Skip directly to search Skip directly to A to Z list Skip directly to site content




Table A-1. Estimation of Daily Human Exposure Dose for Ingestion
of PCE in Water

Daily Exposure Dose (mg/kg) = Conc (mg/kg) * Ingestion Rate (L/d)
Body weight (kg)


Conc = PCE concentration in milligrams per kilogram (mg/kg)
Ingestion Rate
= Average daily inhalation rate (liters per day, L/d); assumed to be
      2 L/d for adults and 1 L/d for a child (1-6 yr.) (ATSDR, 1994)
Body Weight
= Assumed to be 85 kilograms (kg) for adult males; 70 kg for adult
      females, and 16 kg for a child (1-6 yr.)

Table A-2. Estimation of Daily Human Exposure Dose for Inhalation
of Contaminants in Air

Daily Exposure Dose (mg/kg) = Conc (mg/m3) * Inhalation Rate (m3/d)
Body weight (kg)


Conc = Concentration of PCE in milligrams per meter cubed (mg/m3)
Inhalation Rate
= Average daily inhalation rate (meters cubed per day, m3/d); assumed
       to be 23 m3/d for adult males, 21 m3/d for adult females, and 16 m3/d for a child
       (1-6 yr.) (ATSDR, 1994)
Body Weight
= Assumed to be 85 kilograms (kg) for adult males; 70 kg for adult
       females, and 16 kg for a child (1-6 yr.)


Conversion of chemical concentration (conc) from ppm to mg/m3:

Chemical Conc (mg/m3) = Chemical Conc (ppm)
(24.45 / molecular weight of chemical)

Table A-3. Summary of Animal Studies -- Oral exposure

Species Exposure Outcome Findings NOAEL/ LOAEL Reference
Mice (NMRI) 10 days old to 60 dys old Acute: 5 and 320 mg/kg PCE and 50 and 290 mg/kg TCE, via gavage oil, daily for 7 dys beginning at 10 days of age Developmental/ nervous system: Spontaneous activity, observed as locomotion, rearing and total activity (hyperactivity) measured over three 20-min periods at ages of 17 and 60 days. No significant effects in any outcome observed at 17 days of age. At 60 days of age, significant increases in locomotion and total activity (hyperactivity) were observed at both doses of PCE ( 5 and 320 mg/kg/day). Rearing was significantly decreased at high dose group, but not low dose, of PCE. TCE-exposed mice had decreased rearing behavior (only) at both dose groups. No observable signs of toxicity or changes in weight gain were observed at any age. LOAEL = 5 mg/kg/day; Basis for ATSDR acute oral MRL = 0.05 mg/kg/day Fredriksson et al., 1993
Mouse (B6C3F1) Rats (Osborne-Mendel) Chronic: 78 wk, 5 dy/wk, PCE via gavage oil, followed by an observation period of 32 wk (rats) and 12 wk (mice). PCE = 371 and 941 mg/kg/day for male rats; 474 and 949 mg/kg/day for female rats; 536 and 1,072 mg/kg/day for male mice; 386 and 772 mg/kg/day for female mice Cancer Carcinogenic bioassay. High early mortality observed at both doses in both sexes of rats and mice was related to PCE-induced kidney toxicity, and indicates that a maximum tolerated dose was exceeded. Because of reduced survival, the study is not considered adequate for evaluation of carcinogenicity in rats. Statistically significant increases in liver carcinomas occurred in treated mice of both sexes. CEL = 386 mg/kg/day for female mice and 536 mg/kg/day for male mice, for liver cancer (carcinomas) NCI, 1977

Table A-4. Summary of Animal Studies -- Inhalation exposure

Species Exposure Outcome Findings NOAEL/ LOAEL Reference
Rat (Sprague-Dawley) Acute: 4 dy for 6 hr/d, PCE (unspecified purity) at 200 ppm; behavior observed at 1 hr and 17 hr after last exposure Nervous system: Behavior Behavioral alteration (i.e., increased open-field behavior or ambulation) was observed in males 1 hr, but not 17 hr, after last exposure. Biochemical changes in brain included reduced ribonucleic acid (RNA) content and increased non-specified cholinesterase content. No brain histologies performed so no correlation with structural brain damage possible. LOAEL = 200 ppm for increased open-field behavior (ambulation) Savolainen et al., 1977
Rat (Fischer-344) Acute: 2 wk for 5 dy/wk for 6 hr/d to PCE conc of 0,100, 200, 425, 875, and 1,750 ppm Nervous system: Behavior (hypoactivity/ ataxia) Mice exhibited significant dyspnea, hypoactivity, and ataxia at the highest concentration PCE. No histologic lesions were observed in exposed rats. NOAEL = 875 ppm for hypoactivity/ ataxia NTP, 1986
Mouse (B6C3F1) Acute: 2 wk for 5 dy/wk for 6 hr/d to PCE conc of 0,100, 200, 425, 875, and 1,750 ppm Nervous system: Behavior (anesthesia) Mice exhibited significant dyspnea, hypoactivity, anesthesia, and ataxia at the highest concentration PCE. No histologic lesions were observed in exposed mice. NOAEL = 875 ppm and LOAEL = 1750 ppm for anesthesia NTP, 1986
Mouse (B6C3F1) Acute: 4 hr at 2,328 ppm PCE Nervous system Anesthesia. No histologic lesions observed in exposed mice. LOAEL = 2328 ppm for anesthesia NTP, 1986
Gerbil (Mongol-ian) Intermediate: Exposure to 60 ppm PCE?? For 90 dy for 24 hr/d Nervous system Significant decreases in DNA content in portions of the cerebrum LOAEL = 60 ppm for decreased DNA levels in frontal cortex Karlsson et al., 1987
Gerbil (Mongol-ian) Intermediate: Exposure to 60 or 320 ppm PCE?? for 3 mo for 24 hr/d, followed by a 4-mo exposure-free period Nervous system Significant decrease in deoxyribonucleic acid (DNA) content in various brain regions of brain LOAEL = 60 ppm for decreased DNA content in frontal cerebral cortex Rosengren et al., 1986
Rat (Sprague-Dawley) Intermediate: Exposure to 300 or 600 ppm PCE for 4 or 12 wk Nervous system (Study designed to examine effects of PCE on different regions and cell types of the brain). Measured brain weight and neuronal and glial markers. Brain weight was significantly reduced at 600 ppm after 4- and 12-wk exposures. No significant changes in cytosolic, neuron-specific protein (enolase) found in the frontal cortex, hippocampus, and brainstem. The cytosolic marker of glial cells (S-100) was significantly reduced in all three brain regions following exposure at 600 ppm for 12 weeks, with greatest reduction in frontal cerebral cortex. Cytoskeletal elements of neuronal cells (neurofilaments) and glial cells (glial fibrillary acid protein, GFAP) were significantly reduced in frontal cerebral cortex at 600 ppm at 4 and 12 wk (neurofilaments) and at 12 wk (GFAP). In hippocampus and brainstem, only GFAP was significantly reduced after 12 wk exposure at 600 ppm. NOAEL = 300 ppm and LOAEL = 600 ppm in males for decreased brain weight, decreased cytoskeletal protein. Researchers concluded that the frontal cerebral cortex is more sensitive to PCE than other brain regions, that cytoskeletal proteins are more sensitive than cytosolic proteins, and that glial cells are more sensitive to PCE toxicity than neurons. Wang et al., 1993
Rat (Sprague-Dawley) Intermediate: Exposure to 320 ppm PCE for 30 or 90 dy for 24 hr/d Nervous system Rats continuously exposed to 320 ppm PCE for 30 dys had changes in brain cholesterol, lipids, and polyunsaturated fatty acids. Changes in fatty acid composition of the brain were also observed in rats continuously exposed to PCE at 320 ppm for 90 dys. LOAEL = 320 ppm for changes in fatty acid composition of the brain in males Kyrklund et al., 1988, 1990
Gerbil (Mongol-ian) Chronic: Exposure to 120 ppm PCE for 12 mo for 24 hr/d Nervous system Significant changes in phospholipid content (phosphatidylethanolamine) in cerebral cortex and hippocampus LOAEL = 120 ppm in males for phospholipid changes in cerebral cortex and hippocampus Kyrklund et al., 1984
Rat (Sprague-Dawley) Acute: Exposure to 0 100, or 900 ppm PCE for 7 hr/d at Gd 14-20 Developmental Performance ascent test; brain acetylcholinesterase (AChE) levels; activity open-field NOAEL = 100 ppm and LOAEL = 900 ppm for transient decreased performance ascent test; decreased brain AChE; and increased activity open-field Nelson et al., 1980
Rat (Sprague-Dawley) Acute: Exposure to 0 or 300 ppm PCE or TCE for 7 hr/d for 9 dys ( Gd 6-15) Developmental: skeletal and soft tissue anomalies; fetal and maternal weight and external abnormalities; incidence of implantations, litter size, resorptions, fetal body weights, and sex ratios No effect on incidence of skeletal or soft tissue (or external) anomalies in fetuses or mothers; no effect on maternal or fetal body weight for both PCE and TCE exposures. Slight but significant increase in maternal and fetal toxicity (fetal resorptions) in rats exposed to 300 ppm PCE. No effect on reproductive outcome for TCE exposure. LOAEL = 300 ppm (PCE) for increased fetal resorptions; [NOAEL = 300 ppm for developmental effects of TCE exposure in utero] Schwetz et al., 1975
Mouse (Swiss-Webster) Acute: Exposure to 0 or 300 ppm PCE or TCE for 7 hr/d at Gd 6-15 Developmental: skeletal and soft tissue anomalies; fetal and maternal weight and external abnormalities; incidence of implantations, litter size, resorptions, fetal body weights, and sex ratios No effect on incidence of skeletal or soft tissue (or external) anomalies in fetuses or mothers; no effect on maternal or fetal body weight for both PCE and TCE exposures. Slight but significant increase in maternal and fetal toxicity (decreased fetal weight, delayed fetal ossifications) observed at 300 ppm PCE. LOAEL = 300 ppm (PCE) for decreased fetal weight; delayed ossifications [NOAEL = 300 ppm for developmental effects of TCE exposure in utero] Schwetz et al., 1975
Mouse (B6C3F1) Chronic: Exposure to 0, 100, or 200 ppm PCE for 103 wk (2-yr) for 5 dy/wk for 6 hr/d Cancer Significant evidence of carcinogenicity for mice as shown by increased incidence of both hepatocellular (liver) adenomas and carcinomas in males and of hepatocellular carcinomas in females. CEL = 100 ppm for liver (hepatocellular) carcinoma NTP, 1986; Mennear et al., 1986
Rat (Fischer-344) Chronic: Exposure to 0, 200, or 400 ppm PCE for 103 wk (2-yr) for 5 dy/wk for 6 hr/d Cancer Significant evidence for carcinogenicity in male rats as shown by increased incidence of monocellular cell leukemia and an increased incidence (although not statistically significant) of uncommon renal tubular cell neoplasms. Some evidence provided for carcinogenicity in females as shown by increased incidence of monocellular cell leukemia. CEL = 200 ppm for mononuclear cell leukemia NTP, 1986; Mennear et al., 1986

