO 3 0 U 73 u XI CO • -4 £ CO CO co — u 2 XI 3 u a 3 S U-i 3 CO a 4J OJ —1 3C 0) - from animal skins until 1978, when its use was stopped (Pagnotto). It is not known, however, how much monomethylamine may have been in the ambient air. In response to complaints about odor from the lagoons, Eastman Gelatine hired consultants, TRC, to take air samples in the lagoon area. A mobile gas chromatograph-mass spectrometer instrument was used. Samples were taken directly over the lagoons. They were also taken downwind and upwind from the site. The upwind sites were near the Cedar Grove Cemetary and south and north of Sidney's Pond. The downwind sites were east of Meadow Pond, and just southeast of Welsh School. H.5 c The only substances detected we^e f^ur kinds of amines and ammonia. These we»*e detected over the lagoons but not at any of the upwind or downwind sites. The levels were as follows: C-1 amines 1-9 ppt C-2 amines 2-3 ppt C-3 amines 60-900 ppt C-4 amines 0-1 ppb ammonia 25 ppb The samples were taken in April 1983. The day was cloudy and cool. Thus, volatilization of the compounds would be less under these conditions than under warmer conditions. An Eastman Kodak official stated that, at the levels found in this sampling >-ound, none of the amines alone would have caused an odor (Mathews). However, together they may act synergistically to produce an odor. Amines can be converted to nitrosamines under neutral or acidic conditions, and in the presence of nitrogen oxide compounds (see Section 4.1.b). According to an Eastman Kodak spokesman, if nitrosamines had been present in the lagoon area in concentrations greater than 1-5 ppb, they would have been detected. Eastman Kodak also conducted 344 "odor patrols" between May and October 1983. Trained individuals visited 20 sites to determine if an odor was present. The sites bordered an area from Rainbow Road to the west to Sherwood Avenue to the southeast. They found that May, Hourihan, and Granite Streets were impacted the most with regard to odors. Their conclusion was that 2% of the time, the average person would have detected some odor, and 3.7% of the time, a more sensitive person would have detected the odor. The odor was probably originating from the amines. Discussion is currently underway on whether to continue with the odor patrols for the coming year. H.6 t Appendix I Salem Acres A consulting firm, the NUS Corporation, conducted tests at Salem Acres, which has been nominated as a Superfund site. Results of the testing are contained in a site inspection report which has not yet been released. John Panaro of NUS, however, supplied some preliminary results. The site contains sludge waste (grease and grit), possibly some tannery waste, and sewage. NUS sampled soil from the sludge pits, as well as surface wate>- and groundwater. Two surface water samples were from Strongwater Brook, one next to the pits, and one further downstream (near the Peabody/Salem border). Three groundwater samples were taken from the swamp area. One groundwater sample was taken at an Eastman Gelatine well, and another at a private well across Highland Avenue along Swampscott Road in Salem. The latter well was the closest well downstream from the pits. All water samples had "expected levels" of metals (Panaro). The results of tests for organic compounds in all water samples have not yet been received. Soil samples from the pits indicate that chromium levels are elevated in two pits (800 - 900 ppm), and are below 25 ppm in the other two pits. The quality of the data is still being evaluated. I.l Appendix J Pierpont Street Park A number of soil samples have been taken at Pierpont Street Park which was once the site of two leather companies, B.M. Moore and Central Leather. The park is now a playground. Strongwater Brook flows through the site, which is believed to have been used for dumping tannery wastes (Peabody Community Development Department, local citizens). In August 1983. soils of the park we^e analyzed for heavy metals by Skinner & Sherman. The results are shown in Table J.1. The mean and range of values for uncontaminated soils are derived from the EPA Publication "Hazardous Waste Land Treatment", SW-974, September 1980. The testing was insensitive as to whether the chromium was in the more toxic hexavalent form, or the less toxic trivalent form. Inferences on the valence state of the chromium cannot be made, since the half-life of hexava- lent chromium in soil is not known. According to an EPA official, however, the conversion of hexavalent chromium to trivalent chromium is probably dependent on the organic content of the soil. The higher the organic content, the faster the breakdown of hexavalent chromium. J. 1 r i Table J.1. Soil Testing Results at Pie'-por.t Park, Peabody , Massachusetts (Skinner & Sherman, 1983) Metal Levels (ppm) Mean Value and Range (ppm) of Uncontaminated Soils Mean Range Aluminum Arsenic Chromium Coppe*- Iron Lead Magnesium Manganese Nickel Silicon Sodium Titanium Vanadium Zinc 80,000-800,000* 8.7 1368 80 2400-24,000* 2400-24,000* 80,000-800,000* 800-80,000* 8-80* 80,000-800,000* 8000-80,000* 2400-24,000 8-80 80 6 100 20 10 350 40 100 50 (.1-40) (5-3000) (2-100) (2-200) ( 100-4000) (5-5000) (20-500) ( 10-300) * The exact concentration lies somewhere in the range shown While the chromium level found by Skinne*- and Sherman is not beyond tne expected range, it is still elevated. DEQE officials