Table A-5. Summary of Human Health Studies—
Oral and Inhalation Exposure

Human Exposure Outcomes Findings NOAEL/ LOAEL Reference
Human (Volunteers) Acute Inhalation: Exposure to 10 or 50 ppm PCE for 4 hr/d for 4 dy. Nervous system: neurological battery At 50 ppm, visual-evoked potential latencies were significantly increased over the exposure period compared to subjects exposed at 10 ppm and effects correlated significantly with blood PCE levels. At 50 ppm, significant deficits for vigilance and eye-hand coordination were reported compared to subjects exposed at 10 ppm. NOAEL = 10 ppm; Basis for ATSDR acute inhalation MRL = 0.2 ppm Altmann et al., 1990, 1992
Human infant Acute Oral: PCE exposure via breast feeding to a single infant Liver toxicity A 6-week old infant breast-fed PCE developed obstructive jaundice and enlarged liver cells. PCE was found in breastmilk at level of 1 mg/dL (1 hour after exposure) and in maternal blood at 0.3 mg/dL (2 hours after exposure). Rapid improvement following after breast-feeding ended. Results cannot be conclusively attributed to PCE exposure because only one case was reported; however, suggests transfer of PCE in breastmilk. None reported Bagnell and Ellenbarger, 1977
Human (Occupational) Chronic Inhalation: Exposure to various levels PCE for 1-120 months for 5 dy/wk for 8 hr/d Nervous system Increased reporting of subjective symptoms: forgetfulness and dizziness LOAEL = 20 ppm for increase in subjective symptoms including dizziness and forgetfulness Cai et al., 1991
Human Acute Inhalation: Exposure to 0, 20, 100, and 150 ppm PCE for 5 dy for 7.5 hr/d Nervous system Cerebral cortex depression was observed at 100 ppm PCE. No effects on visual-evoked potentials, equilibrium tests, math skills, time discrimination and reaction times were noted (Hake and Stewart, 1977). Changes in EEG pattern, typical of drowsiness, light sleep, or anesthesia, observed at 100 ppm (Stewart et al., 1981). NOAEL = 25 ppm and LOAEL = 100 ppm for cerebral cortex depression Hake and Stewart, 1977; Stewart et al., 1981
Human (Clinical) Acute/ Oral: Exposure to single dose PCE Nervous system: unconscious-ness Use of PCE therapeutically, as a anthelminthic (de-worming agent) LOAEL = 108 mg/kg/dy Kendrick, 1929
Human Acute Inhalation: Exposure to various doses of PCE for 0.05 - 2 hr. Nervous system Dizziness, sleeplessness, incoordination NOAEL = 106 ppm and LOAEL = 216 ppm for dizziness and sleeplessness; LOAEL = 280 ppm for incoordination Rowe et al., 1952
Human Acute Inhalation: Exposure to 100 ppm PCE for 5 dy for 7 hr/d. No control subjects were used. Nervous system Mood and personality changes, headache, dizziness, difficulty speaking, sleepiness were observed at 100 ppm PCE. LOAEL = 100 ppm for mood and personality changes Stewart et al., 1970

Table A-5. Summary of Human Health Studies—
Oral and Inhalation Exposure

Human Exposure Outcome Findings NOAEL/ LOAEL References
Community: > 11,000 live births to women residing in base family housing at U.S. Marine Corps Base at Camp Lejune, NC., who delivered from Jan 1, 1968 - Dec 31, 1985 and were supplied water by 2 sources: Hadnot Point and Tarawa Terrace Oral/ Inhalation: VOC contamina-tion in 2 water supplies. Hadnot Point: 1982-1985 (known) and 1940s-1982 (unknown); max TCE = 1,400 ppb, DCE = 407 ppb. Tarawa Terrace: 1982-1985 (known) and 1954-1982 (unknown); max PCE = 215 ppb, TCE = 8 ppb Reproductive/ Developmental: Mean birth weight (MBW) and small for gestational age (SGA) Retrospective cohort study of exposure to VOCs in drinking water and adverse pregnancy outcomes. Base housing records were linked to birth certificate records. Several factors were evaluated: sex of infant, maternal and paternal age and education, maternal race, military pay grade, maternal parity, adequacy of prenatal care, marital status, and year of birth. Associations between the Tarawa Terrace water supply and decreased MBW and increased SGA were observed in two potentially susceptible subgroups: infants of mothers > 35 years age and infants of mothers with histories of fetal death. For mothers who were supplied with water from Hadnot Point, decreased MBW and increased SGA were observed in male, but not female, infants. These findings were unanticipated because associations were observed in certain groups but not others. The study is considered exploratory, although the outcomes were biologically plausible. None reported ATSDR, 1977
Community: Residents of Cape Cod, MA using public water where PCE was presumed to have leached from plastic liners of water distribution system Oral/ Inhalation: Exposures to PCE classified as high, low or any based on residence and water distribution patterns/ usage; not quantified for each residence Cancer: Bladder, kidney, leukemia A case-control study to evaluate the relationship between cancer and assumed exposure to PCE from public drinking water. Cancer (incident) cases were identified through the State Cancer Registry. Control subjects were "unexposed" residents of the same town. Information about confounding variables including smoking, occupation, residential history, alcohol consumption, and water usage habits were collected by interview. Researchers reported a significantly elevated odds ratio for bladder cancer among "highly" exposed persons; however, the effect was not significant when latency of disease onset in relation to exposure was considered. An elevated, but not significant odds ratio was reported for leukemia among persons who were exposed to "any" level of PCE, and the effect increased further among subjects whose exposure was "high." None reported Aschengrau et al., 1993
Community: 80,938 live births and 594 fetal deaths from 1985-1988 from a 4-county (75- town) area in northern NJ whose residents were mostly served by public water systems and births occurred mostly in state Oral/ Inhalation: 75 towns served by 49 water companies during pregnancy; each supply sampled at least 2 times per year for routine VOCs. Max monthly conc PCE = 26 ppb; TCE = 55 ppb; 1,2-DCA = 19 ppb; benzene = 2 ppb; carbon tetrachloride= 7 ppb Reproductive/ Developmental: For live births, outcomes were: low birth weight (LBW; < 2500 g) and very LBW (< 1500 g) among term births (> 37 gestational weeks), small for gestational age (SGA), and preterm birth (< 37 weeks gestational age). For all live births and fetal deaths, outcomes were: central nervous system (CNS), oral cleft, major cardiac, ventricular septal, and all cardiac defects; and neural tube defects (NTDs) A cross-sectional study using environmental and birth outcome databases (e.g., individual outcomes linked with individual exposure data). Birth outcome status and maternal risk factor data (e.g., age, race, education, prenatal care, birth order) were obtained from vital records and the NJ Birth Defects Registry. Odds ratios > 1.5 were found for: PCE with oral cleft defects; TCE with CNS defects, neural tube defects (NTDs), and oral cleft defects; carbon tetrachloride with LBW, very LBW, SGA, CNS defects, oral cleft defects, and major cardiac defects; benzene with NTDs, and major cardiac defects, total dichloroethylenes with CNS defects and oral cleft defects, and 1,2-dichloroethane with major cardiac defects. The authors reported that by itself, this study cannot resolve whether drinking water contaminants are causally related to adverse birth outcomes, or whether these findings arose because of chance or bias in exposure classification. None reported Bove et al., 1995
Community: Children in elementary school in Rutherford, NJ (1972-1976) Unknown: Exposure to PCE and TCE in air, tap water, and soil in/around Pierremont School (sampling in 1978-79) Cancer: childhood leukemia and Hodgkin's disease ("clusters" in Rutherford, NJ) Evaluated incidence of childhood leukemia and Hodgkin's disease in school-aged children in Rutherford, NJ. Review of historical environmental sampling data (Phase I) found no elevated levels of PCE/TCE. Sampling (in 1978-79) of air, tap water and soil in/around the school (Phase II) revealed low levels of PCE (max = 0.1 ppb in tap water, 1.0 ppb in air, and non-detect, ND, in soil) and TCE (max = ND in tap water, 0.7 ppb in air, and ND in soil). Levels of contaminants similar to other communities in the state, and no past environmental contamination found causally linked or related to the observed cases. None reported Burke et al., 1980
Community: 8 families (28 family members) in East Woburn, MA in which leukemia had occurred from 1969 to 1979 Oral/Inhalation: Refer to Lagakos, et al. (1986) for exposure-related information Cancer: childhood leukemia Related to the study by (Lagakos et al. , 1986). Medical and laboratory evaluations were carried out on 28 family members of the patients with leukemia identified from 1969-1979 (Lagakos et al. , 1986), with emphasis on immune function disorders. Subjects complained of unexplained tachycardia (elevated heart rate) at rest, palpitations, and rashes. Immunological profiles showed depressed non-specific immune function parameters. No control data were provided. The study findings are questionable because exposures to PCE were not verified nor quantified. None reported Byers et al. 1988
Community: 151 residents of Hardeman County, TN: 94 "exposed" residents consuming contaminated private well water near a pesticide waste site which operated from 1964-1972; 57 "unexposed" (control) persons Oral/ Inhalation: Exposure to VOCs, primarily carbon tetrachloride and to a limited extent PCE, in private wells for an unknown period; longest possible exposure period = 14 yr. Liver toxicity: Heptatotoxicity as determined by health questionnaire, clinical exam, biochemical serum screen (liver and kidney function and bile acid determinations) conducted approximately 6 months and 1 year after contaminated well water use discontinued. Sampling of air (with shower running) and tap water from some homes conducted late 1978, approximately 6 mo and 1 year after private well water use discontinued (in summer 1978): found maximum concentration of PCE in water = 2,405 ppb; carbon tetrachloride = 18,700 ppb in water; maximum concentration of PCE in air = 14.3 ug/m3; carbon tetrachloride = 41 ug/m3. Found statistically significant increase in liver enzyme and bile acid profile, and liver enlargement, in persons using contaminated well water (compared with controls), within 6 months after exposures ended, suggesting liver toxicity. Community persons reported symptoms of GI anomalies, respiratory difficulty, neurologic symptoms, muscular weakness, high fever, dizziness, rashes, nausea, numbness of limbs, headache, skin and eye irritation. The liver enzyme and bile acid profiles were reduced by 1 yr after exposure ended, indicating transient liver toxicity from exposure to VOCs. None reported Clark et al., 1982; Meyer, 1983
Community: Residents of 75 municipali- ties in New Jersey, Oral/ Inhalation: Exposures to PCE (and TCE) were categorized as: 0; 0.1-5 ppb; and > 5 ppb. Cancer: Leukemia and non-Hodgkin's lymphoma Study examined relationship between incidence of leukemia and non-Hodgkin's lymphoma and PCE exposures in drinking water in 75 municipalities in New Jersey. Non-Hodgkin's lymphoma was increased in females in the highest exposure group. The increase was significant when only 1984-1985 exposure data were analyzed, but was insignificant when 1978-1983 exposure data were also included. Exposure classification and quantification were incomplete and included both TCE and PCE exposures. None reported Cohn et al., 1994
Community: 12 cases of childhood leukemia (children < 20 yrs age) in Woburn, MA diagnosed from 1969-1979. Oral/ Inhalation: Public water supply (Wells G & H) found to be contaminated by inorganics and VOCs at max conc PCE = 21 ppb, TCE = 267 ppb, chloroform = 12 ppb in May, 1979. No sampling data prior to 1979. Cancer: Childhood leukemia Retrospective study of leukemia occurrence, diagnosed from 1969-1979, in children < 20 yrs age living in Woburn, MA. Cases were confirmed by review of hospital and pathology records. The cases were distributed uniformly over the 11-year period during which cases were identified for the study. When analyzed by residence at diagnosis, 6 of the 12 cases were located close to each other within one census tract in East Woburn, at a rate 7.5 times greater than expected. Water from Wells G & H was principally distributed to East Woburn, until 1979 when the wells were closed. Parents of children were interviewed about medical history, mother's pregnancy history, school history, and environmental exposures. Nine of the 12 cases had acute lymphocytic leukemia (ALL). There were no significant differences between leukemia cases and non-cases in terms of environmental exposures or any other risk factors investigated. None reported Cutler et al., 1986
Community: 688,055 Residents of 27 towns served by public water (95-100% total town population) contaminated with VOCs (1984-85) in New Jersey Oral/ Inhalation: Exposures to VOCs in drinking water were classified as high (++), low (+) or none (-) based on sampling from 1984 to 1985; Highest town average conc PCE = 16 µg/L, TCE = 46 µg/L; total VOCs (PCE, TCE, trichloro- ethane, dichloro-ethylenes) = 72 µg/L Cancer: Leukemia An ecologic study to determine the relationship between the incidence of leukemias and the occurrence of VOCs in public drinking water supplies. Leukemia incidence data (1979-84) were obtained from the state cancer registry. Age and sex distribution and numbers of persons in each town were obtained from 1980 Census. Standardized Incidence Rates (SIRs) were calculated using NJ statewide 1982 age-and sex-specific rates of total leukemia to determine expected number of case for towns with same exposure levels. The SIR was elevated in females only in the highest exposure category for total VOCs and appeared to increase with increasing level of exposure (across all age categories). The rate ratio for females at the highest exposure level as compared to the lowest was 1.68. No association was observed in males in any exposure category. The authors suggested that exposure to drinking water contaminated with VOCs may be related to an increase the incidence of leukemia. However, this type of study design cannot be used to determine causality. None reported Fagliano et al., 1990
Community: 26 children of Mellery, Belgium residing near a quarry where illegal dumping occurred. Local residents, including adults, were studied in 1990 and in 1992 (follow-up). This 1993 study was an extended follow-up of children of these adults. Inhalation: PCE (max = 2.5 µg/m3) and TCE (max = 1.3 µg/m3) in air in village of Mellery sampled in Dec. 1989, Oct. 1992 and March 1993. Site remediation occurred from Sept. 1991 to Jan. 1992. Genotoxicity/ Cancer: Biological markers of genotoxicity (blood analyses of frequency of sister chromatid exchange, SCE; micronucleus assay; DNA single strand breaks) In the 1990 study, mean sister chromatid exchange (SCE) frequencies were higher in "exposed" residents than the unexposed group, and the difference was greater for children than adults. The 1992 follow-up study of the same exposed persons confirmed the previous results with even higher SCE frequencies for exposed compared with unexposed persons. In this study (in 1993), the same parameters as studied previously were evaluated to determine "reversibility" of effects following site remediation. In addition, "exposed" children were tested for lymphocyte SCE frequencies as a function of time. No significant differences in any outcome were observed for exposed, as compared to unexposed, children. Results suggest that previous effects on genotoxicity were transient and maybe due to chemical exposures. None reported Klemans et al., 1995
Community: childhood leukemia cases among residents of Woburn, MA from 1969-1979 and 4,396 pregnancies among Woburn residents from 1960-1980. Oral/Inhalation: Drinking water exposure to inorganics and VOCs from 2 town wells, Wells G & H from 1974 to 1979 (possible). Max conc of PCE (21 ppb), TCE (267 ppb), and chloroform (12 ppb) found in 1979. No data prior to 1979. Exposures determined from sampling data and water usage patterns. 2 Studies reported: 1) Cancer: childhood leukemia and 2) Reproductive/ Developmental (self-reported adverse effects) Two phase study involving: 1) 20 childhood leukemia cases in Woburn identified from pregnancies ending between 1960 and 1982, including 12 cases identified from 1969 to 1979, 1 case identified before 196, and 7 cases identified through 1980; and 2) health survey of pregnancies in Woburn from 1960-1980 identified in 1982 by phone census.