qualified the sampling results as follows: "It is difficult to characterize the pollution potential of a site based upon an analysis of a single grab sample of soil. Site location and history, pollutant types and quantities, hydrogeological setting, pathways of exposure, and the types and locations of sensitive receptors must necessarily be considered. Additionally, since sampling procedures were not observed, the quality of submitted data cannot be verified." (In a letter from Richard Chalpin to Peter Angeramo). Thus, DEQE officials believe that more study is needed to characterize the extent of contamination of the site. Goldberg 4 Zoino tested soils adjacent to this site in 1981. The soils were analyzed for heavy metals (arsenic, cadmium, chromium, lead, mercury, thallium) (see Table J. 2). The analysis used an elutriate test procedure where soil samples are subjected to processes intended to remove any leachable metals from soil particles. The results of this test can be used to predict the potential for groundwater contamination. The "leachate" produced is then analyzed for the metals. The results are compared to EPA drinking water standards, as well as to levels which would classify the soil as a hazardous waste under the Resource Conservation and Recovery Act (RCRA). For chromium and lead, two samples showed levels greater than drinking water standards. No levels, however, were great enough to classify the soil as a hazardous waste. J. 3 Table J .2. Soil Testing Results at Pie»-pont Par*, Peabody , Massachuset ts (by Goldberg & Zoir.o, 1981) Samples *# 1 2 3 4 5 6 7 Arsenic <.002* <.002 <.002 .110 <.002 <.002 <.002 Cadmium <.01 <.01 <.01 <.01 <.01 <.01 <.01 Chromium .09 <.01 .33 .03 <.01 .04 <.01 Lead <.05 .07 <.05 .07 <.05 <.05 <.05 Mercury <.001 <.001 <.001 <.001 <.001 <.001 <.001 Thallium <. 1 <. 1 <. 1 <. 1 <. 1 <. 1 <. 1 * All concentrations are in parts per million (ppm) ** Samples 1-4 were collected beneath a concrete floor slab of an old tannery building. Samples 5- -7 were collected beneath the pavement of the adjacent parking and -oad a»-eas. EPA Drinking Water Standard (ppm) RCRA Level (ppm) Arsenic .05 5.0 Cadmium .01 1.0 Chromium .05 5.0 Lead .05 5.0 Mercury .02 0.2 Thallium J .4 Appendix K Strongwater Brook In 197^, Green Engineering Affiliates, Inc. conducted a study called "Industrial Discharge Survey of North River and Tributaries". One of the tributaries to the North River is Strongwater Brook. Two samples of the brook's water were collected. One was immediately downstream of the bridge between Swampscott Street and the William Welsh School. The other sample was taken near the junction of the brook with the North River. Results of the sampling are shown in Table K. 1 . These pertain to both locations. K. 1 I Table K.I. St^or.gwater Brook Sampling Results, Peabody, Massachusetts. (Greer. Engineering Affiliates, Inc., 1974) Levels (ppm) Massachusetts Groundwater Standards (ppm) Cadmium Cobalt Copper Lead Chromium Nickel <0.05 <0. 1 <0.05 <0. 1 <0.05 <0. 1 0.01 1.0 0.05 0.05 K.2 f i ( Appendix L Peabody & GCR Landfills Testing was conducted by a consultant to DEQE, Tighe and Bond/SCI, of groundwater, surface water, and leachate at the Peabody Landfill. The landfill has been used by the city since the 1950s and is located in the Goldthwaite Brook Drainage Basin. Waters within this basin are used by Eastman Gelatine for industrial purposes. Tests done in 1980 and 1981 showed that groundwater underneath the landfill had become highly contaminated, particularly with chlorides, iron, chromium, and total dissolved solids (EIR, Tighe and Bond, pp. 4-10 and 4-11). The report noted, "Two significant factors which impact the water quality of the basin are leachates from the local landfill operations and runoff from the regional highway system" (EIR, Tighe and Bond, pp. 4-10). Tighe and 3ond also tested for certain organic compounds (PCBs and pesticides) in both surface and groundwaters, and none were detected. Five tests were conducted in March-August 1983 and January 1 984. Three sampling sites were on the site itself, one in a well east of the landfill, and one at pumphouse #3t a well owned by Eastman Gelatine. According to a spokesman from Tighe and Bond, no unusual levels were found. In the volatile organic group, most were not detected. Of those detected, all were within state or federal guidelines. Surface and subsurface waters in the basin where the landfill is located eventually discharge into Goldthwaite Brook (EIR, Tighe and Bond, p. 4-9). L . 1 ( Adjacent to the Peabody landfill is the GCR landfill, whe^e groundwate^ testing was done in September 1983 by GHR Engineering. The standard water tests as well as tests for volatile organic compounds were conducted in four monitoring wells around the site. Leachate from the landfill is contaminating groundwater (GHR, pp. A-13). Of the metals detected (iron, manganese, zinc, calcium, magnesium), concentrations were well below the Massachusetts Groundwater Standards for Class I and II groundwater. Five different volatile organic compounds were detected in two wells, and none were detected in the other two wells. None of the compounds, however, exceeded EPA's health advisory levels (where such levels exist) for these contaminants in drinking water. The groundwater of this region drains eventually to the south and southeast and does not impact on Peabody' s water supply. L.2 t < < I