For cancer study: The researchers reported a statistical association between the occurrence of childhood leukemia and the potential for exposure to water from Wells G & H. Several outside researchers evaluated the data from this study and identified a number of shortcomings. Many issues focused on incomplete determination of exposure status (e.g., which cases had an opportunity for exposure to water from Wells G & H and which did not), and exposure levels, duration and frequency. Available sampling data were not adequate to determine exposures and precluded quantitative exposure assessment.

For the reproductive health survey: Within East Woburn, a water distribution model was used to determine usage areas in which high, moderate, or no proportion of potentially-contaminated water was supplied during pregnancy. Researchers reported significant elevations in perinatal deaths since 1970 and two categories of congenital anomalies: eye/ear and CNS/chromosomal/oral cleft anomalies. Exposures were not verified nor quantified. Survey subject to extensive recall bias. Several researchers questioned biological relevance of grouping of anomalies (as they were done) for statistical analysis.

None reported Lagakos et al., 1986
Community: 65 residents (Londonderry Township, PA) where chemical dumping took place from 1959-61 and resulted in past exposure to VOCs in private wells; 66 controls Oral/ Inhalation: Private wells tested for VOCs in July 1983 (TCE max= 140 ppb; PCE max = 1 ppb); max possible exposure period of 23 yrs General: Self-reported health survey conducted 3 weeks after contaminated well water use discontinued. Cross-sectional study design (exposure and self-reported health outcome data, via health survey, collected at same time). Mean length of residency since 1959 for the "exposed" group was 9 years; for "unexposed" group was 14 years. Survey used to collect self-reported prevalence of 14 diseases or medical conditions and 15 health symptoms. The exposed group reported significantly higher rates of 3 symptoms: 1) eye irritation, 2) diarrhea, and 3) sleeplessness. No significant correlation with levels of VOC exposure and reported health symptoms were found. Researchers report that data indicate no long-term health effects could be ascribed to VOC exposure in this population. None reported Logue, et al., 1985
Community: 2 cohorts: 1) Birth certificates for 1975-1979 among Woburn residents; and 2) 4396 pregnancies in Woburn from 1960-1980, identified by 1982 phone census. Oral/ Inhalation: Refer to Lagakos et al., 1986 for exposure information. Reproductive/ Developmental: 2 outcomes evaluated: 1) small for gestational age (SGA) as defined by < 10th percentile for gestational age-specific weight, determined from birth certificate; 2) low birth weight (LBW) as reported by mother to be < 6 lb. Two analyses conducted: 1) Prevalence of SGA based on birth certificates obtained for 1975-1979 among Woburn residents. Street address on birth certificate used to assign residence. Within East Woburn, a water distribution model was used to determine usage areas in which a high, moderate, or no proportion of water supplied by water from Wells G & H during pregnancy. Confounders measured: mother's age, education, and race and sex of infant. Odds Ratio (OR) for exposure to water in 3rd trimester and increased SGA (with 95% CI) = 1.6 (0.9-2.8) ; 2) Prevalence of LBW determined for pregnancies occurring from 1960-1980 in Woburn identified by a 1982 phone census. Confounders measured: maternal age, smoking, socioeconomic status defined by census tract. Odds Ratio (95% CI) = 1.0 (0.6-1.8). None reported MDPH, 1994
Community: Childhood (< 19 yrs age) leukemia cases diagnosed Jan 1, 1969-Dec 31, 1989 while residents of Woburn at time of diagnosis Oral/ Inhalation: Refer to Lagakos et al., 1986 for contaminant exposure information. Woburn divided into 5 zones of graduated monthly exposure based on relative proportion of water from Wells G & H. Etiologic period defined as two years prior to conception to date of diagnosis. Cancer: Childhood leukemia. A re-analysis of data for 12 leukemia cases diagnosed from Jan 1969- Dec 1979 and expanded to include an additional 9 cases diagnosed as of Aug. 1989 (from hospitals and State Cancer Registry) Retrospective study of exposure to Woburn (Wells G & H) waters supply and childhood leukemia. Included 12 leukemia cases identified from 1969-1978 and 9 additional cases identified through Aug. 1989 (total diagnosis period: 1969-1989). Information on confounding factors (residential, occupational, health history) collected during interview for the etiologic period. Non-significant but positive associations found for: maternal alcohol consumption during pregnancy; diagnosis of paternal grandfather with cancer; having father who worked for industries considered high risk for occupational exposures, and subject's consumption of water from Wells G & H as the primary beverage. A significant association was found for developing childhood leukemia and breast-feeding as a child. A strong, relationship, although not statistically significant, was reported for exposure to water from Wells G & H during pregnancy and childhood leukemia (OR = 8.33, CI = 0.73, 94.67). A significant qualitative dose-response effect of increasing opportunity for exposure to contaminated water and occurrence of leukemia was observed for exposures to water during pregnancy. The authors reported that the relative risk of developing childhood leukemia (from 1969 to 1989) was greater for mothers having consumed water from Woburn during pregnancy. Analytical data were inadequate to determine potential exposure types (e.g., exposures occurred to a variety of VOCs and inorganic compounds), as well as levels, duration and frequency of exposure. The MDPH has never identified a specific compound or group of compounds that may be responsible for the finding of leukemia in children born to mothers with the greatest opportunity for exposure to water from Wells G & H None reported MDPH, 1997
Community: 725 Persons diagnosed with (and/or died from) bladder cancer between 1978-1985 in 8 counties and 97 Zip codes areas in northwest IL near a landfill which operated from late 1950s - 1972. Oral/ Inhalation: Two of 4 municipal water wells ½-mile of landfill had VOC contamina-tion: Max conc TCE = 15 ppb in both wells, and PCE = 5.1 ppb in one well from 1982-1988. Wells installed in 1955; longest possible exposure period = 30 yr. Cancer: Bladder cancer incidence and/or death Descriptive epidemiological study: calculated age-adjusted bladder cancer SIRs and standardized mortality rates (SMRs) for 1978-1981 and 1982-1985 for each of 8 counties and 97 Zip code areas in northwest IL. Expected numbers of cases obtained from National Cancer Institute's Surveillance Epidemiology End Result (SEER) database. No excesses of bladder cancer (CA) incidence were found for any county; however, significant cancer incidence excesses were reported for 2 Zip code areas; one area had excesses in males (SIR = 1.5) and females (SIR = 1.9), and the excesses were found in one town within that one Zip code area (SIR males = 1.7; SIR females = 2.6). Significant bladder cancer mortality excesses were found in one county (Winnebago). None reported Mallin, 1990
Community: 1049 "Exposed" residents (> 18 yr) of Lowell, MA living within 1/4-mile of a site where VOCs were dumped from 1971-1977; remediation of site in 1981; 948 "control" persons Inhalation: Air monitoring during a 7-dy period in Aug 1982 showed VOCs in residential air at levels < 10 ppb for PCE, TCE, and benzene, and < 20 ppb for toluene. Respiratory, Gastrointestinal (GI), cardiac, eye/nose irritation Symptom prevalence study of a neighborhood potentially "exposed" to airborne VOCs (e.g., toluene, benzene, PCE, TCE). Determined the frequency of self-reported health complaints: respiratory (wheezing, shortness of breath, chest discomfort, persistent colds, coughs), GI complaints (bowel dysfunction) and other (fatigue, irregular heart beat, irritation of eyes and nose). Researchers noted a biological gradient for self-reported complaints: higher frequency of complaints for residents living closest to center of site. No medical verification of self-reported health problems. Collected info on occupation, smoking, demographics. Problems with recall bias because residents concerned about possible site-related health problems prior to survey. No correlation of actual exposure levels of VOCs with residence of subjects with adverse health outcomes was conducted. None reported Ozonoff et al., 1987
Community: Births to persons living near hazardous waste sites in the San Francisco Bay area, CA Unknown: Exposures were not quantified but defined by residential proximity to a hazardous waste site Reproductive/ Developmental Low birthweight Ecologic study of hazardous waste sites in 5 counties of the San Francisco Bay area, CA. No associations between living in proximity to a hazardous waste site and low birth weight were reported. None reported Shaw et al., 1992
Community: Births to persons living near U.S. EPA's National Priority List hazardous waste sites during pregnancy in 1990. Many types and levels of exposure; none verified or quantified. Reproductive/ Developmental: Low birth- weight, very low birthweight, infant and fetal death, prematurity, and congenital malformations Data linkage "study" involving outcome data from the 1988 National Maternal and Infant Health Survey Files and locations of U.S. EPA's National Priority List sites in 1990. Purpose was to determine any possible relationships between living in proximity to a hazardous waste site and low birthweight. Many types of chemical exposures were possible; exposure types and levels were not verified or quantified. No significant odds ratios were reported for low birthweight, very low birthweight, infant and fetal death, prematurity, and congenital malformations and living near a site during pregnancy. Information on maternal age, parity, prenatal care, and behavioral and socioeconomic factors related to these outcomes was collected. Findings are, at best, hypothesis-generating. None reported Sosniak et al., 1994


Figure 1. Census Tracts in Natick, Massachusetts
(Source: MDPH, 1997a)

Figure 1 was excluded from both the hard copy and electronic copy.

Appendix C - Glossary

Occurring over a short time, usually within a few hours and less than 24 hours. An acute exposure can result in short-term or long-term health effects. An acute effect happens a short time (up to 1 year) after exposure.

Ambulation (Activity)
In behavioral studies of animals, ambulation refers tofree movement of the animal, such as when the animal is placed in large unconfined spaces. Normally, when animals are placed in a new environment they will move around freely to explore that environment. Increased ambulation activity denotes excessive movement in unconfined spaces.

A carcinogen is any substance that may produce cancer. Carcinogenesis is the process by which cancer occurs.

Central Nervous System (CNS)
The CNS is the part of the nervous system that includes the brain and spinal cord.

Chronic refers to occurring over a long period of time (more than 1 year).

The amount of one substance dissolved or contained in a given amount of another. For example, sea water contains a higher concentration of salt than fresh water.

Congenital Malformations
Congenital malformations generally refer to abnormalities in infants that are recognizable at birth (birth defects). Examples of congenital malformations include structural defects or anomalies (e.g., neural tube defects, oral clefts) and mental retardation.

Any substance or material that enters a system (the environment, human body, food, etc.) where it is not normally found.

Cross-Sectional Study
The cross-sectional study is a type of descriptive (epidemiology) study design whereby the status of an individual with respect to the presence or absence of both exposure and disease is assessed at the same point in time. Cross-sectional studies, because they are descriptive in nature, are useful for raising questions about the presence of associations between exposure and disease rather than for testing hypotheses.

Referring to the skin. Dermal absorption means absorption through the skin (into the body).

Descriptive (Epidemiology) Study
Descriptive epidemiology is concerned with describing the distribution of disease, including consideration of what populations or subgroups do or do not develop a disease, in what geographic locations is it most or least common, and how the frequency of occurrence varies over time. Due to limitations inherent in their study design, descriptive (epidemiology) studies are primarily useful for formulating hypotheses about associations between exposure and disease occurrence that can be subsequently tested in an analytic study design.

The amount of substance to which a person is exposed, taking body weight into account. For example, exposure dose represented in units of mg/kg denotes milligram of chemical per kilogram of body weight.

The relationship between the dose of a chemical administered or received and the incidence of adverse health effects in the exposed population. Dose-response relationships can be derived using data from studies in laboratory animals or human populations. Information from a quantitative dose-response relationship can be used to estimate the likelihood of adverse effects occurring at different exposure dose levels.

Environmental contamination
The presence of hazardous substances in the environment. From the public health perspective, environmental contamination is addressed when it potentially affects the health and quality of life of people living and working near the contamination.

The study of the occurrence and causes of health effects in human populations. An epidemiological study often compares two groups of people who are similar except for one factor, either an exposure or a presence of a health effect of interest.

Contact with a chemical by swallowing (ingestion), by breathing (inhalation), or by direct skin contact (dermal exposure). Exposure my be short term (acute) or long-term (chronic).

Gavage (Gavage-oil, Gavage-water)
A route of chemical administration (exposure) used in studies of laboratory animals where the agent is dissolved in solution and mechanically introduced into the stomach of the animal via a tube passing through the mouth. Gavage is often used in lieu of dietary administration because it provides a more accurate measure of actual exposure dose and can be used when aversive chemical taste is an issue. Gavage-oil refers to administration of a chemical that is dissolved in oil. Gavage-water refers to administration of a chemical that is dissolved in water.

Genotoxic is a shortened word for genetic toxicity and refers to adverse effects on the genetic material (DNA) of an organism. Genotoxicity may involve small changes in DNA base pairing or sequence to gross changes in chromosome structure or number. Genotoxicity may result in inherited diseases or cancer.

Gestation is the period of pregnancy (carrying the young in the uterus).

A source of risk that does not necessarily imply potential for occurrence. A hazard produces risk only if an exposure pathway exists, and if exposure creates the possibility of adverse consequences.

Health Consultation
A response to a specific question or request for information pertaining to a hazardous substance or hazardous waste facility (site). It often addresses an issue that necessitates a more rapid response, and thus is more limited in scope, than a public health assessment.

Health Outcome
An adverse health effect of interest, such as cancer or birth defects.

Human Health Study (or Investigation)
Any investigation of a defined population, using epidemiological methods, which would assist in determining exposures (risk factors) or possible public health impact (occurrence of health effects).

In human and animal studies, incidence refers to occurrence or frequency, as in the incidence of an adverse effect being observed in persons or animals who have been exposed to a particular chemical as compared to those who have not.

In utero
Refers to the period of gestation (after conception and before birth) of an embryo or fetus. Chemical exposure to a pregnant women may result in in utero exposures to the embryo or fetus.

Swallowing, such as eating or drinking. Chemicals can get in or on food, drink, utensils, cigarettes, or hands where they can be ingested. After ingestion, chemicals can be absorbed into the blood and distributed throughout the body.

Intermediate exposure is exposure for more than 2 weeks and up to 1 year (e.g., 15 to 364 days).

Low Birthweight (LBW)/Very Low Birthweight (VLBW)
LBW infants are commonly defined as infants weighing less than 2500 grams at birth. Very low birthweight (VLBW) infants weigh less than 1500 grams at birth.

Lowest-Observed-Adverse-Effect Level (LOAEL)
In dose-response experiments, the LOAEL is the lowest exposure level at which there are statistically or biologically significant increases in the frequency or severity of adverse effects between the exposed population and its appropriate control (unexposed) group. The LOAEL may also be referred to as the minimum-effective level in this consultation.

Maximum Contaminant Level (MCL)
The MCL is the drinking water standard established by the U.S. Environmental Protection Agency (EPA). It is the maximum permissible level of a contaminant in water that is delivered to the free-flowing outlet (for public distribution). MCLs are considered protective of public health over a lifetime (70 years) for individuals consuming 2 liters of water a day.

Mean Birth Weight (MBW)
Mean birth weight means average weight at birth.

Metabolism/Metabolic Pathways
Metabolism refers to the breakdown or transformation of chemical in the body. Metabolism may occur at several locations in the body, such as the gut, lung, liver and kidney. The process by which the breakdown occurs is referred to as a metabolic pathway. Depending on the chemical and its concentration in the body, and the species of animal, one or more metabolic pathways may be used in the body for chemical breakdown. In mice, metabolism of perchloroethylene (PCE) to trichloroacetic acid (TCA) by the liver is a more important metabolic pathway than for rats and humans. In rats, metabolism of perchloroethylene (PCE) by combining (conjugation) with glutathione in the kidney is a more important metabolic pathway than for mice and humans.

Metabolites are the products of chemical breakdown or transformation.

Minimum-Effect Level
The minimum-effect level refers to the lowest level of a chemical reported in the scientific literature to be associated with adverse health effects. In most cases, it is the same as the lowest-observed-adverse-effect level (LOAEL).

Minimal Risk Level
An estimate of daily human exposure to a substance that is likely to be without appreciable risk of adverse non-carcinogenic effects over a specified duration of exposure. MRLs are derived when reliable and sufficient data exist to identify target organ(s) of effect or the most sensitive health effect(s) for a specified duration via a given route of exposure. MRLs are based on non-cancer health effects only, and can be derived for acute, intermediate, and chronic duration of exposure by the inhalation and oral routes of exposure.

National Priorities List
The Environmental Protection Agency's listing of sites that have undergone preliminary assessment and site investigation to determination which locations pose immediate threat to persons living or working near the release or to the environment. These sites are most in need of cleanup.

Neurotoxicity or Nervous System Toxicity
Neurotoxicity is a general term which means adverse effects on the nervous system, including both the central nervous system (brain and spinal cord) and the peripheral nervous system (outside the central nervous system).

No-Observed-Adverse-Effect Level (NOAEL)
In dose-response experiments, the NOAEL is the exposure level at which there are no statistically or biologically significant increases in the frequency or severity of any effect between the exposed population and its appropriate control (unexposed) group.

Perchloroethylene (PCE)
Perchloroethylene (PCE), also called tetrachloroethylene, is a solvent (volatile organic compound) commonly used in industry (as a metal degreaser) and in dry-cleaning.

Pharmacokinetics refers to the dynamics of absorption, distribution, metabolism and elimination of chemicals in the body following exposure by any route (ingestion, inhalation, skin contact).

Physiologically-Based Pharmacokinetic (PB/PK) Model
A PB/PK model is a mathematical description of the disposition of chemicals either in an entire organism or in a part of an organism (e.g., an organ such as the kidney or liver) that is derived primarily from physiological and physiochemical information or from experimental data.

Public Health Assessment (PHA)
The evaluation of data and information on the release of hazardous substances in the environment in order to assess any current or future impact on public health, develop health advisories or other recommendations, and identify studies or actions needed to evaluate and mitigate or prevent human health effects; also, the document resulting from that evaluation.

Risk is the probability that something will cause injury, combined with the potential severity of that injury.

Route of Exposure
The route of exposure is the way in which a person may contact a chemical substance. For example, drinking (ingestion), and bathing (skin contact) are two different routes of exposure to contaminants that may be found in water.

Sister Chromatid Exchange (SCE)
SCE is a laboratory test used to determine whether a chemical causes genetic damage (genotoxicity).

Small for Gestational Age (SGA)
A condition where infants are born after 37 weeks of pregnancy (gestation), and therefore have sufficient time to grow and mature, but are nonetheless small in terms of weight. SGA infants are often within the bottom tenth percentile of the weight distribution at any given gestational age. Some SGA infants are simply smaller than average, while others are growth retarded. SGA and preterm infants (infants who are born small because they were born before 37 weeks completed gestation) are separate outcomes but are often grouped together in epidemiological studies and measured as low birth weight infants.

The study of the adverse effects of chemicals and other agents in biological systems. For purposes of this health consultation, toxicological studies refer to studies conducted in laboratory animals.

Trichloroethylene (TCE)
Trichloroethylene (TCE) is a synthetic chemical (volatile organic compound) commonly used as a solvent in industry (as a metal degreaser) and in dry-cleaning.

Volatile Organic Compounds (VOCs)
A group of substances containing carbon and different proportions of other elements such as hydrogen, oxygen, fluorine, chlorine, bromine, sulfur, or nitrogen; these substances easily become vapors or gases. A number of VOCs are commonly used as solvents (paint thinners, lacquer thinner, degreasers, and dry cleaning fluids). Examples of VOCs include: perchloroethylene (PCE), trichloroethylene (TCE), 1,1,1-trichloroethane, carbon tetrachloride, 1,2-dichloroethane, benzene, and dichloroethylenes.

Appendix D: Response to Public Comments

ATSDR received the comments during the public comment period for the Natick Research Army Laboratories (a/k/a U.S. Army Soldier Systems Command Natick [SSCOM]) Health Consultation evaluating public health impacts from exposure to VOCs in the Natick public water supply (October, 1997). For comments that questioned the validity of statements made in the consultation, ATSDR verified or corrected the statements. The list of comments does not include editorial comments concerning word spelling or sentence syntax.

  1. Comment: Insert information showing whether or not contamination in Evergreen Well No. 1 of the Natick public water supply was increasing between 1988 to 1991.

    Response: Average and maximum contaminant levels detected in Evergreen Well No. 1 from 1988 to 1991 are presented in Table 1 of the consultation. Maximum PCE levels were relatively constant over this time period, ranging from approximately 9 to 16 µg/L, and were actually lowest during the middle years (1989 and 1990). The maximum PCE concentrations were used as the focus of this consultation.

    Average PCE levels increased slightly from 1988 to 1991 (e.g., from 5 to 11 µg/L). However, given the small number of samples (4 to 6) for each year and normal variances inherent in laboratory analyses, there does not appear to be a significantly increasing trend in PCE concentrations over time.

  2. Comment: Suggest inserting a table showing the levels for PCE and TCE in water delivered to households, to serve as a comparison to levels found in the Evergreen Well No. 1.

    Response: The average annual levels of PCE and TCE in water delivered to households never exceeded the Maximum Contaminant Level (MCL) of 5 µg/L during the period from 1988 to 1991. This is stated on pages 4, 5, and 26 of the consultation. An additional column has been added to Table 1 including this information.

  3. Comment: Suggest moving the sentences contained in the last paragraph on page 8 (beginning with "It was most efficient...", and ending with "...detected in this well") to an appendix.

    Response: This information is important to understanding that the chemical toxicity of PCE and TCE, following exposure to Natick public water, is not dependent solely on the levels of contaminant present in the water. Rather, it depends on a number of processes that occur in the body following exposure, including absorption, distribution, metabolism, and elimination of chemical from target tissues. Because of these processes, the amount of TCE and PCE that ultimately reaches target tissues in the body, where adverse effects could occur, is considerably less than the amount contained in the Natick public drinking water. In fact, ingestion of TCE and PCE in drinking water, even at the highest levels detected in the Natick public water supply wells, is likely to be completely eliminated by the liver and lungs, prior to reaching the blood and being transported throughout the body. Because this is important information for understanding potential toxicity to Natick residents, the information is best presented in the body of the document.

  4. Comment: Recommend having a glossary of terms, list of definitions of acronyms, and table of contents.

    Response: All of these items have been added to the consultation.

  5. Comment: The first paragraph under "Oral Exposure" on page 11 should be restructured to improve clarity.

    Response: The paragraph has been restructured.

  6. Comment: Under "Nervous System Toxicity" the wording should be changed to: "A detailed summary including doses, length of study, outcomes, and findings on the effects of PCE exposure on nervous system toxicity is provided in Appendix A."

    Response: We have included this sentence, or a similar one, under each of the three headings: "Reproductive and Developmental Toxicity", "Nervous System Toxicity", and "Cancer" in the consultation.

  7. Comment: Suggest including a map depicting the Census tracts included in the MDPH (1997b) investigation.

    Response: A map (Figure 1) has been added.

  8. Comment: The methods used in the consultation are not consistent with standard risk assessment approaches used by the U.S. Environmental Protection Agency (EPA), as defined in the EPA Risk Assessment Guidance for Superfund (RAGS) and the EPA Integrated Risk Information System (IRIS). Specifically, uncertainty factors were not used when comparing estimated human exposures to literature-based no-observed-adverse-effect (NOAEL) and lowest-observed-adverse-effect (LOAEL) levels.

    Response: The consultation involved a health evaluation (assessment), as opposed to a risk assessment, approach. The two tasks are methodologically different and serve different purposes. Risk assessment is generally a quantitative approach to evaluate health and environmental impacts in order to select on-site remedial alternatives. This approach involves standard procedures, as defined in RAGS and IRIS, which are used systematically across sites nationwide.

    By comparison, health assessment is generally a qualitative, weight-of-evidence approach used to evaluate public health impact to residential communities living near hazard waste sites. The health assessment is highly dependent on site- and community-specific information to evaluate and estimate exposures and related health impacts. It is the first phase in determining whether additional health-related activities (e.g., health studies, exposure registries, health education) are recommended for a community.

    This consultation was conducted as a follow-up to the public health assessment (PHA) conducted by ATSDR for the SSCOM site. This is described in more detail on page 3 of the consultation. In the PHA, ATSDR used screening-level approaches whereby toxicity values, namely ATSDR Minimal Risk Values (MRLs) and EPA Reference Doses (RfDs) and cancer slope factors (CSFs), were used to estimate potential cancer and non-cancer impact to Natick residents. These toxicity values are derived from literature-based NOAELs and LOAELs (and CELs), reported in studies primarily using laboratory animals, to which uncertainty factors have been applied.

    In this consultation, ATSDR used the information contained in the PHA and applied an additional weight-of-evidence approach to evaluate public health impact to Natick residents. This generally involved a more in-depth and less conservative approach, but one that better reflects what is known and not known about the hazards posed by PCE and the scientific uncertainties of the available literature. Therefore, implementation of standard EPA methods, as defined in RAGS and IRIS, was not warranted for this consultation.

  9. Comment: The human health studies were eliminated from consideration largely because of methodological issues, such as the inability to define individual exposures to VOCs. Specific examples include the studies by ATSDR (1997) and the Massachusetts Department of Public Health (MDPH, 1997a). Because it is not certain whether higher quality drinking water exposure data for PCE contamination will emerge in the future, EPA's approach is generally to employ conservative strategy that allows for the use of all the available data in evaluating potential human health risks. Problems with specific studies can be detailed in an "uncertainty section."

    Response: The human health studies were not eliminated from the evaluation in any way. Rather, they were reviewed together with the findings from animal studies of PCE exposure to determine whether they provide adequate and appropriate scientific information with which to evaluate public health impact to Natick residents. Specifically, they were evaluated to determine if they provide evidence for a cause-and-effect (dose-response) relationship between PCE exposure and adverse effects on reproduction and development, nervous system toxicity, or cancer. All studies involving potential or actual exposure to PCE, either alone or together with other chemicals, were reviewed.

    Taken together, the studies do not provide sufficient evidence for such an association between PCE exposure and adverse outcomes of pregnancy. Therefore, ATSDR feels that using them as a quantitative basis for evaluating public health impact for Natick residents would not be appropriate even if factors of uncertainty were included.

    In the ATSDR (1997) study, health outcomes were evaluated for the period of 1968 to 1985. Environmental sampling data used to define potential exposure were collected on few occasions between 1982 and 1985, a very small portion of the total period evaluated in the study. Exposures occurred to several VOCs simultaneously, the highest levels being dichloroethylenes (as opposed to PCE). The available exposure data are inadequate to determine any cause-and-effect relationships between exposure to PCE and reproductive and developmental toxicity.

    The MDPH (1997a) study provides qualitative evidence for an association between exposure to water from Woburn Wells G & H and adverse pregnancy outcomes. The study does not provide sufficient evidence of a quantitative, cause-and-effect (dose-response) relationship between PCE exposure and adverse pregnancy outcomes. This is primarily because of incomplete information about chemical exposure. The available environmental sampling data are not adequate to determine how long people may have been exposed to chemicals in the water, and at what levels. Water from Wells G & H contained several types of contaminants including PCE, TCE, chloroform and inorganic compounds (e.g., metals). It cannot be determined whether PCE was the cause of the increased childhood leukemia observed. The MDPH has never identified a specific compound or group of compounds that may be responsible for the finding of leukemia in children born to mothers with the greatest opportunity for exposure to the water (see Comment # 15 by MDPH, below). ATSDR does not feet it is appropriate to use the findings of this study as a sole basis to evaluate public health impact to Natick residents potentially exposed to PCE, no matter how conservative a strategy is employed and what type of uncertainties are included.

  10. Comment: Well-documented animal studies were overly discounted in the evaluation, and this may have underestimated the potential for human exposure to fumes from VOC-contaminated groundwater.

    Response: The animal studies were in no way discounted in the evaluation. Rather, they were reviewed together with the findings from human health studies, as a weight-of-evidence approach, to evaluate public health impact to Natick residents. All studies were reviewed to determine whether they provide evidence for a cause-and-effect (dose-response) relationship between PCE exposure and adverse effects on reproduction and development, nervous system toxicity, or cancer. Because of inherent differences between animals and humans, the animal studies were also reviewed to determine whether the findings were generalizable to humans. The results of animal studies are not always generalizable to humans because of important differences in how exposures occur (e.g., there is no human equivalent for exposure by gavage in animals), the levels of exposure experienced (e.g., animal exposures generally occur at higher levels than human environmental exposures), and most importantly, in the biological mechanisms of disease. For example, the kidney and liver cancers observed in rodents exposed to PCE may occur because of biological processes that are not important to humans. Additionally, the types of cancers observed in animals have not been documented in human populations. The weight-of-evidence approach used in this consultation did not discount animal studies; rather, it involved an additional consideration of the relevance of findings from animal studies to evaluate human health hazards.

  11. Comment: The statements about historic levels of PCE in the Evergreen Well No. 1 and the Natick public water supplied to households need to be more clearly stated. That is, levels of PCE in either water supply were not known prior to 1988 and should not be overstated. Also, the newly installed Natick water treatment system should be mentioned in the text.

    Response: The text has been modified to improve clarity and better reflect the available historic sampling data. The fact that no contamination by VOCs was detected in the Evergreen Well No. 1 during voluntary sampling events in 1980 and 1984 (pp. 6), which was prior to 1988, was also highlighted as a source of historic sampling data. The recently installed Natick water treatment system is already mentioned in the text (pp. 4).

  12. Comment: Remove the reference MDPH (1997b) from the list of human health studies reviewed for cancer and add a new sentence: "In addition, MDPH conducted a descriptive analysis of cancer incidence for Natick during the period 1982 to 1992 (MDPH, 1997b)."

    Response: These changes and additions were made to the text.

  13. Comment: The sentence starting on page 22, describing cancer cases by Census tract, states: "Additionally, most bladder and kidney cases occurred among current and former smokers, and smoking is a potential risk factor for these types of cancers" and should be modified to describe the reported smoking status at the time of diagnosis. The sentence implies that smoking was the cause of all bladder and kidney cancers in Natick. Most men (7 out of 9 cases) with bladder cancer reported being current or former smokers at the time of diagnosis, but this was not the case for kidney cancer. Suggested wording changes are provided.

    Response: The MDPH (1997b) report states that "Smoking is the most important known risk factor for cancers of the lung, bladder, kidney and pancreas (pp. 14)." The report also states that "...more than 50% of the individuals diagnosed with kidney cancer in this Census tract were current or former smokers which may have contributed to the increase observed in this area of town." (pp. 26).

    ATSDR has revised the wording of the sentence to read: "Additionally, the majority (approximately 78%) of bladder cancer cases reported being current or former smokers at the time of diagnosis, while more than half (e.g., 56%) of kidney cancer cases reported being current or former smokers at the time of diagnosis. Smoking is a potential risk factor for these types of cancers and may have contributed to the observed excess numbers of cases."

  14. Comment: ATSDR should delete the sentence on page 24: "Of the 12 cases of leukemia identified, only 6 could be linked to exposure to contaminated water, the other 6 cases had no access to contaminated water from Wells G & H" because this sentence oversimplifies the reason for the Lagakos et al. (1986) estimates for leukemia incidence related to the water. The Lagakos et al. (1986) estimate of the number of cases was derived by statistical estimation and does not represent a precise number. You may also want to delete the previous sentence referring a "number of shortcomings" unless you comment on other shortcomings noted by the referenced authors of the Lagakos et al. (1986) study.

    Response: The Executive Summary of the MDPH (1997a) study states that: " was learned that between January 1969 and December 1979, twelve cases of childhood leukemia had been diagnosed in Woburn, six of these cases resided in a six-block area which was served directly by the contaminated wells." The consultation does not state that these 12 leukemia cases were identified, or even estimated, by Lagakos et al. (1986). The text has been rewritten to further clarify this.

    The consultation references several researchers (MacMahon, 1986; Prentice, 1986; Rogan, 1986; Swan and Robins, 1986) who identified a "number of shortcomings" with the Lagakos et al. (1986) study. Many of these issues appeared to involve incomplete determination of exposure status for the childhood leukemia cases (e.g., which of the cases may have been exposed to water from Wells G & H and when). In the follow-up study of the childhood leukemia cases (MDPH, 1997a), a water distribution model was used to more precisely determine exposure status and history. The consultation has been revised to highlight these improvements.

  15. Comment: It is incorrect to focus on VOCs in Well G & H water when discussing the study by MDPH (1997a). Rather, there was a mixture of chemicals in water, including VOCs and inorganic compounds, and MDPH has never identified a specific compound or group of compounds that may be responsible for the finding of leukemia in children born to mothers with the greatest opportunity for exposure to the water. Also, the MDPH (1997a) report did not discuss "Cancer rates in East Woburn declined to expected levels about 8 years after the contaminated wells were closed, but were increased in West Woburn which had never received water from Wells G & H." This statement is incorrect.

    Response: We concur that the focus of the MDPH (1997a) study should be on exposure to water from Wells G & H and childhood leukemia. This is stated previously in Response #9 (above) as the primary reason why the findings of that study do not provide sufficient evidence for a cause-and-effect relationship between PCE and adverse pregnancy outcomes, and therefore cannot serve as a quantitative basis for evaluating potential health impacts to Natick residents. The text of the consultation has been modified to reflect this important difference.

    The statement regarding cancer rates declining about after eight years after Wells G & H were closed was based on information provided in the MDPH (1997a) study. The wording of the consultation has been modified to reflect exactly the text of that study: "For a period of eight years beginning in 1986 (e.g., 1986-1994), no new cases of leukemia had been diagnosed in Woburn. Between 1994 and the present, three more cases were identified among Woburn residents. The identification of these cases was not greater than would have been expected based upon statewide leukemia rates in children. Since 1986, childhood leukemia incidence in Woburn fell below that which would normally be expected in a community with Woburn's population." (MDPH, 1997a).

    The reference to the shift in geographic distribution of childhood leukemia cases after 1979 from East to West Woburn was based on information contained in Table 2 of Lagakos et al. (1986), which indicates that only one of the seven childhood leukemia cases identified after 1979 had an opportunity for exposure to water from Wells G & H (Lagakos et al., 1986; MacMahon, 1986). The childhood leukemia cases considered in Lagakos et al. (1986) were re-evaluated in the MDPH (1997a) study. The use of the Murphy water distribution model in the MDPH (1997a) study undoubtably provided a more accurate means of exposure assessment (e.g., which cases received contaminated water and which did not), as compared to the Lagakos et al. (1986) study. We have revised the text to reflect this.

  16. Comment: In reference to the MDPH (1997a) study, minimum effect levels of exposure for public health hazard were not determined because of methodologic problems with the study, but rather because there was very little actual contaminant concentration data available for the Woburn wells. The comment that the observed association may have been due to bias introduced by the study interviewers or by mothers interviewed in the study is not an issue because exposure estimates were based solely on the results of the Woburn water model and residential history information gathered during interviews.

    Response: We concur that environmental sampling data were not adequate to determine when exposure to contamination in Well G & H water first occurred, at what levels people may have been exposed, and for how long. The number of childhood leukemia cases was unexpectedly high from 1969 to 1979, although environmental sampling data were not available prior to 1979. We described this as a "methodologic problem" in the consultation because without adequate information about exposure, the identification of causal (dose-response) relationships, and therefore minimum-effective levels, between chemicals and adverse outcomes cannot be made. We have revised the text to state: "...the limitations in exposure assessment make the findings inadequate for determining causal relationships between exposure to PCE, or VOCs in general, and childhood leukemia. Therefore minimum-effective levels of exposure cannot be established from these data and used to evaluate public health hazard for Natick residents."

    The introduction of bias may have minimized by using the water distribution model to define exposure, but may not have been entirely eliminated because information about residential history was also used and was collected by interviewing parents of children with childhood leukemia. The site has received national attention for childhood leukemia and most of these residents had been interviewed previously and were not blinded to the purpose of this study (MDPH, 1997a). For these reasons, the statement about potential bias has been modified, but not deleted.

  17. Comment: We are pleased to hear that there is no substantial risk due to VOCs in the Natick water supply in the late 1980s and early 1990s. However, we remain concerned because this does not explain the heightened occurrence of pancreatic cancers in Natick (MDPH, 1997b); it only says that this was not the cause.

    Response: The MDPH's analysis of cancer incidence in Natick (MDPH, 1997b) identified one Census tract in Natick where pancreatic cancer had occurred in excess of expected numbers from 1982 to 1986. The purpose of this ATSDR consultation was to evaluate public health impact from potential past exposure to PCE in the Natick public water supply. It was not within the scope of this consultation to evaluate potential risk factors for pancreatic cancer cases in Natick.

  18. Comment: The report ignores our oft-repeated complaint about SSCOM's claimed "loss" of hazardous materials inventory reports. By choosing not to report our concern, you contribute to this failure on the part of SSCOM to disclose all data in a timely manner.

    Response: The purpose of this ATSDR consultation was stated previously in Response #17 and again in Response #19. It was not within the scope of this consultation to either collect or review inventory reports of SSCOM. Your concern for this matter is noted.

  19. Comment: Your report was limited to a literature review. We had hoped for a review of actual infant health records in Natick. We were told that this was a new and sensitive environmental indicator, and yet actual infant health data not only were undelivered, but that shortfall received no comment.

    Response: The purpose and scope of this ATSDR consultation was proposed to the Restoration Advisory Board (RAB) for SSCOM during the January, 1997 meeting. The members of the RAB agreed that ATSDR would conduct a literature review of the available animal and human studies pertaining to VOC exposure (e.g., perchloroethylene, PCE, and trichloroethylene, TCE) and adverse effects of reproductive and developmental toxicity and cancer. A strict review of infant mortality rates would not address cause and effect relationships between environmental contaminants and disease.

  20. Comment: The report only considered residential exposure to environmental risks, and continues to ignore our repeated requests to evaluate health histories of SSCOM employees, as well.

    Response: The ATSDR is mandated under CERCLA and SARA to conduct public health assessments and health consultations which determine the public health impact to communities potentially exposed to hazardous substances in the environment. The purpose of this consultation was to evaluate health impact to Natick residents from potential past exposure to contamination in the municipal water supply.

    In past Restoration Advisory Board (RAB) meetings, the issue of worker health has been raised. The installation (SSCOM) is looking into the availability of appropriate records. If records are available, the National Institutes of Occupational Safety and Health (NIOSH) has the authority to address worker issues.

Table of Contents The U.S. Government's Official Web PortalDepartment of Health and Human Services
Agency for Toxic Substances and Disease Registry, 4770 Buford Hwy NE, Atlanta, GA 30341
Contact CDC: 800-232-4636 / TTY: 888-232-6348

A-Z Index

  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #